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 Previously this was permitted.
98 JH/21 Bug 2501: Fix init call in the heimdal authenticator. Previously it
99 adjusted the size of a major service buffer; this failed because the
100 buffer was in use at the time. Change to a compile-time increase in the
101 buffer size, when this authenticator is compiled into exim.
103 JH/22 Taint-checking: move to safe-mode taint checking on all platforms. The
104 previous fast-mode was untenable in the face of glibs using mmap to
105 support larger malloc requests.
107 PP/01 Update the openssl_options possible values through OpenSSL 1.1.1c.
108 New values supported, if defined on system where compiled:
109 allow_no_dhe_kex, cryptopro_tlsext_bug, enable_middlebox_compat,
110 no_anti_replay, no_encrypt_then_mac, prioritize_chacha, tlsext_padding
112 JH/23 Performance improvement in the initial phase of a two-pass queue run. By
113 running a limited number of proceses in parallel, a benefit is gained. The
114 amount varies with the platform hardware and load. The use of the option
115 queue_run_in_order means we cannot do this, as ordering becomes
118 JH/24 Bug 2524: fix the cyrus_sasl auth driver gssapi usage. A previous fix
119 had introduced a string-copy (for ensuring NUL-termination) which was not
120 appropriate for that case, which can include embedded NUL bytes in the
121 block of data. Investigation showed the copy to actually be needless, the
122 data being length-specified.
124 JH/25 Fix use of concurrent TLS connections under GnuTLS. When a callout was
125 done during a receiving connection, and both used TLS, global info was
126 used rather than per-connection info for tracking the state of data
127 queued for transmission. This could result in a connection hang.
129 JH/26 Fix use of the SIZE parameter on MAIL commands, on continued connections.
130 Previously, when delivering serveral messages down a single connection
131 only the first would provide a SIZE. This was due to the size information
132 not being properly tracked.
134 JH/27 Bug 2530: When operating in a timezone with sub-minute offset, such as
135 TAI (at 37 seconds currently), pretend to be in UTC for time-related
136 expansion and logging. Previously, spurious values such as a future
137 minute could be seen.
143 JH/01 OpenSSL: With debug enabled output keying information sufficient, server
144 side, to decode a TLS 1.3 packet capture.
146 JH/02 OpenSSL: Suppress the sending of (stateful) TLS1.3 session tickets.
147 Previously the default library behaviour applied, sending two, each in
150 JH/03 Debug output for ACL now gives the config file name and line number for
153 JH/04 The default received_header_text now uses the RFC 8314 tls cipher clause.
155 JH/05 DKIM: ensure that dkim_domain elements are lowercased before use.
157 JH/06 Fix buggy handling of autoreply bounce_return_size_limit, and a possible
158 buffer overrun for (non-chunking) other transports.
160 JH/07 GnuTLS: Our use of late (post-handshake) certificate verification, under
161 TLS1.3, means that a server rejecting a client certificate is not visible
162 to the client until the first read of encrypted data (typically the
163 response to EHLO). Add detection for that case and treat it as a failed
164 TLS connection attempt, so that the normal retry-in-clear can work (if
165 suitably configured).
167 JB/01 Bug 2375: fix expansions of 822 addresses having comments in local-part
168 and/or domain. Found and fixed by Jason Betts.
170 JH/08 Add hardening against SRV & TLSA lookups the hit CNAMEs (a nonvalid
171 configuration). If a CNAME target was not a wellformed name pattern, a
174 JH/09 Logging: Fix initial listening-on line for multiple ports for an IP when
175 the OS reports them interleaved with other addresses.
177 JH/10 OpenSSL: Fix aggregation of messages. Previously, when PIPELINING was
178 used both for input and for a verify callout, both encrypted, SMTP
179 responses being sent by the server could be lost. This resulted in
180 dropped connections and sometimes bounces generated by a peer sending
183 JH/11 Harden plaintext authenticator against a badly misconfigured client-send
184 string. Previously it was possible to cause undefined behaviour in a
185 library routine (usually a crash). Found by "zerons".
187 JH/12 Bug 2384: fix "-bP smtp_receive_timeout". Previously it returned no
190 JH/13 Bug 2386: Fix builds with Dane under LibreSSL 2.9.0 onward. Some old
191 API was removed, so update to use the newer ones.
193 JH/14 Bug 1891: Close the log file if receiving a non-smtp message, without
194 any timeout set, is taking a long time. Previously we would hang on to a
195 rotated logfile "forever" if the input was arriving with long gaps
196 (a previous attempt to fix addressed lack, for a long time, of initial
199 HS/01 Bug 2390: Use message_id for tempfile creation to avoid races in a
200 shared (NFS) environment. The length of the tempfile name is now
201 4 + 16 ("hdr.$message_exim_id") which might break on file
202 systems which restrict the file name length to lower values.
205 HS/02 Bug 2390: Use message_id for tempfile creation to avoid races in a
206 shared (NFS) environment.
208 HS/03 Bug 2392: exigrep does case sensitive *option* processing (as it
209 did for all versions <4.90). Notably -M, -m, --invert, -I may be
212 JH/15 Use unsigned when creating bitmasks in macros, to avoid build errors
213 on some platforms for bit 31.
215 JH/16 GnuTLS: rework ciphersuite strings under recent library versions. Thanks
216 to changes apparently associated with TLS1.3 handling some of the APIs
217 previously used were either nonfunctional or inappropriate. Strings
218 like TLS1.3:ECDHE_SECP256R1__RSA_PSS_RSAE_SHA256__AES_256_GCM__AEAD:256
219 and TLS1.2:ECDHE_SECP256R1__RSA_SHA256__AES_128_CBC__SHA256:128 replace
220 the previous TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256 .
221 This affects log line X= elements, the $tls_{in,out}_cipher variables,
222 and the use of specific cipher names in the encrypted= ACL condition.
224 JH/17 OpenSSL: the default openssl_options now disables ssl_v3.
226 JH/18 GnuTLS: fix $tls_out_ocsp under hosts_request_ocsp. Previously the
227 verification result was not updated unless hosts_require_ocsp applied.
229 JH/19 Bug 2398: fix listing of a named-queue. Previously, even with the option
230 queue_list_requires_admin set to false, non-admin users were denied the
233 JH/20 Bug 2389: fix server advertising of usable certificates, under GnuTLS in
234 directory-of-certs mode. Previously they were advertised despite the
237 JH/21 The smtp transport option "hosts_noproxy_tls" is now unset by default.
238 A single TCP connection by a client will now hold a TLS connection open
239 for multiple message deliveries, by default. Previoud the default was to
242 JH/22 The smtp transport option "hosts_try_dane" now enables all hosts by
243 default. If built with the facility, DANE will be used. The facility
244 SUPPORT_DANE is now enabled in the prototype build Makefile "EDITME".
246 JH/23 The build default is now for TLS to be included; the SUPPORT_TLS define
247 is replaced with DISABLE_TLS. Either USE_GNUTLS or (the new) USE_OPENSSL
248 must be defined and you must still, unless you define DISABLE_TLS, manage
249 the the include-dir and library-file requirements that go with that
250 choice. Non-TLS builds are still supported.
252 JH/24 Fix duplicated logging of peer name/address, on a transport connection-
255 JH/25 The smtp transport option "hosts_try_fastopen" now enables all hosts by
256 default. If the platform supports and has the facility enabled, it will
257 be requested on all coneections.
259 JH/26 The PIPE_CONNECT facility is promoted from experimental status and is now
260 controlled by the build-time option SUPPORT_PIPE_CONNECT.
262 PP/01 Unbreak heimdal_gssapi, broken in 4.92.
264 JH/27 Bug 2404: Use the main-section configuration option "dsn_from" for
265 success-DSN messages. Previously the From: header was always the default
266 one for these; the option was ignored.
268 JH/28 Fix the timeout on smtp response to apply to the whole response.
269 Previously it was reset for every read, so a teergrubing peer sending
270 single bytes within the time limit could extend the connection for a
271 long time. Credit to Qualsys Security Advisory Team for the discovery.
273 JH/29 Fix DSN Final-Recipient: field. Previously it was the post-routing
274 delivery address, which leaked information of the results of local
275 forwarding. Change to the original envelope recipient address, per
278 JH/30 Bug 2411: Fix DSN generation when RFC 3461 failure notification is
279 requested. Previously not bounce was generated and a log entry of
280 error ignored was made.
282 JH/31 Avoid re-expansion in ${sort } expansion. (CVE-2019-13917)
284 JH/32 Introduce a general tainting mechanism for values read from the input
285 channel, and values derived from them. Refuse to expand any tainted
286 values, to catch one form of exploit.
288 JH/33 Bug 2413: Fix dkim_strict option. Previously the expansion result
289 was unused and the unexpanded text used for the test. Found and
290 fixed by Ruben Jenster.
292 JH/34 Fix crash after TLS shutdown. When the TCP/SMTP channel was left open,
293 an attempt to use a TLS library read routine dereffed a nul pointer,
296 JH/35 Bug 2409: filter out-of-spec chars from callout response before using
297 them in our smtp response.
299 JH/36 Have the general router option retry_use_local_part default to true when
300 any of the restrictive preconditions are set (to anything). Previously it
301 was only for check_local user. The change removes one item of manual
302 configuration which is required for proper retries when a remote router
303 handles a subset of addresses for a domain.
305 JH/37 Appendfile: when evaluating quota use (non-quota_size_regex) take the file
306 link count into consideration.
308 HS/04 Fix handling of very log lines in -H files. If a -<key> <value> line
309 caused the extension of big_buffer, the following lines were ignored.
311 JH/38 Bug 1395: Teach the DNS negative-cache about TTL value from the SOA in
312 accordance with RFC 2308. Previously there was no expiry, so a longlived
313 receive process (eg. due to ACL delays) versus a short SOA value could
316 HS/05 Handle trailing backslash gracefully. (CVE-2019-15846)
318 JH/39 Promote DMARC support to mainline.
320 JH/40 Bug 2452: Add a References: header to DSNs.
322 JH/41 With GnuTLS 3.6.0 (and later) do not attempt to manage Diffie-Hellman
323 parameters. The relevant library call is documented as "Deprecated: This
324 function is unnecessary and discouraged on GnuTLS 3.6.0 or later. Since
325 3.6.0, DH parameters are negotiated following RFC7919."
327 HS/06 Change the default of dnssec_request_domains to "*"
329 JH/42 Bug 2545: Fix CHUNKING for all RCPT commands rejected. Previously we
330 carried on and emitted a BDAT command, even when PIPELINING was not
333 JH/43 Bug 2465: Fix taint-handling in dsearch lookup. Previously a nontainted
334 buffer was used for the filename, resulting in a trap when tainted
335 arguments (eg. $domain) were used.
337 JH/44 With OpenSSL 1.1.1 (onwards) disable renegotiation for TLS1.2 and below;
338 recommended to avoid a possible server-load attack. The feature can be
339 re-enabled via the openssl_options main cofiguration option.
341 JH/45 local_scan API: documented the current smtp_printf() call. This changed
342 for version 4.90 - adding a "more data" boolean to the arguments.
343 Bumped the ABI version number also, this having been missed previously;
344 release versions 4.90 to 4.92.3 inclusive were effectively broken in
345 respect of usage of smtp_printf() by either local_scan code or libraries
346 accessed via the ${dlfunc } expansion item. Both will need coding
347 adjustment for any calls to smtp_printf() to match the new function
348 signature; a FALSE value for the new argument is always safe.
350 JH/46 FreeBSD: fix use of the sendfile() syscall. The shim was not updating
351 the file-offset (which the Linux syscall does, and exim expects); this
352 resulted in an indefinite loop.
354 JH/47 ARC: fix crash in signing, triggered when a configuration error failed
355 to do ARC verification. The Authentication-Results: header line added
356 by the configuration then had no ARC item.
362 JH/01 Remove code calling the customisable local_scan function, unless a new
363 definition "HAVE_LOCAL_SCAN=yes" is present in the Local/Makefile.
365 JH/02 Bug 1007: Avoid doing logging from signal-handlers, as that can result in
366 non-signal-safe functions being used.
368 JH/03 Bug 2269: When presented with a received message having a stupidly large
369 number of DKIM-Signature headers, disable DKIM verification to avoid
370 a resource-consumption attack. The limit is set at twenty.
372 JH/04 Add variables $arc_domains, $arc_oldest_pass for ARC verify. Fix the
373 report of oldest_pass in ${authres } in consequence, and separate out
374 some descriptions of reasons for verification fail.
376 JH/05 Bug 2273: Cutthrough delivery left a window where the received messsage
377 files in the spool were present and unlocked. A queue-runner could spot
378 them, resulting in a duplicate delivery. Fix that by doing the unlock
379 after the unlink. Investigation by Tim Stewart. Take the opportunity to
380 add more error-checking on spoolfile handling while that code is being
383 PP/01 Refuse to open a spool data file (*-D) if it's a symlink.
384 No known attacks, no CVE, this is defensive hardening.
386 JH/06 Bug 2275: The MIME ACL unlocked the received message files early, and
387 a queue-runner could start a delivery while other operations were ongoing.
388 Cutthrough delivery was a common victim, resulting in duplicate delivery.
389 Found and investigated by Tim Stewart. Fix by using the open message data
390 file handle rather than opening another, and not locally closing it (which
391 releases a lock) for that case, while creating the temporary .eml format
392 file for the MIME ACL. Also applies to "regex" and "spam" ACL conditions.
394 JH/07 Bug 177: Make a random-recipient callout success visible in ACL, by setting
395 $sender_verify_failure/$recipient_verify_failure to "random".
397 JH/08 When generating a selfsigned cert, use serial number 1 since zero is not
400 JH/09 Bug 2274: Fix logging of cmdline args when starting in an unlinked cwd.
401 Previously this would segfault.
403 JH/10 Fix ARC signing for case when DKIM signing failed. Previously this would
406 JH/11 Bug 2264: Exim now only follows CNAME chains one step by default. We'd
407 like zero, since the resolver should be doing this for us, But we need one
408 as a CNAME but no MX presence gets the CNAME returned; we need to check
409 that doesn't point to an MX to declare it "no MX returned" rather than
410 "error, loop". A new main option is added so the older capability of
411 following some limited number of chain links is maintained.
413 JH/12 Add client-ip info to non-pass iprev ${authres } lines.
415 JH/13 For receent Openssl versions (1.1 onward) use modern generic protocol
416 methods. These should support TLS 1.3; they arrived with TLS 1.3 and the
417 now-deprecated earlier definitions used only specified the range up to TLS
418 1.2 (in the older-version library docs).
420 JH/14 Bug 2284: Fix DKIM signing for body lines starting with a pair of dots.
422 JH/15 Rework TLS client-side context management. Stop using a global, and
423 explicitly pass a context around. This enables future use of TLS for
424 connections to service-daemons (eg. malware scanning) while a client smtp
425 connection is using TLS; with cutthrough connections this is quite likely.
427 JH/16 Fix ARC verification to do AS checks in reverse order.
429 JH/17 Support a "tls" option on the ${readsocket } expansion item.
431 JH/18 Bug 2287: Fix the protocol name (eg utf8esmtp) for multiple messages
432 using the SMTPUTF8 option on their MAIL FROM commands, in one connection.
433 Previously the "utf8" would be re-prepended for every additional message.
435 JH/19 Reject MAIL FROM commands with SMTPUTF8 when the facility was not advertised.
436 Previously thery were accepted, resulting in issues when attempting to
437 forward messages to a non-supporting MTA.
439 PP/02 Let -n work with printing macros too, not just options.
441 JH/20 Bug 2296: Fix cutthrough for >1 address redirection. Previously only
442 one parent address was copied, and bogus data was used at delivery-logging
443 time. Either a crash (after delivery) or bogus log data could result.
444 Discovery and analysis by Tim Stewart.
446 PP/03 Make ${utf8clean:} expansion operator detect incomplete final character.
447 Previously if the string ended mid-character, we did not insert the
448 promised '?' replacement.
450 PP/04 Documentation: current string operators work on bytes, not codepoints.
452 JH/21 Change as many as possible of the global flags into one-bit bitfields; these
453 should pack well giving a smaller memory footprint so better caching and
454 therefore performance. Group the declarations where this can't be done so
455 that the byte-sized flag variables are not interspersed among pointer
456 variables, giving a better chance of good packing by the compiler.
458 JH/22 Bug 1896: Fix the envelope from for DMARC forensic reports to be possibly
459 non-null, to avoid issues with sites running BATV. Previously reports were
460 sent with an empty envelope sender so looked like bounces.
462 JH/23 Bug 2318: Fix the noerror command within filters. It wasn't working.
463 The ignore_error flag wasn't being returned from the filter subprocess so
464 was not set for later routers. Investigation and fix by Matthias Kurz.
466 JH/24 Bug 2310: Raise a msg:fail:internal event for each undelivered recipient,
467 and a msg:complete for the whole, when a message is manually removed using
468 -Mrm. Developement by Matthias Kurz, hacked on by JH.
470 JH/25 Avoid fixed-size buffers for pathnames in DB access. This required using
471 a "Gnu special" function, asprintf() in the DB utility binary builds; I
472 hope that is portable enough.
474 JH/26 Bug 2311: Fix DANE-TA verification under GnuTLS. Previously it was also
475 requiring a known-CA anchor certificate; make it now rely entirely on the
476 TLSA as an anchor. Checking the name on the leaf cert against the name
477 on the A-record for the host is still done for TA (but not for EE mode).
479 JH/27 Fix logging of proxy address. Previously, a pointless "PRX=[]:0" would be
480 included in delivery lines for non-proxied connections, when compiled with
481 SUPPORT_SOCKS and running with proxy logging enabled.
483 JH/28 Bug 2314: Fire msg:fail:delivery event even when error is being ignored.
484 Developement by Matthias Kurz, tweaked by JH. While in that bit of code,
485 move the existing event to fire before the normal logging of message
486 failure so that custom logging is bracketed by normal logging.
488 JH/29 Bug 2322: A "fail" command in a non-system filter (file) now fires the
489 msg:fail:internal event. Developement by Matthias Kurz.
491 JH/30 Bug 2329: Increase buffer size used for dns lookup from 2k, which was
492 far too small for todays use of crypto signatures stored there. Go all
493 the way to the max DNS message size of 64kB, even though this might be
494 overmuch for IOT constrained device use.
496 JH/31 Fix a bad use of a copy function, which could be used to pointlessly
497 copy a string over itself. The library routine is documented as not
498 supporting overlapping copies, and on MacOS it actually raised a SIGABRT.
500 JH/32 For main options check_spool_space and check_inode_space, where the
501 platform supports 64b integers, support more than the previous 2^31 kB
502 (i.e. more than 2 TB). Accept E, P and T multipliers in addition to
503 the previous G, M, k.
505 JH/33 Bug 2338: Fix the cyrus-sasl authenticator to fill in the
506 $authenticated_fail_id variable on authentication failure. Previously
509 JH/34 Increase RSA keysize of autogen selfsign cert from 1024 to 2048. RHEL 8.0
510 OpenSSL didn't want to use such a weak key. Do for GnuTLS also, and for
511 more-modern GnuTLS move from GNUTLS_SEC_PARAM_LOW to
512 GNUTLS_SEC_PARAM_MEDIUM.
514 JH/35 OpenSSL: fail the handshake when SNI processing hits a problem, server
515 side. Previously we would continue as if no SNI had been received.
517 JH/36 Harden the handling of string-lists. When a list consisted of a sole
518 "<" character, which should be a list-separator specification, we walked
519 off past the nul-terimation.
521 JH/37 Bug 2341: Send "message delayed" warning MDNs (restricted to external
522 causes) even when the retry time is not yet met. Previously they were
523 not, meaning that when (say) an account was over-quota and temp-rejecting,
524 and multiple senders' messages were queued, only one sender would get
525 notified on each configured delay_warning cycle.
527 JH/38 Bug 2351: Log failures to extract envelope addresses from message headers.
529 JH/39 OpenSSL: clear the error stack after an SSL_accept(). With anon-auth
530 cipher-suites, an error can be left on the stack even for a succeeding
531 accept; this results in impossible error messages when a later operation
534 AM/01 Bug 2359: GnuTLS: repeat lowlevel read and write operations while they
535 return error codes indicating retry. Under TLS1.3 this becomes required.
537 JH/40 Fix the feature-cache refresh for EXPERIMENTAL_PIPE_CONNECT. Previously
538 it only wrote the new authenticators, resulting in a lack of tracking of
539 peer changes of ESMTP extensions until the next cache flush.
541 JH/41 Fix the loop reading a message header line to check for integer overflow,
542 and more-often against header_maxsize. Previously a crafted message could
543 induce a crash of the recive process; now the message is cleanly rejected.
545 JH/42 Bug 2366: Fix the behaviour of the dkim_verify_signers option. It had
546 been totally disabled for all of 4.91. Discovery and fix by "Mad Alex".
552 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
553 When redis_servers is set to a list of > 1 element, and the Redis servers
554 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
555 case of MOVED into a DEFER case instead, thus moving the query onto the
556 next server in the list. For a cluster of N elements, all N servers must
557 be defined in redis_servers.
559 GF/02 Catch and remove uninitialized value warning in exiqsumm
560 Check for existence of @ARGV before looking at $ARGV[0]
562 JH/01 Replace the store_release() internal interface with store_newblock(),
563 which internalises the check required to safely use the old one, plus
564 the allocate and data copy operations duplicated in both (!) of the
565 extant use locations.
567 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
568 modifier. This matches the restriction on the commandline.
570 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
571 Previously only the last row was returned.
573 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
574 we assumed that tags in the header were well-formed, and parsed the
575 element content after inspecting only the first char of the tag.
576 Assumptions at that stage could crash the receive process on malformed
579 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
580 While running the DKIM ACL we operate on the Permanent memory pool so that
581 variables created with "set" persist to the DATA ACL. Also (at any time)
582 DNS lookups that fail create cache records using the Permanent pool. But
583 expansions release any allocations made on the current pool - so a dnsdb
584 lookup expansion done in the DKIM ACL releases the memory used for the
585 DNS negative-cache, and bad things result. Solution is to switch to the
586 Main pool for expansions.
587 While we're in that code, add checks on the DNS cache during store_reset,
588 active in the testsuite.
589 Problem spotted, and debugging aided, by Wolfgang Breyha.
591 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
592 When none of the hosts presented to a transport match an already-open
593 connection, close it and proceed with the list. Previously we would
594 queue the message. Spotted by Lena with Yahoo, probably involving
597 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
598 Previously a spurious "250 OK id=" response was appended to the proper
601 JH/08 The "support for" informational output now, which built with Content
602 Scanning support, has a line for the malware scanner interfaces compiled
603 in. Interface can be individually included or not at build time.
605 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
606 by the template makefile "src/EDITME". The "STREAM" support for an older
607 ClamAV interface method is removed.
609 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
610 rows affected is given instead).
612 JH/11 The runtime Berkeley DB library version is now additionally output by
613 "exim -d -bV". Previously only the compile-time version was shown.
615 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
616 SMTP connection. Previously, when one had more receipients than the
617 first, an abortive onward connection was made. Move to full support for
618 multiple onward connections in sequence, handling cutthrough connection
619 for all multi-message initiating connections.
621 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
622 routers. Previously, a multi-recipient message would fail to match the
623 onward-connection opened for the first recipient, and cause its closure.
625 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
626 a timeout on read on a GnuTLS initiating connection, resulting in the
627 initiating connection being dropped. This mattered most when the callout
628 was marked defer_ok. Fix to keep the two timeout-detection methods
631 JH/15 Relax results from ACL control request to enable cutthrough, in
632 unsupported situations, from error to silently (except under debug)
633 ignoring. This covers use with PRDR, frozen messages, queue-only and
636 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
638 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
639 metadata, resulting in a crash in free().
641 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
642 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
643 Broken also in d185889f4, with init system revamp.
645 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
646 Previously we abruptly closed the connection after reading a malware-
647 found indication; now we go on to read the "scan ok" response line,
650 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
651 ACL. Previously, a crash would result.
653 JH/19 Speed up macro lookups during configuration file read, by skipping non-
654 macro text after a replacement (previously it was only once per line) and
655 by skipping builtin macros when searching for an uppercase lead character.
657 JH/20 DANE support moved from Experimental to mainline. The Makefile control
658 for the build is renamed.
660 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
661 was allocated for every new TLS startup, meaning one per message. Fix
662 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
664 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
665 reported the original. Fix to report (as far as possible) the ACL
666 result replacing the original.
668 JH/23 Fix memory leak during multi-message connections using STARTTLS under
669 OpenSSL. Certificate information is loaded for every new TLS startup,
670 and the resources needed to be freed.
672 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
674 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
677 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
678 DATA response info to the (existing) per-recipient response info for
679 the "C=" log element. It can have useful tracking info from the
680 destination system. Patch from Simon Arlott.
682 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
683 length value. Previously this would segfault.
685 HS/02 Support Avast multiline protoocol, this allows passing flags to
686 newer versions of the scanner.
688 JH/28 Ensure that variables possibly set during message acceptance are marked
689 dead before release of memory in the daemon loop. This stops complaints
690 about them when the debug_store option is enabled. Discovered specifically
691 for sender_rate_period, but applies to a whole set of variables.
692 Do the same for the queue-runner and queue-list loops, for variables set
693 from spool message files. Do the same for the SMTP per-message loop, for
694 certain variables indirectly set in ACL operations.
696 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
697 as a multi-recipient message from a mailinglist manager). The coding had
698 an arbitrary cutoff number of characters while checking for more input;
699 enforced by writing a NUL into the buffer. This corrupted long / fast
700 input. The problem was exposed more widely when more pipelineing of SMTP
701 responses was introduced, and one Exim system was feeding another.
702 The symptom is log complaints of SMTP syntax error (NUL chars) on the
703 receiving system, and refused recipients seen by the sending system
704 (propating to people being dropped from mailing lists).
705 Discovered and pinpointed by David Carter.
707 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
708 replaced by the ${authresults } expansion.
710 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
712 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
713 allows proper process termination in container environments.
715 JH/32 Bug 2258: Fix spool_wireformat in combination with LMTP transport.
716 Previously the "final dot" had a newline after it; ensure it is CR,LF.
718 JH/33 SPF: remove support for the "spf" ACL condition outcome values "err_temp"
719 and "err_perm", deprecated since 4.83 when the RFC-defined words
720 "temperror" and "permerror" were introduced.
722 JH/34 Re-introduce enforcement of no cutthrough delivery on transports having
723 transport-filters or DKIM-signing. The restriction was lost in the
724 consolidation of verify-callout and delivery SMTP handling.
725 Extend the restriction to also cover ARC-signing.
727 JH/35 Cutthrough: for a final-dot response timeout (and nonunderstood responses)
728 in defer=pass mode supply a 450 to the initiator. Previously the message
731 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
732 tls_require_ciphers is used as before.
734 HS/03 Malware Avast: Better match the Avast multiline protocol. Add
735 "pass_unscanned". Only tmpfails from the scanner are written to
736 the paniclog, as they may require admin intervention (permission
737 denied, license issues). Other scanner errors (like decompression
738 bombs) do not cause a paniclog entry.
740 JH/36 Fix reinitialisation of DKIM logging variable between messages.
741 Previously it was possible to log spurious information in receive log
744 JH/37 Bug 2255: Revert the disable of the OpenSSL session caching. This
745 triggered odd behaviour from Outlook Express clients.
747 PP/03 Add util/renew-opendmarc-tlds.sh script for safe renewal of public
750 JH/38 DKIM: accept Ed25519 pubkeys in SubjectPublicKeyInfo-wrapped form,
751 since the IETF WG has not yet settled on that versus the original
752 "bare" representation.
754 JH/39 Fix syslog logging for syslog_timestamp=no and log_selector +millisec.
755 Previously the millisecond value corrupted the output.
756 Fix also for syslog_pid=no and log_selector +pid, for which the pid
757 corrupted the output.
763 JH/01 Rework error string handling in TLS interface so that the caller in
764 more cases is responsible for logging. This permits library-sourced
765 string to be attached to addresses during delivery, and collapses
766 pairs of long lines into single ones.
768 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
769 during configuration. Wildcards are allowed and expanded.
771 JH/02 Rework error string handling in DKIM to pass more info back to callers.
772 This permits better logging.
774 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
775 do not close it down and have the child transport start it up again on
776 the passed-on TCP connection. Instead, proxy the child (and any
777 subsequent ones) for TLS via a unix-domain socket channel. Logging is
778 affected: the continued delivery log lines do not have any DNSSEC, TLS
779 Certificate or OCSP information. TLS cipher information is still logged.
781 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
782 identical IP addresses on different listening ports. Will also affect
785 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
786 add noisy ifdef guards to special-case this sillyness.
787 Patch from Bernd Kuhls.
789 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
790 than 255 are no longer allowed.
792 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
793 Disable the session-cache too, which might reduce our load. Since we
794 currrectly use a new context for every connection, both as server and
795 client, there is no benefit for these.
796 GnuTLS appears to not support tickets server-side by default (we don't
797 call gnutls_session_ticket_enable_server()) but client side is enabled
798 by default on recent versions (3.1.3 +) unless the PFS priority string
801 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
802 <https://reproducible-builds.org/specs/source-date-epoch/>.
804 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
805 the check for any unsuccessful recipients did not notice the limit, and
806 erroneously found still-pending ones.
808 JH/08 Pipeline CHUNKING command and data together, on kernels that support
809 MSG_MORE. Only in-clear (not on TLS connections).
811 JH/09 Avoid using a temporary file during transport using dkim. Unless a
812 transport-filter is involved we can buffer the headers in memory for
813 creating the signature, and read the spool data file once for the
814 signature and again for transmission.
816 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
817 4.77 as the kernel support then wasn't solid, having issues in 64bit
818 mode. Now, it's been long enough. Add support for FreeBSD also.
820 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
821 case where the routing stage had gathered several addresses to send to
822 a host before calling the transport for the first, we previously failed
823 to close down TLS in the old transport process before passing the TCP
824 connection to the new process. The new one sent a STARTTLS command
825 which naturally failed, giving a failed delivery and bloating the retry
826 database. Investigation and fix prototype from Wolfgang Breyha.
828 JH/12 Fix check on SMTP command input synchronisation. Previously there were
829 false-negatives in the check that the sender had not preempted a response
830 or prompt from Exim (running as a server), due to that code's lack of
831 awareness of the SMTP input buffering.
833 PP/04 Add commandline_checks_require_admin option.
834 Exim drops privileges sanely, various checks such as -be aren't a
835 security problem, as long as you trust local users with access to their
836 own account. When invoked by services which pass untrusted data to
837 Exim, this might be an issue. Set this option in main configuration
838 AND make fixes to the calling application, such as using `--` to stop
841 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
842 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
843 responses to those, into a single TLS record each way (this usually means
844 a single packet). As a side issue, smtp_enforce_sync now works on TLS
847 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
848 affects you only if you're dancing at the edge of the param size limits.
849 If you are, and this message makes sense to you, then: raise the
850 configured limit or use OpenSSL 1.1. Nothing we can do for older
853 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
854 cmdline element to get the documented default one. Previously it was
857 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
858 are unsuable for non-message-specific reasons - eg. connection timeout,
859 banner-time rejection.
861 JH/16 Fix logging of delivery remote port, when specified by router, under
864 PP/06 Repair manualroute's ability to take options in any order, even if one
865 is the name of a transport.
868 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
870 JH/17 Change the list-building routines interface to use the expanding-string
871 triplet model, for better allocation and copying behaviour.
873 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
874 Previously it was constructed the first time a possibly-matching string
875 was met in the configuration file input during startup; now it is done
878 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
879 compatible one, to avoid the (poorly documented) possibility of a config
880 file in the working directory redirecting the DB files, possibly correpting
881 some existing file. CVE-2017-10140 assigned for BDB.
883 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
884 cache-hot. Previously, although the result was properly cached, the
885 initial verify call returned a defer.
887 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
888 the main verify for receipient in uncached-mode.
890 JH/22 Retire historical build files to an "unsupported" subdir. These are
891 defined as "ones for which we have no current evidence of testing".
893 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
894 if present. Previously it was ignored.
896 JH/24 Start using specified-initialisers in C structure init coding. This is
897 a C99 feature (it's 2017, so now considered safe).
899 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
900 if was a fixed-sized field and bitmask ops via macros; it is now more
903 PP/07 GitHub PR 56: Apply MariaDB build fix.
904 Patch provided by Jaroslav Škarvada.
906 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
907 during Coverity cleanups [4.87 JH/47]
908 Diagnosis and fix provided by Michael Fischer v. Mollard.
910 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
911 the right size to place the terminating semicolon on its own folded
912 line, the header hash was calculated to an incorrect value thanks to
913 the (relaxed) space the fold became.
915 HS/02 Fix Bug 2130: large writes from the transport subprocess were chunked
916 and confused the parent.
918 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
919 which could crash as a result. This could lead to undeliverable messages.
921 JH/28 Logging: "next input sent too soon" now shows where input was truncated
924 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
925 matters on fast-turnover and PID-randomising systems, which were getting
926 out-of-order delivery.
928 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
929 a possibly-overlapping copy. The symptom was that "Remote host closed
930 connection in response to HELO" was logged instead of the actual 4xx
933 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
934 Previously only that bufferd was discarded, resulting in SYMTP command
937 JH/32 DKIM: when a message has multiple signatures matching an identity given
938 in dkim_verify_signers, run the dkim acl once for each. Previously only
939 one run was done. Bug 2189.
941 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
942 "panic the current process" to "deliberately defer". The panic log is
943 still written with the problem list name; the mail and reject logs now
944 get a temp-reject line for the message that was being handled, saying
945 something like "domains check lookup or other defer". The SMTP 451
946 message is still "Temporary local problem".
948 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
949 A crafted sequence of BDAT commands could result in in-use memory beeing
950 freed. CVE-2017-16943.
952 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
953 from SMTP input. Previously it was always done; now only done for DATA
954 and not BDAT commands. CVE-2017-16944.
956 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
957 to the message (such as an overlong header line). Previously this was
958 not done and we did not exit BDAT mode. Followon from the previous item
959 though a different problem.
965 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
966 than -2003 did; needs libidn2 in addition to libidn.
968 JH/02 The path option on a pipe transport is now expanded before use.
970 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
971 Patch provided by "Björn", documentation fix added too.
973 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
974 missing a wire-to-host endian conversion.
976 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
977 close after a BDAT command line could be taken as a following command,
978 giving a synch failure. Fix by only checking for synch immediately
979 before acknowledging the chunk.
981 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
982 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
983 macro. Patches provided by Josh Soref.
985 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
986 Previously we did not; the RFC seems ambiguous and VRFY is not listed
987 by IANA as a service extension. However, John Klensin suggests that we
990 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
991 the dkim code may be unix-mode line endings rather than smtp wire-format
992 CRLF, so prepend a CR to any bare LF.
994 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
995 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
997 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
998 the first were themselves being wrongly included in the feed into dkim
999 processing; with most chunk sizes in use this resulted in an incorrect
1000 body hash calculated value.
1002 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
1003 DKIM signature block, for verification. Although advised against by
1004 standards it is specifically not ruled illegal.
1006 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
1008 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
1009 missing a body hash (the bh= tag).
1011 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
1012 It seems that HAProxy sends the Proxy Protocol information in clear and
1013 only then does a TLS startup, so do the same.
1015 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
1016 TCP connections (such as for Spamd) unless the daemon successfully set
1017 Fast Open mode on its listening sockets. This fixes breakage seen on
1018 too-old kernels or those not configured for Fast Open, at the cost of
1019 requiring both directions being enabled for TFO, and TFO never being used
1020 by non-daemon-related Exim processes.
1022 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
1023 endings, at least on the first header line. Try to canonify any that get
1024 past that check, despite the cost.
1026 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
1027 now limited to an arbitrary five deep, while parsing addresses with the
1028 strip_excess_angle_brackets option enabled.
1030 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
1031 instead leave the unprompted TLS handshake in socket buffer for the
1032 TLS library to consume.
1034 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
1036 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
1038 JH/16 Drop variables when they go out of scope. Memory management drops a whole
1039 region in one operation, for speed, and this leaves assigned pointers
1040 dangling. Add checks run only under the testsuite which checks all
1041 variables at a store-reset and panics on a dangling pointer; add code
1042 explicitly nulling out all the variables discovered. Fixes one known
1043 bug: a transport crash, where a dangling pointer for $sending_ip_address
1044 originally assigned in a verify callout, is re-used.
1046 PP/06 Drop '.' from @INC in various Perl scripts.
1048 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
1050 PP/08 Reduce a number of compilation warnings under clang; building with
1051 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
1052 should be warning-free.
1054 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
1056 HS/01 Fix portability problems introduced by PP/08 for platforms where
1057 realloc(NULL) is not equivalent to malloc() [SunOS et al].
1059 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
1060 chunk. This allows us to accept broken chunked messages. We need a more
1061 general solution here.
1063 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
1064 already-broken messages in the queue.
1066 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
1068 JH/19 Fix reference counting bug in routing-generated-address tracking.
1074 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
1075 supports it and a size is available (ie. the sending peer gave us one).
1077 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
1078 years of being deprecated). The replacements are the ACLs
1079 acl_smtp_mime and acl_not_smtp_mime.
1081 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
1082 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
1083 or even an in-clear connection were permitted. Now, if the host lookup
1084 was dnssec and dane was requested then the host is only used if the
1085 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
1086 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
1087 if one fails this test.
1088 This means that a poorly-configured remote DNS will make it incommunicado;
1089 but it protects against a DNS-interception attack on it.
1091 JH/04 Bug 1810: make continued-use of an open smtp transport connection
1092 non-noisy when a race steals the message being considered.
1094 JH/05 If main configuration option tls_certificate is unset, generate a
1095 self-signed certificate for inbound TLS connections.
1097 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
1098 in rewrites and routers.
1100 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
1101 and logged a warning sing 4.83; now they are a configuration file error.
1103 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
1104 (lacking @domain). Apply the same qualification processing as RCPT.
1106 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
1108 JH/10 Support ${sha256:} applied to a string (as well as the previous
1111 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
1112 a cutthrough deliver is pending, as we always want to make a connection.
1113 This also avoids re-routing the message when later placing the cutthrough
1114 connection after a verify cache hit.
1115 Do not update it with the verify result either.
1117 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
1118 when routing results in more than one destination address.
1120 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
1121 signing (which inhibits the cutthrough capability). Previously only
1122 the presence of an option was tested; now an expansion evaluating as
1123 empty is permissible (obviously it should depend only on data available
1124 when the cutthrough connection is made).
1126 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
1127 the relevant preceding SMTP command did not note the pipelining mode.
1129 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
1130 Previously they were not counted.
1132 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
1133 as one having no matching records. Previously we deferred the message
1134 that needed the lookup.
1136 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
1137 distinguished as "(=".
1139 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
1140 for missing MX records. Previously it only worked for missing A records.
1142 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
1144 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
1145 after the data-go-ahead and data-ack. Patch from Jason Betts.
1147 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
1148 even for a "none" policy. Patch from Tony Meyer.
1150 JH/22 Fix continued use of a connection for further deliveries. If a port was
1151 specified by a router, it must also match for the delivery to be
1154 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
1155 When one of the recipients of a message was unsuitable for the connection
1156 (has no matching addresses), we lost track of needing to mark it
1157 deferred. As a result mail would be lost.
1159 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
1161 JH/25 Decoding ACL controls is now done using a binary search; the source code
1162 takes up less space and should be simpler to maintain. Merge the ACL
1163 condition decode tables also, with similar effect.
1165 JH/26 Fix problem with one_time used on a redirect router which returned the
1166 parent address unchanged. A retry would see the parent address marked as
1167 delivered, so not attempt the (identical) child. As a result mail would
1170 JH/27 Fix a possible security hole, wherein a process operating with the Exim
1171 UID can gain a root shell. Credit to http://www.halfdog.net/ for
1172 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
1175 JH/28 Enable {spool,log} filesystem space and inode checks as default.
1176 Main config options check_{log,spool}_{inodes,space} are now
1177 100 inodes, 10MB unless set otherwise in the configuration.
1179 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
1180 Previously it only applied to the main-section connection policy
1183 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
1185 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
1186 by me. Added RFC7919 DH primes as an alternative.
1188 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
1189 are not in the system include path.
1191 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
1192 GnuTLS, when a session startup failed (eg because the client disconnected)
1193 Exim did stdio operations after fclose. This was exposed by a recent
1194 change which nulled out the file handle after the fclose.
1196 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
1197 signed directly by the cert-signing cert, rather than an intermediate
1198 OCSP-signing cert. This is the model used by LetsEncrypt.
1200 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
1202 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
1203 an incoming connection.
1205 HS/02 Bug 1802: Do not half-close the connection after sending a request
1208 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
1209 fallback to "prime256v1".
1211 JH/34 SECURITY: Use proper copy of DATA command in error message.
1212 Could leak key material. Remotely exploitable. CVE-2016-9963.
1218 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
1219 and 3.4.4 - once the server is enabled to respond to an OCSP request
1220 it does even when not requested, resulting in a stapling non-aware
1221 client dropping the TLS connection.
1223 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
1224 support variable-length bit vectors. No functional change.
1226 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
1227 The I= interface field on outgoing lines is now after the H= remote
1228 host field, same as incoming lines. There is a separate
1229 outgoing_interface log selector which allows you to disable the
1232 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
1233 If not running log_selector +smtp_connection the mainlog would be held
1234 open indefinitely after a "too many connections" event, including to a
1235 deleted file after a log rotate. Leave the per net connection logging
1236 leaving it open for efficiency as that will be quickly detected by the
1237 check on the next write.
1239 HS/01 Bug 1671: Fix post transport crash.
1240 Processing the wait-<transport> messages could crash the delivery
1241 process if the message IDs didn't exist for some reason. When
1242 using 'split_spool_directory=yes' the construction of the spool
1243 file name failed already, exposing the same netto behaviour.
1245 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
1246 mime_regex ACL conditions.
1248 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
1249 to DSN fail messages (bounces): remote IP, remote greeting, remote response
1250 to HELO, local diagnostic string.
1252 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
1253 log line to debug. Even when configured with a tls authenticator many
1254 client connections are expected to not authenticate in this way, so
1255 an authenticate fail is not an error.
1257 HS/02 Add the Exim version string to the process info. This way exiwhat
1258 gives some more detail about the running daemon.
1260 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
1261 matter for fast-change records such as DNSBLs.
1263 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
1264 retry record, even if constant. There may be multiple transports with
1265 different interface settings and the retry behaviour needs to be kept
1268 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
1270 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
1272 JH/10 Bug 840: fix log_defer_output option of pipe transport
1274 JH/11 Bug 830: use same host for all RCPTS of a message, even under
1275 hosts_randomize. This matters a lot when combined with mua_wrapper.
1277 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
1278 ${quote_pgsql:<string>} operator.
1280 JH/13 Bug 1708: avoid misaligned access in cached lookup.
1282 JH/14 Change header file name for freeradius-client. Relevant if compiling
1283 with Radius support; from the Gentoo tree and checked under Fedora.
1285 JH/15 Bug 1712: Introduce $prdr_requested flag variable
1287 JH/16 Bug 1714: Permit an empty string as expansion result for transport
1288 option transport_filter, meaning no filtering.
1290 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
1292 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
1293 defaults to "*" (all hosts). The variable is now available when not built
1294 with TLS, default unset, mainly to enable keeping the testsuite sane.
1295 If a server certificate is not supplied (via tls_certificate) an error is
1296 logged, and clients will find TLS connections fail on startup. Presumably
1297 they will retry in-clear.
1298 Packagers of Exim are strongly encouraged to create a server certificate
1299 at installation time.
1301 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
1302 with the $config_file variable.
1304 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
1305 in transport context, after the attempt, and per-recipient. The latter type
1306 is per host attempted. The event data is the error message, and the errno
1307 information encodes the lookup type (A vs. MX) used for the (first) host,
1308 and the trailing two digits of the smtp 4xx response.
1310 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
1311 to write to mainlog (or rejectlog, paniclog) in the window between file
1312 creation and permissions/ownership being changed. Particularly affects
1313 installations where exicyclog is run as root, rather than exim user;
1314 result is that the running daemon panics and dies.
1316 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
1318 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
1319 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
1320 "pri" and "weight". Note that the previous implicit priority given by the
1321 list order is no longer honoured.
1323 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
1324 for DKIM processing.
1326 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
1327 by defining SUPPORT_SOCKS.
1329 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
1330 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
1331 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
1332 variables are renamed to proxy_{local,external}_{address,port}.
1334 JH/27 Move Internationalisation support from Experimental to mainline, enabled
1335 for a build by defining SUPPORT_I18N
1337 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
1338 of the query string, and make ${quote_redis:} do that quoting.
1340 JH/29 Move Events support from Experimental to mainline, enabled by default
1341 and removable for a build by defining DISABLE_EVENT.
1343 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
1345 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
1346 cached by the daemon.
1348 JH/32 Move Redis support from Experimental to mainline, enabled for a build
1349 by defining LOOKUP_REDIS. The libhiredis library is required.
1351 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
1352 keys are given for lookup.
1354 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
1355 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
1356 only supported when built with TLS support. The PolarSSL SHA routines
1357 are still used when the TLS library is too old for convenient support.
1359 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
1360 openssl_options), for security. OpenSSL forces this from version 1.1.0
1361 server-side so match that on older versions.
1363 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
1364 allocation for $value could be released as the expansion processing
1365 concluded, but leaving the global pointer active for it.
1367 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
1368 and to use the domains and local_parts ACL conditions.
1370 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
1371 incorrectly not doubled on cutthrough transmission, hence seen as a
1372 body-termination at the receiving system - resulting in truncated mails.
1373 Commonly the sender saw a TCP-level error, and retransmitted the message
1374 via the normal store-and-forward channel. This could result in duplicates
1375 received - but deduplicating mailstores were liable to retain only the
1376 initial truncated version.
1378 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
1380 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
1382 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
1383 we're in there, support oversigning also; bug 1309.
1385 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
1387 HS/04 Add support for keep_environment and add_environment options.
1389 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
1390 either intentional arithmetic overflow during PRNG, or testing config-
1393 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
1394 delivery resulted in actual delivery. Cancel cutthrough before DATA
1397 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
1398 rejects a recipient: pass the reject to the originator.
1400 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
1401 Many were false-positives and ignorable, but it's worth fixing the
1404 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
1405 for the new environment-manipulation done at startup. Move the routines
1406 from being local to tls.c to being global via the os.c file.
1408 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
1409 an extract embedded as result-arg for a map, the first arg for extract
1410 is unavailable so we cannot tell if this is a numbered or keyed
1411 extraction. Accept either.
1417 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
1420 JH/02 The smtp transport option "multi_domain" is now expanded.
1422 JH/03 The smtp transport now requests PRDR by default, if the server offers
1425 JH/04 Certificate name checking on server certificates, when exim is a client,
1426 is now done by default. The transport option tls_verify_cert_hostnames
1427 can be used to disable this per-host. The build option
1428 EXPERIMENTAL_CERTNAMES is withdrawn.
1430 JH/05 The value of the tls_verify_certificates smtp transport and main options
1431 default to the word "system" to access the system default CA bundle.
1432 For GnuTLS, only version 3.0.20 or later.
1434 JH/06 Verification of the server certificate for a TLS connection is now tried
1435 (but not required) by default. The verification status is now logged by
1436 default, for both outbound TLS and client-certificate supplying inbound
1439 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
1442 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
1443 Status Notification (bounce) messages are now MIME format per RFC 3464.
1444 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
1445 under the control of the dsn_advertise_hosts option, and routers may
1446 have a dsn_lasthop option.
1448 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
1449 default, modifiable by a malware= option. The list separator for
1450 the options can now be changed in the usual way. Bug 68.
1452 JH/10 The smtp_receive_timeout main option is now expanded before use.
1454 JH/11 The incoming_interface log option now also enables logging of the
1455 local interface on delivery outgoing connections.
1457 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
1458 if the interface and destination host and port all match.
1460 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
1463 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
1464 Patch from Andrew Lewis.
1466 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
1467 now supports optional time-restrictions, weighting, and priority
1468 modifiers per server. Patch originally by <rommer@active.by>.
1470 JH/16 The spamd_address main option now supports a mixed list of local
1471 and remote servers. Remote servers can be IPv6 addresses, and
1472 specify a port-range.
1474 JH/17 Bug 68: The spamd_address main option now supports an optional
1475 timeout value per server.
1477 JH/18 Bug 1581: Router and transport options headers_add/remove can
1478 now have the list separator specified.
1480 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
1483 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
1486 JH/21 Support for the A6 type of dns record is withdrawn.
1488 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
1489 rather than the verbs used.
1491 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
1492 from 255 to 1024 chars.
1494 JH/24 Verification callouts now attempt to use TLS by default.
1496 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
1497 are generic router options now. The defaults didn't change.
1499 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
1500 Original patch from Alexander Shikoff, worked over by JH.
1502 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
1503 files if ZCAT_COMMAND is not executable.
1505 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
1507 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
1509 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
1510 Normally benign, it bites when the pair was led to by a CNAME;
1511 modern usage is to not canonicalize the domain to a CNAME target
1512 (and we were inconsistent anyway for A-only vs AAAA+A).
1514 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
1516 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
1517 when evaluating $sender_host_dnssec.
1519 JH/31 Check the HELO verification lookup for DNSSEC, adding new
1520 $sender_helo_dnssec variable.
1522 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
1524 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
1526 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
1528 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
1529 documented as working, but never had. Support all but $spam_report.
1531 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
1532 added for tls authenticator.
1534 HS/03 Add perl_taintmode main config option
1540 TL/01 When running the test suite, the README says that variables such as
1541 no_msglog_check are global and can be placed anywhere in a specific
1542 test's script, however it was observed that placement needed to be near
1543 the beginning for it to behave that way. Changed the runtest perl
1544 script to read through the entire script once to detect and set these
1545 variables, reset to the beginning of the script, and then run through
1546 the script parsing/test process like normal.
1548 TL/02 The BSD's have an arc4random API. One of the functions to induce
1549 adding randomness was arc4random_stir(), but it has been removed in
1550 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
1551 function when detected.
1553 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
1554 cause callback expansion.
1556 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
1557 syntax errors in an expansion can be treated as a string instead of
1558 logging or causing an error, due to the internal use of bool_lax
1559 instead of bool when processing it.
1561 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
1562 server certificates when making smtp deliveries.
1564 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
1566 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
1568 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1570 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1571 Merged patch from Sebastian Wiedenroth.
1573 JH/05 Fix results-pipe from transport process. Several recipients, combined
1574 with certificate use, exposed issues where response data items split
1575 over buffer boundaries were not parsed properly. This eventually
1576 resulted in duplicates being sent. This issue only became common enough
1577 to notice due to the introduction of connection certificate information,
1578 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1580 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1581 size buffer was used, resulting in syntax errors when an expansion
1584 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1585 version 3.3.6 or later.
1587 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1588 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1589 both become "event_action", the variables become $event_name, $event_data
1590 and $event_defer_errno. There is a new variable $verify_mode, usable in
1591 routers, transports and related events. The tls:cert event is now also
1592 raised for inbound connections, if the main configuration event_action
1595 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1596 early OCSP support, but no stapling (appears to be less than 1.0.0).
1598 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1599 server certificate names available under the smtp transport option
1600 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1603 JH/10 Time-related extraction expansions from certificates now use the main
1604 option "timezone" setting for output formatting, and are consistent
1605 between OpenSSL and GnuTLS compilations. Bug 1541.
1607 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1608 encoded parameter in the incoming message. Bug 1558.
1610 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1611 include certificate info, eximon was claiming there were spoolfile
1614 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1616 JH/14 Log delivery-related information more consistently, using the sequence
1617 "H=<name> [<ip>]" wherever possible.
1619 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1620 are problematic for Debian distribution, omit them from the release
1623 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1625 JH/16 Fix string representation of time values on 64bit time_t architectures.
1628 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1629 output list separator was used.
1634 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1635 checkers that were complaining about end of non-void function with no
1638 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1639 This was a regression introduced in 4.83 by another bugfix.
1641 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1643 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1644 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1650 TF/01 Correctly close the server side of TLS when forking for delivery.
1652 When a message was received over SMTP with TLS, Exim failed to clear up
1653 the incoming connection properly after forking off the child process to
1654 deliver the message. In some situations the subsequent outgoing
1655 delivery connection happened to have the same fd number as the incoming
1656 connection previously had. Exim would try to use TLS and fail, logging
1657 a "Bad file descriptor" error.
1659 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1660 utilities have not been installed.
1662 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1663 temporary space as the ACL may create new global variables.
1665 TL/01 LDAP support uses per connection or global context settings, depending
1666 upon the detected version of the libraries at build time.
1668 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1669 to extract and use the src ip:port in logging and expansions as if it
1670 were a direct connection from the outside internet. PPv2 support was
1671 updated based on HAProxy spec change in May 2014.
1673 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1675 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1676 Properly escape header and check for NULL return.
1678 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1681 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1683 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1684 characters in header names, implemented as a verify condition.
1685 Contributed by Michael Fischer v. Mollard.
1687 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1688 results permerror and temperror. Previous values are deprecated but
1689 still accepted. In a future release, err_perm and err_temp will be
1690 completely removed, which will be a backward incompatibility if the
1691 ACL tests for either of these two old results. Patch contributed by
1692 user bes-internal on the mailing list.
1694 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1696 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1697 selectors, in both main and reject logs.
1699 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1702 JH/07 Add malware type "sock" for talking to simple daemon.
1704 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1706 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1707 routers/transports under cutthrough routing.
1709 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1710 numbers. Touch up "bool" conditional to keep the same definition.
1712 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1714 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1716 JH/12 Expand items in router/transport headers_add or headers_remove lists
1717 individually rather than the list as a whole. Bug 1452.
1719 Required for reasonable handling of multiple headers_ options when
1720 they may be empty; requires that headers_remove items with embedded
1721 colons must have them doubled (or the list-separator changed).
1723 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1724 view the policy declared in the DMARC record. Currently, $dmarc_status
1725 is a combined value of both the record presence and the result of the
1728 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1730 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1731 dnslookup router and the smtp transport (applying to the forward
1734 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1735 of ldap servers used for a specific lookup. Patch provided by Heiko
1738 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1739 New variable $lookup_dnssec_authenticated for observability.
1741 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1742 Patch submitted by Lars Timman.
1744 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1746 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1747 Requires trusted mode and valid format message id, aborts otherwise.
1748 Patch contributed by Heiko Schlichting.
1750 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1751 certextract with support for various fields. Bug 1358.
1753 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1754 is requested by default, modifiable by smtp transport option
1757 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1758 operate on certificate variables to give certificate fingerprints
1759 Also new ${sha256:cert_variable}.
1761 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1763 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1764 Christian Aistleitner.
1766 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1768 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1769 file. Patch from Wolfgang Breyha.
1771 JH/25 Expand the coverage of the delivery $host and $host_address to
1772 client authenticators run in verify callout. Bug 1476.
1774 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1775 align with daemon_smtp_ports. Bug 72.
1777 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1778 support and error reporting did not work properly.
1780 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1781 and is readable. Patch from Andrew Colin Kissa.
1783 TL/14 Enhance documentation of ${run expansion and how it parses the
1784 commandline after expansion, particularly in the case when an
1785 unquoted variable expansion results in an empty value.
1787 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1789 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1790 by renaming away from it. Addresses GH issue 15, reported by
1793 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1795 TL/15 SECURITY: prevent double expansion in math comparison functions
1796 (can expand unsanitized data). Not remotely exploitable.
1803 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1805 PP/02 Make -n do something, by making it not do something.
1806 When combined with -bP, the name of an option is not output.
1808 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1811 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1812 $sender_host_name and config options to manage this, and basic check
1815 PP/05 DSCP support for outbound connections and control modifier for inbound.
1817 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1818 (Only plugin which currently uses this is kerberos4, which nobody should
1819 be using, but we should make it available and other future plugins might
1820 conceivably use it, even though it would break NAT; stuff *should* be
1821 using channel bindings instead).
1823 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1824 name; added for Sendmail compatibility; requires admin caller.
1825 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1826 just ignore it); requires trusted caller.
1827 Also parse but ignore: -Ac -Am -X<logfile>
1830 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1832 TL/02 Add +smtp_confirmation as a default logging option.
1834 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1835 Patch by Magnus Holmgren from 2007-02-20.
1837 TL/04 Bugzilla 1281 - Spec typo.
1838 Bugzilla 1283 - Spec typo.
1839 Bugzilla 1290 - Spec grammar fixes.
1841 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1843 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1845 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1846 mailing list by by Dmitry Isaikin.
1848 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1850 JH/02 Support "G" suffix to numbers in ${if comparisons.
1852 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1854 NM/01 Bugzilla 1197 - Spec typo
1855 Bugzilla 1196 - Spec examples corrections
1857 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1859 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1860 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1863 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1864 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1866 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1867 "acl {{name}{arg}...}", and optional args on acl condition
1870 JH/05 Permit multiple router/transport headers_add/remove lines.
1872 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1874 JH/07 Avoid using a waiting database for a single-message-only transport.
1875 Performance patch from Paul Fisher. Bugzilla 1262.
1877 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1880 JH/09 Add $headers_added variable, with content from use of ACL modifier
1881 add_header (but not yet added to the message). Bugzilla 199.
1883 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1884 Pulled from Bugzilla 817 by Wolfgang Breyha.
1886 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1888 (nb: this is the same fix as in Exim 4.80.1)
1890 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1893 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1896 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1898 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1899 advertise SMTP AUTH mechanism to us, instead of a generic
1900 protocol violation error. Also, make Exim more robust to bad
1901 data from the Dovecot auth socket.
1903 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1905 When a queue runner is handling a message, Exim first routes the
1906 recipient addresses, during which it prunes them based on the retry
1907 hints database. After that it attempts to deliver the message to
1908 any remaining recipients. It then updates the hints database using
1911 So if a recipient address works intermittently, it can get repeatedly
1912 deferred at routing time. The retry hints record remains fresh so the
1913 address never reaches the final cutoff time.
1915 This is a fairly common occurrence when a user is bumping up against
1916 their storage quota. Exim had some logic in its local delivery code
1917 to deal with this. However it did not apply to per-recipient defers
1918 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1920 This change adds a proper retry rule check during routing so that the
1921 final cutoff time is checked against the message's age. We only do
1922 this check if there is an address retry record and there is not a
1923 domain retry record; this implies that previous attempts to handle
1924 the address had the retry_use_local_parts option turned on. We use
1925 this as an approximation for the destination being like a local
1926 delivery, as in LMTP.
1928 I suspect this new check makes the old local delivery cutoff check
1929 redundant, but I have not verified this so I left the code in place.
1931 TF/02 Correct gecos expansion when From: is a prefix of the username.
1933 Test 0254 submits a message to Exim with the header
1937 When I ran the test suite under the user fanf2, Exim expanded
1938 the header to contain my full name, whereas it should have added
1939 a Resent-Sender: header. It erroneously treats any prefix of the
1940 username as equal to the username.
1942 This change corrects that bug.
1944 GF/01 DCC debug and logging tidyup
1945 Error conditions log to paniclog rather than rejectlog.
1946 Debug lines prefixed by "DCC: " to remove any ambiguity.
1948 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1950 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1951 Bug spotted by Jeremy Harris; was flawed since initial commit.
1952 Would have resulted in OCSP responses post-SNI triggering an Exim
1953 NULL dereference and crash.
1955 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1957 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1958 Bug detection, analysis and fix by Samuel Thibault.
1959 Bugzilla 1331, Debian bug #698092.
1961 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1963 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1964 Server implementation by Todd Lyons, client by JH.
1965 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1966 config variable "prdr_enable" controls whether the server
1967 advertises the facility. If the client requests PRDR a new
1968 acl_data_smtp_prdr ACL is called once for each recipient, after
1969 the body content is received and before the acl_smtp_data ACL.
1970 The client is controlled by both of: a hosts_try_prdr option
1971 on the smtp transport, and the server advertisement.
1972 Default client logging of deliveries and rejections involving
1973 PRDR are flagged with the string "PRDR".
1975 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1976 fclose(). Diagnosis by Todd Lyons.
1978 PP/17 Update configure.default to handle IPv6 localhost better.
1979 Patch by Alain Williams (plus minor tweaks).
1982 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1983 This is now consistent with GnuTLS, and is now documented: the
1984 previous undocumented portable approach to treating the option as
1985 unset was to force an expansion failure. That still works, and
1986 an empty string is now equivalent.
1988 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1989 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1990 not performing validation itself.
1992 PP/20 Added force_command boolean option to pipe transport.
1993 Patch from Nick Koston, of cPanel Inc.
1995 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1998 TF/04 Added udpsend ACL modifier and hexquote expansion operator
2000 PP/21 Fix eximon continuous updating with timestamped log-files.
2001 Broken in a format-string cleanup in 4.80, missed when I repaired the
2002 other false fix of the same issue.
2003 Report and fix from Heiko Schlichting.
2006 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
2007 Report from Prashanth Katuri.
2009 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
2010 It's SecureTransport, so affects any MacOS clients which use the
2011 system-integrated TLS libraries, including email clients.
2013 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
2014 using a MIME ACL for non-SMTP local injection.
2015 Report and assistance in diagnosis by Warren Baker.
2017 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
2019 JH/16 Fix comparisons for 64b. Bugzilla 1385.
2021 TL/09 Add expansion variable $authenticated_fail_id to keep track of
2022 last id that failed so it may be referenced in subsequent ACL's.
2024 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
2027 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
2028 ldap library initialization, allowing self-signed CA's to be
2029 used. Also properly sets require_cert option later in code by
2030 using NULL (global ldap config) instead of ldap handle (per
2031 session). Bug diagnosis and testing by alxgomz.
2033 TL/12 Enhanced documentation in the ratelimit.pl script provided in
2034 the src/util/ subdirectory.
2036 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
2037 renamed to Transport Post Delivery Action by Jeremy Harris, as
2040 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
2041 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
2042 redis_servers = needs to be configured which will be used by the redis
2043 lookup. Patch from Warren Baker, of The Packet Hub.
2045 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
2047 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
2048 hostname or reverse DNS when processing a host list. Used suggestions
2049 from multiple comments on this bug.
2051 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
2053 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
2054 Missed a few lines, added it to make the runtest require no keyboard
2057 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
2058 contains upper case chars. Make router use caseful_local_part.
2060 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
2061 support when GnuTLS has been built with p11-kit.
2067 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
2069 This, or similar/improved, will also be change PP/11 of 4.82.
2075 PP/01 Handle short writes when writing local log-files.
2076 In practice, only affects FreeBSD (8 onwards).
2077 Bugzilla 1053, with thanks to Dmitry Isaikin.
2079 NM/01 Bugzilla 949 - Documentation tweak
2081 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
2084 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
2086 PP/02 Implemented gsasl authenticator.
2088 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
2090 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
2091 `pkg-config foo` for cflags/libs.
2093 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
2094 with rest of GSASL and with heimdal_gssapi.
2096 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
2097 `pkg-config foo` for cflags/libs for the TLS implementation.
2099 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
2100 properties get this fed in as external SSF. A number of robustness
2101 and debugging improvements to the cyrus_sasl authenticator.
2103 PP/08 cyrus_sasl server now expands the server_realm option.
2105 PP/09 Bugzilla 1214 - Log authentication information in reject log.
2106 Patch by Jeremy Harris.
2108 PP/10 Added dbmjz lookup type.
2110 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
2112 PP/12 MAIL args handles TAB as well as SP, for better interop with
2113 non-compliant senders.
2114 Analysis and variant patch by Todd Lyons.
2116 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
2117 Bug report from Lars Müller <lars@samba.org> (via SUSE),
2118 Patch from Dirk Mueller <dmueller@suse.com>
2120 PP/13 tls_peerdn now print-escaped for spool files.
2121 Observed some $tls_peerdn in wild which contained \n, which resulted
2122 in spool file corruption.
2124 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
2125 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
2126 or write after TLS renegotiation, which otherwise led to messages
2129 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
2130 as a tracking header (ie: a signed header comes before the signature).
2131 Patch from Wolfgang Breyha.
2133 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
2134 comma-sep list; embedded commas doubled.
2136 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
2138 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
2140 Report and patch from Dmitry Banschikov.
2142 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
2143 Removed SSL_clear() after SSL_new() which led to protocol negotiation
2144 failures. We appear to now support TLS1.1+ with Exim.
2146 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
2147 lets Exim select keys and certificates based upon TLS SNI from client.
2148 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
2149 before an outbound SMTP session. New log_selector, +tls_sni.
2151 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
2152 NULL dereference. Report and patch from Alun Jones.
2154 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
2155 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
2156 Not seeing resolver debug output on NetBSD, but suspect this is a
2157 resolver implementation change.
2159 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
2160 Left warnings. Added "eximon gdb" invocation mode.
2162 PP/21 Defaulting "accept_8bitmime" to true, not false.
2164 PP/22 Added -bw for inetd wait mode support.
2166 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
2167 locate the relevant includes and libraries. Made this the default.
2169 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
2170 Bugzilla 1246, report and most of solution from Tomasz Kusy.
2172 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
2173 This may cause build issues on older platforms.
2175 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
2176 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
2177 gnutls_require_mac & gnutls_require_protocols (no longer supported).
2178 Added SNI support via GnuTLS too.
2179 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
2181 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
2183 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
2184 Applied second patch from Janne, implementing suggestion to default
2185 multiple-strings-in-record handling to match SPF spec.
2187 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
2189 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
2190 read-only, out of scope).
2191 Patch from Wolfgang Breyha, report from Stuart Northfield.
2193 PP/29 Fix three issues highlighted by clang analyser static analysis.
2194 Only crash-plausible issue would require the Cambridge-specific
2195 iplookup router and a misconfiguration.
2196 Report from Marcin Mirosław.
2198 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
2200 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
2201 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
2202 As part of this, removing so much warning spew let me fix some minor
2203 real issues in debug logging.
2205 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
2206 assignment on my part. Fixed.
2208 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
2209 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
2210 Janne Snabb (who went above and beyond: thank you).
2212 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
2213 string otherwise requires a connection and a bunch more work and it's
2214 relatively easy to get wrong. Should also expose TLS library linkage
2217 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
2218 64-bit ${eval} (JH/03).
2220 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
2221 GNU libc to support some of the 64-bit stuff, should not lead to
2222 conflicts. Defined before os.h is pulled in, so if a given platform
2223 needs to override this, it can.
2225 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
2226 protection layer was required, which is not implemented.
2227 Bugzilla 1254, patch from Wolfgang Breyha.
2229 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
2230 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
2231 tls_dhparam take prime identifiers. Also unbreak combination of
2232 OpenSSL+DH_params+TLSSNI.
2234 PP/39 Disable SSLv2 by default in OpenSSL support.
2240 PP/01 Solaris build fix for Oracle's LDAP libraries.
2241 Bugzilla 1109, patch from Stephen Usher.
2243 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
2245 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
2248 TF/02 Fix a couple more cases where we did not log the error message
2249 when unlink() failed. See also change 4.74-TF/03.
2251 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
2252 lock up or crash if it happened to be inside a call to libc when it
2253 got a SIGUSR1 from exiwhat.
2255 The SIGUSR1 handler appends the current process status to the process
2256 log which is later printed by exiwhat. It used to use the general
2257 purpose logging code to do this, but several functions it calls are
2258 not safe for signals.
2260 The new output code in the SIGUSR1 handler is specific to the process
2261 log, and simple enough that it's easy to inspect for signal safety.
2262 Removing some special cases also simplifies the general logging code.
2263 Removing the spurious timestamps from the process log simplifies
2266 TF/04 Improved ratelimit ACL condition.
2268 The /noupdate option has been deprecated in favour of /readonly which
2269 has clearer semantics. The /leaky, /strict, and /readonly update modes
2270 are mutually exclusive. The update mode is no longer included in the
2271 database key; it just determines when the database is updated. (This
2272 means that when you upgrade Exim will forget old rate measurements.)
2274 Exim now checks that the per_* options are used with an update mode that
2275 makes sense for the current ACL. For example, when Exim is processing a
2276 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
2277 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
2278 must specify per_mail/readonly. If you omit the update mode it defaults to
2279 /leaky where that makes sense (as before) or /readonly where required.
2281 The /noupdate option is now undocumented but still supported for
2282 backwards compatibility. It is equivalent to /readonly except that in
2283 ACLs where /readonly is required you may specify /leaky/noupdate or
2284 /strict/noupdate which are treated the same as /readonly.
2286 A useful new feature is the /count= option. This is a generalization
2287 of the per_byte option, so that you can measure the throughput of other
2288 aggregate values. For example, the per_byte option is now equivalent
2289 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
2291 The per_rcpt option has been generalized using the /count= mechanism
2292 (though it's more complicated than the per_byte equivalence). When it is
2293 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
2294 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
2295 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
2296 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
2297 Note that using per_rcpt with a non-readonly update mode in more than
2298 one ACL will cause the recipients to be double-counted. (The per_mail
2299 and per_byte options don't have this problem.)
2301 The handling of very low rates has changed slightly. If the computed rate
2302 is less than the event's count (usually one) then this event is the first
2303 after a long gap. In this case the rate is set to the same as this event's
2304 count, so that the first message of a spam run is counted properly.
2306 The major new feature is a mechanism for counting the rate of unique
2307 events. The new per_addr option counts the number of different
2308 recipients that someone has sent messages to in the last time period. It
2309 behaves like per_rcpt if all the recipient addresses are different, but
2310 duplicate recipient addresses do not increase the measured rate. Like
2311 the /count= option this is a general mechanism, so the per_addr option
2312 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
2313 example, measure the rate that a client uses different sender addresses
2314 with the options per_mail/unique=$sender_address. There are further
2315 details in the main documentation.
2317 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
2319 TF/06 Removed a few PCRE remnants.
2321 TF/07 Automatically extract Exim's version number from tags in the git
2322 repository when doing development or release builds.
2324 PP/02 Raise smtp_cmd_buffer_size to 16kB.
2325 Bugzilla 879. Patch from Paul Fisher.
2327 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
2328 Heavily based on revision 40f9a89a from Simon Arlott's tree.
2331 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
2333 PP/05 Variable $av_failed, true if the AV scanner deferred.
2334 Bugzilla 1078. Patch from John Horne.
2336 PP/06 Stop make process more reliably on build failure.
2337 Bugzilla 1087. Patch from Heiko Schlittermann.
2339 PP/07 Make maildir_use_size_file an _expandable_ boolean.
2340 Bugzilla 1089. Patch from Heiko Schlittermann.
2342 PP/08 Handle ${run} returning more data than OS pipe buffer size.
2343 Bugzilla 1131. Patch from Holger Weiß.
2345 PP/09 Handle IPv6 addresses with SPF.
2346 Bugzilla 860. Patch from Wolfgang Breyha.
2348 PP/10 GnuTLS: support TLS 1.2 & 1.1.
2350 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
2353 PP/11 match_* no longer expand right-hand-side by default.
2354 New compile-time build option, EXPAND_LISTMATCH_RHS.
2355 New expansion conditions, "inlist", "inlisti".
2357 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
2359 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
2361 PP/14 fix log_write() format string regression from TF/03.
2362 Bugzilla 1152. Patch from Dmitry Isaikin.
2368 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
2370 PP/02 Harmonised TLS library version reporting; only show if debugging.
2371 Layout now matches that introduced for other libraries in 4.74 PP/03.
2373 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
2375 PP/04 New "dns_use_edns0" global option.
2377 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
2380 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
2381 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
2383 TK/01 Updated PolarSSL code to 0.14.2.
2384 Bugzilla 1097. Patch from Andreas Metzler.
2386 PP/07 Catch divide-by-zero in ${eval:...}.
2387 Fixes bugzilla 1102.
2389 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
2392 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
2393 format-string attack -- SECURITY: remote arbitrary code execution.
2395 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
2396 time unintentionally subject to list matching rules, letting the header
2397 cause arbitrary Exim lookups (of items which can occur in lists, *not*
2398 arbitrary string expansion). This allowed for information disclosure.
2400 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
2401 INT_MIN/-1 -- value coerced to INT_MAX.
2407 NM/01 Workaround for PCRE version dependency in version reporting
2410 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
2411 This fixes portability to compilers other than gcc, notably
2412 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
2414 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
2415 makefiles for portability to HP-UX and POSIX correctness.
2417 PP/01 Permit LOOKUP_foo enabling on the make command-line.
2418 Also via indented variable definition in the Makefile.
2419 (Debugging by Oliver Heesakkers).
2421 PP/02 Restore caching of spamd results with expanded spamd_address.
2422 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
2424 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
2425 Improves build reliability. Fix from: Frank Elsner
2427 NM/02 Fix wide character breakage in the rfc2047 coding
2428 Fixes bug 1064. Patch from Andrey N. Oktyabrski
2430 NM/03 Allow underscore in dnslist lookups
2431 Fixes bug 1026. Patch from Graeme Fowler
2433 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
2434 Code patches from Adam Ciarcinski of NetBSD.
2436 NM/04 Fixed exiqgrep to cope with mailq missing size issue
2439 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
2440 is logged, to avoid truncation. Patch from John Horne.
2442 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
2443 Patch from Jakob Hirsch.
2445 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
2446 SQL string expansion failure details.
2447 Patch from Andrey Oktyabrski.
2449 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
2450 Patch from Simon Arlott.
2452 PP/09 New lookups functionality failed to compile on old gcc which rejects
2453 extern declarations in function scope.
2454 Patch from Oliver Fleischmann
2456 PP/10 Use sig_atomic_t for flags set from signal handlers.
2457 Check getgroups() return and improve debugging.
2458 Fixed developed for diagnosis in bug 927 (which turned out to be
2461 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
2462 Patch from Mark Zealey.
2464 PP/12 Bugzilla 1056: Improved spamd server selection.
2465 Patch from Mark Zealey.
2467 PP/13 Bugzilla 1086: Deal with maildir quota file races.
2468 Based on patch from Heiko Schlittermann.
2470 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
2471 Patch from Uwe Doering, sign-off by Michael Haardt.
2473 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
2474 variable declaration deep within a block. Bug and patch from
2477 PP/15 lookups-Makefile IRIX compatibility coercion.
2479 PP/16 Make DISABLE_DKIM build knob functional.
2481 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
2482 Patch by Simon Arlott
2484 TF/03 Fix valgrind.h portability to C89 compilers that do not support
2485 variable argument macros. Our copy now differs from upstream.
2491 TF/01 Failure to get a lock on a hints database can have serious
2492 consequences so log it to the panic log.
2494 TF/02 Log LMTP confirmation messages in the same way as SMTP,
2495 controlled using the smtp_confirmation log selector.
2497 TF/03 Include the error message when we fail to unlink a spool file.
2499 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
2500 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
2501 for maintaining out-of-tree patches for some time.
2503 PP/01 Bugzilla 139: Documentation and portability issues.
2504 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
2505 Handle per-OS dynamic-module compilation flags.
2507 PP/02 Let /dev/null have normal permissions.
2508 The 4.73 fixes were a little too stringent and complained about the
2509 permissions on /dev/null. Exempt it from some checks.
2510 Reported by Andreas M. Kirchwitz.
2512 PP/03 Report version information for many libraries, including
2513 Exim version information for dynamically loaded libraries. Created
2514 version.h, now support a version extension string for distributors
2515 who patch heavily. Dynamic module ABI change.
2517 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
2518 privilege escalation vulnerability whereby the Exim run-time user
2519 can cause root to append content of the attacker's choosing to
2522 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
2525 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
2526 If dropping privileges for untrusted macros, we disabled normal logging
2527 on the basis that it would fail; for the Exim run-time user, this is not
2528 the case, and it resulted in successful deliveries going unlogged.
2529 Fixed. Reported by Andreas Metzler.
2535 PP/01 Date: & Message-Id: revert to normally being appended to a message,
2536 only prepend for the Resent-* case. Fixes regression introduced in
2537 Exim 4.70 by NM/22 for Bugzilla 607.
2539 PP/02 Include check_rfc2047_length in configure.default because we're seeing
2540 increasing numbers of administrators be bitten by this.
2542 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
2544 PP/03 Bugzilla 994: added openssl_options main configuration option.
2546 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
2548 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
2550 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
2552 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
2553 without a peer certificate, leading to a segfault because of an
2554 assumption that peers always have certificates. Be a little more
2555 paranoid. Problem reported by Martin Tscholak.
2557 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
2558 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
2559 NB: ClamAV planning to remove STREAM in "middle of 2010".
2560 CL also introduces -bmalware, various -d+acl logging additions and
2561 more caution in buffer sizes.
2563 PP/09 Implemented reverse_ip expansion operator.
2565 PP/10 Bugzilla 937: provide a "debug" ACL control.
2567 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2569 PP/12 Bugzilla 973: Implement --version.
2571 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2573 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2575 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2577 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2578 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2579 ignore trailing whitespace.
2581 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2583 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2586 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2587 Notification from Dr Andrew Aitchison.
2589 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2590 ExtendedDetectionInfo response format.
2591 Notification from John Horne.
2593 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2596 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2597 XSL and documented dependency on system catalogs, with examples of how
2600 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2603 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2604 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2605 configuration file which is writeable by the Exim user or group.
2607 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2608 of configuration files to cover files specified with the -C option if
2609 they are going to be used with root privileges, not just the default
2612 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2613 option (effectively making it always true).
2615 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2616 files to be used while preserving root privileges.
2618 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2619 that rogue child processes cannot use them.
2621 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2622 run-time user, instead of root.
2624 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2625 Exim run-time user without dropping privileges.
2627 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2628 result string, instead of calling string_vformat() twice with the same
2631 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2632 for other users. Others should always drop root privileges if they use
2633 -C on the command line, even for a whitelisted configure file.
2635 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2637 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2643 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2644 $data_path, and $header_path variables; fixed documentation bugs and
2647 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2648 exipick to access non-standard spools, including the "frozen" queue
2651 NM/01 Bugzilla 965: Support mysql stored procedures.
2652 Patch from Alain Williams
2654 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2656 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2657 Patch from Andreas Metzler
2659 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2660 Patch from Kirill Miazine
2662 NM/05 Bugzilla 671: Added umask to procmail example.
2664 JJ/03 installed exipick 20100323.0, fixing doc bug
2666 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2667 directory. Notification and patch from Dan Rosenberg.
2669 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2671 TK/02 Improve log output when DKIM signing operation fails.
2673 MH/01 Treat the transport option dkim_domain as a colon separated
2674 list, not as a single string, and sign the message with each element,
2675 omitting multiple occurences of the same signer.
2677 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2678 Bugzilla 985, 986. Patch by Simon Arlott
2680 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2681 Patch by Simon Arlott
2683 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2684 MBX locking. Notification from Dan Rosenberg.
2690 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2692 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2694 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2696 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2698 NM/04 Bugzilla 915: Flag broken perl installation during build.
2704 TK/01 Added patch by Johannes Berg that expands the main option
2705 "spamd_address" if it starts with a dollar sign.
2707 TK/02 Write list of recipients to X-Envelope-Sender header when building
2708 the mbox-format spool file for content scanning (suggested by Jakob
2711 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2712 (http://www.dcc-servers.net/) support via dccifd. Activated by
2713 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2715 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2716 by Mark Daniel Reidel <mr@df.eu>.
2718 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2719 When building exim an external PCRE library is now needed -
2720 PCRE is a system library on the majority of modern systems.
2721 See entry on PCRE_LIBS in EDITME file.
2723 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2724 conversation. Added nologin parameter to request.
2725 Patch contributed by Kirill Miazine.
2727 TF/01 Do not log submission mode rewrites if they do not change the address.
2729 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2731 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2732 log files in place. Contributed by Roberto Lima.
2734 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2736 TF/03 Bugzilla 615: When checking the local_parts router precondition
2737 after a local_part_suffix or local_part_prefix option, Exim now
2738 does not use the address's named list lookup cache, since this
2739 contains cached lookups for the whole local part.
2741 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2742 Robert Millan. Documentation is in experimental-spec.txt.
2744 TF/04 Bugzilla 668: Fix parallel build (make -j).
2746 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2748 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2749 Patch provided by Jan Srzednicki.
2751 TF/05 Leading white space used to be stripped from $spam_report which
2752 wrecked the formatting. Now it is preserved.
2754 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2755 that they are available at delivery time.
2757 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2759 TF/08 TLS error reporting now respects the incoming_interface and
2760 incoming_port log selectors.
2762 TF/09 Produce a more useful error message if an SMTP transport's hosts
2763 setting expands to an empty string.
2765 NM/06 Bugzilla 744: EXPN did not work under TLS.
2766 Patch provided by Phil Pennock.
2768 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2769 Patch provided by Richard Godbee.
2771 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2772 acl_smtp_notquit, added index entry.
2774 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2775 Patch provided by Eugene Bujak.
2777 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2778 accept(). Patch provided by Maxim Dounin.
2780 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2781 Patch provided by Phil Pennock.
2783 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2785 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2786 Patch provided by Brad "anomie" Jorsch.
2788 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2789 Patch provided by Dean Brooks.
2791 TK/05 Add native DKIM support (does not depend on external libraries).
2793 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2794 Patch provided by Graeme Fowler.
2796 NM/16 Bugzilla 851: Documentation example syntax fix.
2798 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2800 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2803 NM/19 Bugzilla 745: TLS version reporting.
2804 Patch provided by Phil Pennock.
2806 NM/20 Bugzilla 167: bool: condition support.
2807 Patch provided by Phil Pennock.
2809 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2810 clients. Patch provided by Phil Pennock.
2812 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2813 Patch provided by Brad "anomie" Jorsch.
2815 NM/23 Bugzilla 687: Fix misparses in eximstats.
2816 Patch provided by Heiko Schlittermann.
2818 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2819 Patch provided by Heiko Schlittermann.
2821 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2822 plus update to original patch.
2824 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2826 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2827 Patch provided by David Brownlee.
2829 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2831 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2833 NM/30 Bugzilla 888: TLS documentation bugfixes.
2835 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2837 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2838 Unlike the original bugzilla I have changed all shell scripts in src tree.
2840 NM/33 Bugzilla 898: Transport filter timeout fix.
2841 Patch by Todd Rinaldo.
2843 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2844 Patch by Serge Demonchaux.
2846 NM/35 Bugzilla 39: Base64 decode bug fixes.
2847 Patch by Jakob Hirsch.
2849 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2851 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2853 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2855 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2861 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2862 ALT-N's libdkim that I have put here:
2863 http://duncanthrax.net/exim-experimental/
2865 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2866 were called 'true' and it seems that C99 defines that as a reserved
2867 keyword to be used with 'bool' variable types. That means you could
2868 not include C99-style headers which use bools without triggering
2869 build errors in sieve.c.
2871 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2872 as mailq or other aliases. Changed the --help handling significantly
2873 to do whats expected. exim_usage() emits usage/help information.
2875 SC/01 Added the -bylocaldomain option to eximstats.
2877 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2879 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2881 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2887 PH/01 Another patch from the Sieve maintainer.
2889 PH/02 When an IPv6 address is converted to a string for single-key lookup
2890 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2891 dots are used instead of colons so that keys in lsearch files need not
2892 contain colons. This was done some time before quoting was made available
2893 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2894 (notated using the quote facility) so as to distinguish them from IPv4
2895 keys. This meant that lookups for IP addresses in host lists did not work
2896 for iplsearch lookups.
2898 This has been fixed by arranging for IPv6 addresses to be expressed with
2899 colons if the lookup type is iplsearch. This is not incompatible, because
2900 previously such lookups could never work.
2902 The situation is now rather anomalous, since one *can* have colons in
2903 ordinary lsearch keys. However, making the change in all cases is
2904 incompatible and would probably break a number of configurations.
2906 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2909 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2910 conversion specification without a maximum field width, thereby enabling
2911 a rogue spamd server to cause a buffer overflow. While nobody in their
2912 right mind would setup Exim to query an untrusted spamd server, an
2913 attacker that gains access to a server running spamd could potentially
2914 exploit this vulnerability to run arbitrary code as the Exim user.
2916 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2917 $primary_hostname instead of what libspf2 thinks the hosts name is.
2919 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2920 a directory entry by the name of the lookup key. Previously, if a
2921 symlink pointed to a non-existing file or a file in a directory that
2922 Exim lacked permissions to read, a lookup for a key matching that
2923 symlink would fail. Now it is enough that a matching directory entry
2924 exists, symlink or not. (Bugzilla 503.)
2926 PH/03 The body_linecount and body_zerocount variables are now exported in the
2929 PH/04 Added the $dnslist_matched variable.
2931 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2932 This means they are set thereafter only if the connection becomes
2935 PH/06 Added the client_condition to authenticators so that some can be skipped
2936 by clients under certain conditions.
2938 PH/07 The error message for a badly-placed control=no_multiline_responses left
2939 "_responses" off the end of the name.
2941 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2943 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2944 (without spaces) instead of just copying the configuration text.
2946 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2948 PH/11 Added $max_received_linelength.
2950 PH/12 Added +ignore_defer and +include_defer to host lists.
2952 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2953 way in which PCRE > 7.0 is built.
2955 PH/14 Implemented queue_only_load_latch.
2957 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2958 MAIL command. The effect was to mangle the value on 64-bit systems.
2960 PH/16 Another patch from the Sieve maintainer.
2962 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2964 PH/18 If a system quota error occurred while trying to create the file for
2965 a maildir delivery, the message "Mailbox is full" was not appended to the
2966 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2967 only to a quota excession during the actual writing of the file.
2969 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2970 characters?) which causes problems in log lines. The DN values are now
2971 passed through string_printing() before being added to log lines.
2973 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2974 and InterBase are left for another time.)
2976 PH/21 Added message_body_newlines option.
2978 PH/22 Guard against possible overflow in moan_check_errorcopy().
2980 PH/23 POSIX allows open() to be a macro; guard against that.
2982 PH/24 If the recipient of an error message contained an @ in the local part
2983 (suitably quoted, of course), incorrect values were put in $domain and
2984 $local_part during the evaluation of errors_copy.
2990 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2991 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2994 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2995 issue a MAIL command.
2997 PH/02 In an ACL statement such as
2999 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
3001 if a client was not listed at all, or was listed with a value other than
3002 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
3003 the condition was not true (as it should be), so access was not denied.
3004 The bug was that the ! inversion was incorrectly passed on to the second
3005 item. This has been fixed.
3007 PH/03 Added additional dnslists conditions == and =& which are different from
3008 = and & when the dns lookup returns more than one IP address.
3010 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
3011 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
3013 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
3014 FSYNC, which compiles an option called disable_fsync that allows for
3015 bypassing fsync(). The documentation is heavily laced with warnings.
3017 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
3019 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
3020 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
3021 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
3022 including adding "make clean"; (3) Added -fPIC when compiling the test
3023 dynamically loaded module, to get rid of a warning.
3025 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
3026 message fails, move_frozen_messages = true and ignore_bounce_errors_after
3027 = 0s. The bug is otherwise harmless.
3029 PH/07 There was a bug in the dovecot authenticator such that the value of
3030 $auth1 could be overwritten, and so not correctly preserved, after a
3031 successful authentication. This usually meant that the value preserved by
3032 the server_setid option was incorrect.
3034 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
3036 PH/09 Installed PCRE release 7.0.
3038 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
3039 run for batched SMTP input. It is now run at the start of every message
3040 in the batch. While fixing this I discovered that the process information
3041 (output by running exiwhat) was not always getting set for -bs and -bS
3042 input. This is fixed, and it now also says "batched" for BSMTP.
3044 PH/11 Added control=no_pipelining.
3046 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
3047 patch, slightly modified), and move the expansion of helo_data till after
3048 the connection is made in the smtp transport (so it can use these
3051 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
3053 PH/14 Added log_selector = +pid.
3055 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
3057 PH/16 Add ${if forany and ${if forall.
3059 PH/17 Added dsn_from option to vary the From: line in DSNs.
3061 PH/18 Flush SMTP output before performing a callout, unless control =
3062 no_callout_flush is set.
3064 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
3065 was true (the default) a successful delivery failed to delete the retry
3066 item, thus causing premature timeout of the address. The bug is now
3069 PH/20 Added hosts_avoid_pipelining to the smtp transport.
3071 PH/21 Long custom messages for fakedefer and fakereject are now split up
3072 into multiline responses in the same way that messages for "deny" and
3073 other ACL rejections are.
3075 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
3076 with slight modification.
3078 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
3079 draft, changing the syntax and factoring some duplicate code".
3081 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
3082 for deliveries of the second and subsequent messages over the same SMTP
3085 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
3086 ${reduce, with only minor "tidies".
3088 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
3090 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
3091 expansion side effects.
3093 PH/27 When a message times out after an over-quota error from an Exim-imposed
3094 quota, the bounce message says "mailbox is full". This message was not
3095 being given when it was a system quota that was exceeded. It now should
3098 MH/03 Made $recipients available in local_scan(). local_scan() already has
3099 better access to the recipient list through recipients_list[], but
3100 $recipients can be useful in postmaster-provided expansion strings.
3102 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
3103 in the case of a MAIL command with additional options following the
3104 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
3105 were accidentally chopped off.
3107 PH/29 SMTP synchronization checks are implemented when a command is read -
3108 there is a check that no more input is waiting when there shouldn't be
3109 any. However, for some commands, a delay in an ACL can mean that it is
3110 some time before the response is written. In this time, more input might
3111 arrive, invalidly. So now there are extra checks after an ACL has run for
3112 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
3113 pipelining has not been advertised.
3115 PH/30 MH's patch to allow iscntrl() characters to be list separators.
3117 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
3118 returned in the SMTP response when smtp_return_error_details was false.
3119 This has been fixed.
3121 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
3122 instead of the C I/O that was originally supplied, because problems were
3123 reported on Solaris.
3125 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
3126 Exim which did not show up earlier: it was assuming that a call to
3127 SSL_CTX_set_info_callback() might give an error value. In fact, there is
3128 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
3129 was a macro that became an assignment, so it seemed to work. This has
3130 changed to a proper function call with a void return, hence the compile
3131 error. Exim's code has been fixed.
3133 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
3136 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
3138 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
3140 PH/37 If a message is not accepted after it has had an id assigned (e.g.
3141 because it turns out to be too big or there is a timeout) there is no
3142 "Completed" line in the log. When some messages of this type were
3143 selected by exigrep, they were listed as "not completed". Others were
3144 picked up by some special patterns. I have improved the selection
3145 criteria to be more general.
3147 PH/38 The host_find_failed option in the manualroute router can now be set
3148 to "ignore", to completely ignore a host whose IP address cannot be
3149 found. If all hosts are ignored, the behaviour is controlled by the new
3150 host_all_ignored option.
3152 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
3153 homing or because of multiple MX records with /mx) generated more than
3154 one IP address, and the following item turned out to be the local host,
3155 all the secondary addresses of the first item were incorrectly removed
3156 from the list, along with the local host and any following hosts (which
3157 is what is supposed to happen).
3159 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
3160 whoever called Exim into the -H file. In the case of the daemon it was
3161 behaving confusingly. When first started, it used values for whoever
3162 started the daemon, but after a SIGHUP it used the Exim user (because it
3163 calls itself on a restart). I have changed the code so that it now always
3166 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
3167 message are rejected with the same error (e.g. no authentication or bad
3168 sender address), and a DATA command is nevertheless sent (as can happen
3169 with PIPELINING or a stupid MUA), the error message that was given to the
3170 RCPT commands is included in the rejection of the DATA command. This is
3171 intended to be helpful for MUAs that show only the final error to their
3174 PH/42 Another patch from the Sieve maintainer.
3176 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
3177 Eximstats - Fixed some broken HTML links and added missing column headers
3179 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
3180 columns for Rejects, Temp Rejects, Ham, and Spam rows.
3182 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
3184 PH/43 Yet another patch from the Sieve maintainer.
3186 PH/44 I found a way to check for a TCP/IP connection going away before sending
3187 the response to the final '.' that terminates a message, but only in the
3188 case where the client has not sent further data following the '.'
3189 (unfortunately, this is allowed). However, in many cases there won't be
3190 any further data because there won't be any more messages to send. A call
3191 to select() can be used: if it shows that the input is "ready", there is
3192 either input waiting, or the socket has been closed. An attempt to read
3193 the next input character can distinguish the two cases. Previously, Exim
3194 would have sent an OK response which the client would never have see.
3195 This could lead to message repetition. This fix should cure that, at
3196 least in a lot of common cases.
3198 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
3199 advertised in response to EHLO.
3205 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
3206 fixed by 4.65/MH/01 (is this a record?) are fixed:
3208 (i) An empty string was always treated as zero by the numeric comparison
3209 operators. This behaviour has been restored.
3211 (ii) It is documented that the numeric comparison operators always treat
3212 their arguments as decimal numbers. This was broken in that numbers
3213 starting with 0 were being interpreted as octal.
3215 While fixing these problems I realized that there was another issue that
3216 hadn't been noticed. Values of message_size_limit (both the global option
3217 and the transport option) were treated as octal if they started with 0.
3218 The documentation was vague. These values are now always treated as
3219 decimal, and I will make that clear in the documentation.
3225 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
3226 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
3229 MH/01 Don't check that the operands of numeric comparison operators are
3230 integers when their expansion is in "skipping" mode (fixes bug
3231 introduced by 4.64-PH/07).
3233 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
3234 child addresses, Exim now panics and dies. Previously, because the count
3235 is held in a short int, deliveries were likely to be lost. As such a
3236 large number of recipients for a single message is ridiculous
3237 (performance will be very, very poor), I have chosen to impose a limit
3238 rather than extend the field.
3244 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
3245 leftover -K file (the existence of which was triggered by #402).
3246 While we were at it, introduced process PID as part of the -K
3247 filename. This should rule out race conditions when creating
3250 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
3251 processing considerably. Previous code took too long for large mails,
3252 triggering a timeout which in turn triggers #401.
3254 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
3255 in the DK code in transports.c. sendfile() is not really portable,
3256 hence the _LINUX specificness.
3258 TF/01 In the add_headers option to the mail command in an Exim filter,
3259 there was a bug that Exim would claim a syntax error in any
3260 header after the first one which had an odd number of characters
3263 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
3264 callout verification, Exim cached a "reject" for the entire domain. This
3265 is correct for most verifications, but it is not correct for a recipient
3266 verification with use_sender or use_postmaster set, because in that case
3267 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
3268 case of MAIL FROM:<> rejection from other early rejections (e.g.
3269 rejection of HELO). When verifying a recipient using a non-null MAIL
3270 address, the cache is ignored if it shows MAIL FROM:<> rejection.
3271 Whatever the result of the callout, the value of the domain cache is
3272 left unchanged (for any other kind of callout, getting as far as trying
3273 RCPT means that the domain itself is ok).
3275 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
3278 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
3279 manifest itself as EPIPE rather than ECONNECT. When tidying away a
3280 session, the daemon ignores ECONNECT errors and logs others; it now
3281 ignores EPIPE as well.
3283 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
3284 (quoted-printable decoding).
3286 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
3287 later the small subsequent patch to fix an introduced bug.
3289 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
3291 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
3293 PH/08 An error is now given if message_size_limit is specified negative.
3295 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
3296 to be given (somewhat) arbitrary names.
3298 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
3301 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
3302 miscellaneous code fixes
3304 PH/10 Added the log_reject_target ACL modifier to specify where to log
3307 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
3308 hostname. This is wrong, because it relates to the incoming message (and
3309 probably the interface on which it is arriving) and not to the outgoing
3310 callout (which could be using a different interface). This has been
3311 changed to use the value of the helo_data option from the smtp transport
3312 instead - this is what is used when a message is actually being sent. If
3313 there is no remote transport (possible with a router that sets up host
3314 addresses), $smtp_active_hostname is used.
3316 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
3317 tweaks were necessary in order to get it to work (see also 21 below):
3318 (a) The code assumed that strncpy() returns a negative number on buffer
3319 overflow, which isn't the case. Replaced with Exim's string_format()
3321 (b) There were several signed/unsigned issues. I just did the minimum
3322 hacking in of casts. There is scope for a larger refactoring.
3323 (c) The code used strcasecmp() which is not a standard C function.
3324 Replaced with Exim's strcmpic() function.
3325 (d) The code set only $1; it now sets $auth1 as well.
3326 (e) A simple test gave the error "authentication client didn't specify
3327 service in request". It would seem that Dovecot has changed its
3328 interface. Fortunately there's a specification; I followed it and
3329 changed what the client sends and it appears to be working now.
3331 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
3334 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
3335 address A is aliased to B and C, where B exists and C does not. Without
3336 -v the output is "A verified" because verification stops after a
3337 successful redirection if more than one address is generated. However,
3338 with -v the child addresses are also verified. Exim was outputting "A
3339 failed to verify" and then showing the successful verification for C,
3340 with its parentage. It now outputs "B failed to verify", showing B's
3341 parentage before showing the successful verification of C.
3343 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
3344 look up a TXT record in a specific list after matching in a combined
3347 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
3348 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
3349 they consult the DNS. I had assumed they would set it the way they
3350 wanted; and indeed my experiments on Linux seem to show that in some
3351 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
3352 To be on the safe side, however, I have now made the interface to
3353 host_find_byname() similar to host_find_bydns(), with an argument
3354 containing the DNS resolver options. The host_find_byname() function now
3355 sets these options at its start, just as host_find_bydns() does. The smtp
3356 transport options dns_qualify_single and dns_search_parents are passed to
3357 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
3358 of host_find_byname() use the default settings of RES_DEFNAMES
3359 (qualify_single) but not RES_DNSRCH (search_parents).
3361 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
3362 spool_read_header() do less string testing, by means of a preliminary
3363 switch on the second character of optional "-foo" lines. (This is
3364 overdue, caused by the large number of possibilities that now exist.
3365 Originally there were few.) While I was there, I also converted the
3366 str(n)cmp tests so they don't re-test the leading "-" and the first
3367 character, in the hope this might squeeze out yet more improvement.
3369 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
3370 flag allowing group syntax was set by the header_syntax check but not
3371 turned off, possible causing trouble later; (2) The flag was not being
3372 set at all for the header_verify test, causing "group"-style headers to
3373 be rejected. I have now set it in this case, and also caused header_
3374 verify to ignore an empty address taken from a group. While doing this, I
3375 came across some other cases where the code for allowing group syntax
3376 while scanning a header line wasn't quite right (mostly, not resetting
3377 the flag correctly in the right place). These bugs could have caused
3378 trouble for malformed header lines. I hope it is now all correct.
3380 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
3381 with the "reply" argument non-NULL. The code, however (which originally
3382 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
3383 but it didn't always do it. This confused somebody who was copying the
3384 code for some other use. I have removed all the tests.
3386 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
3387 feature that was used to support insecure browsers during the U.S. crypto
3388 embargo. It requires special client support, and Exim is probably the
3389 only MTA that supported it -- and would never use it because real RSA is
3390 always available. This code has been removed, because it had the bad
3391 effect of slowing Exim down by computing (never used) parameters for the
3392 RSA_EXPORT functionality.
3394 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
3395 authenticator to fail if there's a tab character in the incoming data
3396 (there should never be unless someone is messing about, as it's supposed
3397 to be base64-encoded). Also added, on Timo's advice, the "secured" option
3398 if the connection is using TLS or if the remote IP is the same as the
3399 local IP, and the "valid-client-cert option" if a client certificate has
3402 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
3403 authenticators. This can be used for authorization after authentication
3404 succeeds. (In the case of plaintext, it servers for both authentication
3407 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
3408 if any retry times were supplied.
3410 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
3411 connection, where there is no client IP address to check. In this
3412 situation, the verify now always succeeds.
3414 PH/25 Applied John Jetmore's -Mset patch.
3416 PH/26 Added -bem to be like -Mset, but loading a message from a file.
3418 PH/27 In a string expansion for a processed (not raw) header when multiple
3419 headers of the same name were present, leading whitespace was being
3420 removed from all of them, but trailing whitespace was being removed only
3421 from the last one. Now trailing whitespace is removed from each header
3422 before concatenation. Completely empty headers in a concatenation (as
3423 before) are ignored.
3425 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
3426 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
3428 PH/29 [Removed. This was a change that I later backed out, and forgot to
3429 correct the ChangeLog entry (that I had efficiently created) before
3430 committing the later change.]
3432 PH/30 Exim was sometimes attempting to deliver messages that had suffered
3433 address errors (4xx response to RCPT) over the same connection as other
3434 messages routed to the same hosts. Such deliveries are always "forced",
3435 so retry times are not inspected. This resulted in far too many retries
3436 for the affected addresses. The effect occurred only when there were more
3437 hosts than the hosts_max_try setting in the smtp transport when it had
3438 the 4xx errors. Those hosts that it had tried were not added to the list
3439 of hosts for which the message was waiting, so if all were tried, there
3440 was no problem. Two fixes have been applied:
3442 (i) If there are any address or message errors in an SMTP delivery, none
3443 of the hosts (tried or untried) are now added to the list of hosts
3444 for which the message is waiting, so the message should not be a
3445 candidate for sending over the same connection that was used for a
3446 successful delivery of some other message. This seems entirely
3447 reasonable: after all the message is NOT "waiting for some host".
3448 This is so "obvious" that I'm not sure why it wasn't done
3449 previously. Hope I haven't missed anything, but it can't do any
3450 harm, as the worst effect is to miss an optimization.
3452 (ii) If, despite (i), such a delivery is accidentally attempted, the
3453 routing retry time is respected, so at least it doesn't keep
3454 hammering the server.
3456 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
3457 in ${readsocket because some servers need this prod.
3459 PH/32 Added some extra debug output when updating a wait-xxx database.
3461 PH/33 The hint "could be header name not terminated by colon", which has been
3462 given for certain expansion errors for a long time, was not being given
3463 for the ${if def:h_colon_omitted{... case.
3465 PH/34 The spec says: "With one important exception, whenever a domain list is
3466 being scanned, $domain contains the subject domain." There was at least
3467 one case where this was not true.
3469 PH/35 The error "getsockname() failed: connection reset by peer" was being
3470 written to the panic log as well as the main log, but it isn't really
3471 panic-worthy as it just means the connection died rather early on. I have
3472 removed the panic log writing for the ECONNRESET error when getsockname()
3475 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
3476 runs only) independently of the message's sender address. This meant
3477 that, if the 4xx error was in fact related to the sender, a different
3478 message to the same recipient with a different sender could confuse
3479 things. In particular, this can happen when sending to a greylisting
3480 server, but other circumstances could also provoke similar problems.
3481 I have changed the default so that the retry time for these errors is now
3482 based a combination of the sender and recipient addresses. This change
3483 can be overridden by setting address_retry_include_sender=false in the
3486 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
3487 remote server are returned as part of bounce messages. This was not
3488 happening for LMTP over a pipe (the lmtp transport), but now it is the
3489 same for both kinds of LMTP.
3491 PH/38 Despite being documented as not happening, Exim was rewriting addresses
3492 in header lines that were in fact CNAMEs. This is no longer the case.
3494 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
3495 and queue runs started by the daemon processed all messages. This has
3496 been fixed so that -R and -S can now usefully be given with -q<time>.
3498 PH/40 Import PCRE release 6.7 (fixes some bugs).
3500 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
3502 PH/42 Give an error if -q is specified more than once.
3504 PH/43 Renamed the variables $interface_address and $interface_port as
3505 $received_ip_address and $received_port, to make it clear that these
3506 values apply to message reception, and not to the outgoing interface when
3507 a message is delivered. (The old names remain recognized, of course.)
3509 PH/44 There was no timeout on the connect() call when using a Unix domain
3510 socket in the ${readsocket expansion. There now is.
3512 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
3513 be meaningful with "accept".
3515 SC/01 Eximstats V1.43
3516 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
3518 SC/02 Eximstats V1.44
3519 Use a glob alias rather than an array ref in the generated
3520 parser. This improves both readability and performance.
3522 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
3523 Collect SpamAssassin and rejection statistics.
3524 Don't display local sender or destination tables unless
3525 there is data to show.
3526 Added average volumes into the top table text output.
3528 SC/04 Eximstats V1.46
3529 Collect data on the number of addresses (recipients)
3530 as well as the number of messages.
3532 SC/05 Eximstats V1.47
3533 Added 'Message too big' to the list of mail rejection
3534 reasons (thanks to Marco Gaiarin).
3536 SC/06 Eximstats V1.48
3537 Mainlog lines which have GMT offsets and are too short to
3538 have a flag are now skipped.
3540 SC/07 Eximstats V1.49 (Alain Williams)
3541 Added the -emptyok flag.
3543 SC/08 Eximstats V1.50
3544 Fixes for obtaining the IP address from reject messages.
3546 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
3547 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
3548 whitespace changes from 4.64-PH/27
3550 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
3553 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
3554 are found, allow negative numbers in numeric criteria)
3556 JJ/06 exipick.20061117.2, added new $message_body_missing variable
3558 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
3559 to match changes made in 4.64-PH/43
3561 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
3563 PH/47 Put in an explicit test for a DNS lookup of an address record where the
3564 "domain" is actually an IP address, and force a failure. This locks out
3565 those revolvers/nameservers that support "A-for-A" lookups, in
3566 contravention of the specifications.
3568 PH/48 When a host name was looked up from an IP address, and the subsequent
3569 forward lookup of the name timed out, the host name was left in
3570 $sender_host_name, contrary to the specification.
3572 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3573 restricted to single-key lookups, Exim was not diagnosing an error if
3574 * or *@ was used with a query-style lookup.
3576 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3578 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3579 long ago, but noone interested enough thought of it. Let's just say that
3580 the "1.1" means that there are some new functions that weren't there at
3581 some point in the past.
3583 PH/51 Error processing for expansion failure of helo_data from an smtp
3584 transport during callout processing was broken.
3586 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3587 tested/used via the -bh/-bhc/-bs options.
3589 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3590 bug, fixed in subsequent PCRE releases).
3592 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3593 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3595 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3601 SC/01 Use a glob alias rather than an array ref in eximstats generated
3602 parser. This improves both readability and performance.
3604 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3605 Don't display local sender or destination tables in eximstats unless
3606 there is data to show.
3607 Added average volumes into the eximstats top table text output.
3609 SC/03 Collect data on the number of addresses (recipients) as well
3610 as the number of messages in eximstats.
3612 TF/01 Correct an error in the documentation for the redirect router. Exim
3613 does (usually) call initgroups() when daemonizing.
3615 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3616 with consistent privilege compared to when running as a daemon.
3618 TF/03 Note in the spec that $authenticated_id is not set for local
3619 submissions from trusted users.
3621 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3622 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3624 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3625 by adding some example configuration directives to the default
3626 configuration file. A little bit of work is required to uncomment the
3627 directives and define how usernames and passwords are checked, but
3628 there is now a framework to start from.
3630 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3631 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3632 without this. I don't know how relevant this is to other LDAP libraries.
3634 PH/02 Add the verb name to the "unknown ACL verb" error.
3636 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3638 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3640 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3641 directory not expanded when it should be if an expanded home directory
3642 was set for the address (which is overridden by the transport).
3644 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3647 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3648 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3649 because it is too late at that time, and has no effect.
3651 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3652 security issue with \' (bugzilla #107). I could not use the
3653 PQescapeStringConn() function, because it needs a PGconn value as one of
3656 PH/08 When testing addresses using -bt, indicate those final addresses that
3657 are duplicates that would not cause an additional delivery. At least one
3658 person was confused, thinking that -bt output corresponded to deliveries.
3659 (Suppressing duplicates isn't a good idea as you lose the information
3660 about possibly different redirections that led to the duplicates.)
3662 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3663 systems where poll() doesn't work, in particular OS X.
3665 PH/10 Added more information to debugging output for retry time not reached.
3667 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3668 operations in malware.c.
3670 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3673 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3674 syslog_duplication set false, log lines that would normally be written
3675 both the the main log and to the reject log were not written to syslog at
3678 PH/14 In the default configuration, change the use of "message" in ACL warn
3679 statements to "add_header".
3681 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3682 not followed by a command (e.g. "seen endif").
3684 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3685 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3688 PH/17 Added extra conditions to the default value of delay_warning_condition
3692 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3693 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3694 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3697 The Auto-Submitted: and various List- headers are standardised, whereas I
3698 don't think Precedence: ever was.
3700 PH/18 Refactored debugging code in route_finduser() to show more information,
3701 in particular, the error code if getpwnam() issues one.
3703 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3704 This is apparently needed in addition to the PH/07 change above to avoid
3705 any possible encoding problems.
3707 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3708 but not after initializing Perl.
3710 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3711 output them only if debugging. By default they are written stderr,
3712 apparently, which is not desirable.
3714 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3717 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3720 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3722 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3723 authenticated or an ident call has been made. Suppress the default
3724 values for $authenticated_id and $authenticated_sender (but permit -oMai
3725 and -oMas) when testing with -bh.
3727 PH/24 Re-jigged the order of the tests in the default configuration so that the
3728 tests for valid domains and recipients precede the DNS black list and CSA
3729 tests, on the grounds that those ones are more expensive.
3731 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3732 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3733 This bug exists in every version of Exim that I still have, right back to
3736 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3737 However, an attempt to turn on case-sensitivity in a regex key by
3738 including (?-i) didn't work because the subject string was already
3739 lowercased, and the effects were non-intuitive. It turns out that a
3740 one-line patch can be used to allow (?-i) to work as expected.
3746 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3747 other effects) broke the use of negated acl sub-conditions.
3749 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3752 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3753 "Deny" causes Exim to reject the incoming connection with a 554 error.
3754 Unfortunately, if there is a major crisis, such as a disk failure,
3755 tcp-wrappers gives "deny", whereas what one would like would be some
3756 kind of temporary error. A kludge has been added to help with this.
3757 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3758 554 error is used if errno is still zero or contains ENOENT (which occurs
3759 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3762 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3764 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3765 errors. Otherwise a message that provokes a temporary error (when other
3766 messages do not) can cause a whole host to time out.
3768 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3769 addresses were routed directly to files or pipes from a redirect router.
3770 File deliveries just didn't batch; pipe deliveries might have suffered
3773 PH/06 A failure to get a lock for a hints database would erroneously always say
3774 "Failed to get write lock", even when it was really a read lock.
3776 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3777 of 0600. This has been changed to use the value of the lockfile_mode
3778 option (which defaults to 0600).
3780 PH/08 Applied small patch from the Sieve maintainer.
3782 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3783 folder from quota calculations, a direct delivery into this folder messed
3784 up the contents of the maildirsize file. This was because the regex was
3785 used only to exclude .Trash (or whatever) when the size of the mailbox
3786 was calculated. There was no check that a delivery was happening into an
3787 excluded directory. This bug has been fixed by ignoring all quota
3788 processing for deliveries into excluded directories.
3790 PH/10 Added the maildirfolder_create_regex option to appendfile.
3796 PH/01 The code for finding all the local interface addresses on a FreeBSD
3797 system running IPv6 was broken. This may well have applied to all BSD
3798 systems, as well as to others that have similar system calls. The broken
3799 code found IPv4 interfaces correctly, but gave incorrect values for the
3800 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3801 that it would not match correctly against @[] and not recognize the IPv6
3804 PH/02 The ipliteral router was not recognizing addresses of the form user@
3805 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3807 PH/03 Added disable_ipv6.
3809 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3810 decoded form, because it is most often used to construct To: headers
3811 lines in autoreplies, and the decoded form may well be syntactically
3812 invalid. However, $reply_address has leading white space removed, and all
3813 newlines turned into spaces so that the autoreply transport does not
3816 PH/05 If group was specified without a user on a router, and no group or user
3817 was specified on a transport, the group from the router was ignored.
3819 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3820 for visible compile-time settings that can be used to change these
3821 numbers, for those that want even more. Backwards compatibility with old
3822 spool files has been maintained. However, going back to a previous Exim
3823 release will lost any variables that are in spool files.
3825 PH/07 Two small changes when running in the test harness: increase delay when
3826 passing a TCP/IP connection to a new process, in case the original
3827 process has to generate a bounce, and remove special handling of
3828 127.0.0.2 (sic), which is no longer necessary.
3830 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3831 be the same on different OS.
3833 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3836 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3837 whether --show-vars was specified or not
3839 JJ/02 exipick: Added support for new ACL variable spool format introduced
3842 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3843 syntactically invalid From: or Reply-to: line, and a filter used this to
3844 generate an autoreply, and therefore failed to obtain an address for the
3845 autoreply, Exim could try to deliver to a non-existent relative file
3846 name, causing unrelated and misleading errors. What now happens is that
3847 it logs this as a hard delivery error, but does not attempt to create a
3850 PH/11 The exinext utility has a -C option for testing purposes, but although
3851 the given file was scanned by exinext itself; it wasn't being passed on
3852 when Exim was called.
3854 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3855 an end-of-file indication when reading a command response.
3857 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3858 compiled. In many other places in Exim, IPv6 addresses are always
3859 recognized, so I have changed this. It also means that IPv4 domain
3860 literals of the form [IPV4:n.n.n.n] are now always recognized.
3862 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3863 used if the router is not running as root, for example, when verifying at
3864 ACL time, or when using -bh. The debugging output from this situation was
3865 non-existent - all you got was a failure to exec. I have made two
3868 (a) Failures to set uid/gid, the current directory, or a process leader
3869 in a subprocess such as that created by queryprogram now generate
3870 suitable debugging output when -d is set.
3872 (b) The queryprogram router detects when it is not running as root,
3873 outputs suitable debugging information if -d is set, and then runs
3874 the subprocess without attempting to change uid/gid.
3876 PH/15 Minor change to Makefile for building test_host (undocumented testing
3879 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3880 additional section of a DNS packet that returns MX or SRV records.
3881 Instead, it always explicitly searches for A/AAAA records. This avoids
3882 major problems that occur when a DNS server includes only records of one
3883 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3884 fixed another bug: if SRV records were looked up and the corresponding
3885 address records were *not* found in the additional section, the port
3886 values from the SRV records were lost.
3888 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3889 using the correct key (the original address) when searching the retry
3890 rules in order to find which one to use for generating the retry hint.
3892 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3893 adding the default one. Similarly, if it contains a Reply-To: header, the
3894 errors_reply_to option, if set, is not used.
3896 PH/19 When calculating a retry time, Exim used to measure the "time since
3897 failure" by looking at the "first failed" field in the retry record. Now
3898 it does not use this if it is later than than the arrival time of the
3899 message. Instead it uses the arrival time. This makes for better
3900 behaviour in cases where some deliveries succeed, thus re-setting the
3901 "first failed" field. An example is a quota failure for a huge message
3902 when small messages continue to be delivered. Without this change, the
3903 "time since failure" will always be short, possible causing more frequent
3904 delivery attempts for the huge message than are intended.
3905 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3907 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3908 $1, $2, $3) because the numerical variables can be reset during some
3909 expansion items (e.g. "match"), thereby losing the authentication data.
3911 PH/21 Make -bV show the size of off_t variables so that the test suite can
3912 decide whether to run tests for quotas > 2G.
3914 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3915 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3916 filecount value is greater than 2G or if a quota value is greater than 2G
3917 on a system where the size of off_t is not greater than 4, a panic error
3920 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3921 never match. The debug and -bh output now contains an explicit error
3922 message indicating a malformed IPv4 address or mask.
3924 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3925 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3926 PH/23 above applies.
3928 PH/25 Do not write to syslog when running in the test harness. The only
3929 occasion when this arises is a failure to open the main or panic logs
3930 (for which there is an explicit test).
3932 PH/26 Added the /no_tell option to "control=freeze".
3934 PH/27 If a host name lookup failed very early in a connection, for example, if
3935 the IP address matched host_lookup and the reverse lookup yielded a name
3936 that did not have a forward lookup, an error message of the form "no IP
3937 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3938 could be logged. Now it outputs the IP address instead of "NULL".
3940 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3941 allows the sender and the authenticated sender to be set when
3942 submitting a message from within Exim. Since child_open_exim() is
3943 documented for local_scan(), the new function should be too.
3945 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3946 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3947 results in an empty string is now treated as unset.
3949 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3951 PH/31 Added sender_verify_fail logging option.
3953 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3954 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3955 tidied the source and removed it altogether.
3957 PH/33 When a queue run was abandoned because the load average was too high, a
3958 log line was always written; now it is written only if the queue_run log
3959 selector is set. In addition, the log line for abandonment now contains
3960 information about the queue run such as the pid. This is always present
3961 in "start" and "stop" lines but was omitted from the "abandon" line.
3963 PH/34 Omit spaces between a header name and the colon in the error message that
3964 is given when verify = headers_syntax fails (if there are lots of them,
3965 the message gets confusing).
3967 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3968 names, as there are now some PTR records that contain slashes. This check
3969 is only to protect against broken name servers that fall over on strange
3970 characters, so the fact that it applies to all lookups doesn't matter.
3972 PH/36 Now that the new test suite is complete, we can remove some of the
3973 special code in Exim that was needed for the old test suite. For example,
3974 sorting DNS records because real resolvers return them in an arbitrary
3975 order. The new test suite's fake resolver always returns records in the
3978 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3979 bounces) except when queue_only is set, to avoid logging races between
3980 the different processes.
3982 PH/38 Panic-die if .include specifies a non-absolute path.
3984 PH/39 A tweak to the "H" retry rule from its user.
3986 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3987 a label. They prevented compilation on older perls.
3989 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3990 a warning to be raised on newish perls.
3992 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3993 on queue. Changes to match documented behaviour of showing count of
3994 messages matching specified criteria.
3996 PH/40 Changed the default ident timeout from 30s to 5s.
3998 PH/41 Added support for the use of login_cap features, on those BSD systems
3999 that have them, for controlling the resources used by pipe deliveries.
4001 PH/42 The content-scanning code uses fopen() to create files in which to put
4002 message data. Previously it was not paying any attention to the mode of
4003 the files. Exim runs with umask(0) because the rest of the code creates
4004 files with open(), and sets the required mode explicitly. Thus, these
4005 files were ending up world-writeable. This was not a big issue, because,
4006 being within the spool directory, they were not world-accessible. I have
4007 created a function called modefopen, which takes an additional mode
4008 argument. It sets umask(777), creates the file, chmods it to the required
4009 mode, then resets the umask. All the relevant calls to fopen() in the
4010 content scanning code have been changed to use this function.
4012 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
4013 to 24 hours. This avoids potential overflow problems when processing G
4014 and H retry rules. I suspect nobody ever tinkers with this value.
4016 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
4018 PH/45 When the plaintext authenticator is running as a client, the server's
4019 challenges are checked to ensure they are valid base64 strings. By
4020 default, the authentication attempt is cancelled if an invalid string is
4021 received. Setting client_ignore_invalid_base64 true ignores these errors.
4022 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
4023 they are received. Thus, the responses can be made to depend on the
4024 challenges. If an invalid string is ignored, an empty string is placed in
4027 PH/46 Messages that are created by the autoreply transport now contains a
4028 References: header, in accordance with RFCs 2822 and 3834.
4030 PH/47 Added authenticated_sender_force to the smtp transport.
4032 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
4034 PH/49 Installed latest patch from the Sieve maintainer.
4036 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
4037 was also set, the appendfile transport was unnecessarily scanning a
4038 directory of message files (e.g. for maildir delivery) to find the count
4039 of files (along with the size), even though it did not need this
4040 information. It now does the scan only if it needs to find either the
4041 size of the count of files.
4043 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
4045 PH/52 Two bugs concerned with error handling when the smtp transport is
4048 (i) Exim was not creating retry information for temporary errors given
4049 for individual recipients after the DATA command when the smtp transport
4050 was used in LMTP mode. This meant that they could be retried too
4051 frequently, and not timed out correctly.
4053 (ii) Exim was setting the flag that allows error details to be returned
4054 for LMTP errors on RCPT commands, but not for LMTP errors for individual
4055 recipients that were returned after the DATA command.
4057 PH/53 This is related to PH/52, but is more general: for any failing address,
4058 when detailed error information was permitted to be returned to the
4059 sender, but the error was temporary, then after the final timeout, only
4060 "retry timeout exceeded" was returned. Now it returns the full error as
4061 well as "retry timeout exceeded".
4063 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
4064 do this, and (what is worse) MTAs that accept it.
4066 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
4067 will now be deprecated.
4069 PH/56 New os.c-cygwin from the Cygwin maintainer.
4071 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
4072 formats (previously only available in exim formats via -bpr, -bpru,
4073 and -bpra. Now also available in native and exiqgrep formats)
4075 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
4076 with very large, slow to parse queues
4078 JJ/08 exipick: added ! as generic prefix to negate any criteria format
4080 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
4082 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
4083 responses to authentication challenges, though it was showing the
4084 challenges; (ii) I've removed the CR characters from the debug output for
4087 PH/58 Allow for the insertion of a newline as well as a space when a string
4088 is turned into more than one encoded-word during RFC 2047 encoding. The
4089 Sieve code now uses this.
4091 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
4092 data_4xx, lost_connection, tls_required.
4094 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
4095 message was being sent as an SMTP response.
4097 PH/61 Add -l and -k options to exicyclog.
4099 PH/62 When verifying, if an address was redirected to one new address, so that
4100 verification continued, and the new address failed or deferred after
4101 having set something in $address_data, the value of $address_data was not
4102 passed back to the ACL. This was different to the case when no
4103 redirection occurred. The value is now passed back in both cases.
4105 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
4106 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
4107 use login_cap.h, so on its own it isn't the distinguishing feature. The
4108 new name refers directly to the setclassresources() function.
4110 PH/65 Added configuration files for NetBSD3.
4112 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
4114 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
4115 is preferred over IPv4.
4117 PH/68 The bounce_return_message and bounce_return_body options were not being
4118 honoured for bounces generated during the reception of non-SMTP messages.
4119 In particular, this applied to messages rejected by the ACL. This bug has
4120 been fixed. However, if bounce_return_message is true and bounce_return_
4121 body is false, the headers that are returned for a non-SMTP message
4122 include only those that have been read before the error was detected.
4123 (In the case of an ACL rejection, they have all been read.)
4125 PH/69 The HTML version of the specification is now built in a directory called
4126 spec_html instead of spec.html, because the latter looks like a path with
4127 a MIME-type, and this confuses some software.
4129 PH/70 Catch two compiler warnings in sieve.c.
4131 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
4132 function verify_get_ident() calls ip_connect() to connect a socket, but
4133 if the "connect()" function timed out, ip_connect() used to close the
4134 socket. However, verify_get_ident() also closes the socket later, and in
4135 between Exim writes to the log, which may get opened at this point. When
4136 the socket was closed in ip_connect(), the log could get the same file
4137 descriptor number as the socket. This naturally causes chaos. The fix is
4138 not to close the socket in ip_connect(); the socket should be closed by
4139 the function that creates it. There was only one place in the code where
4140 this was missing, in the iplookup router, which I don't think anybody now
4141 uses, but I've fixed it anyway.
4143 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
4144 well as to direct DNS lookups. Otherwise the handling of names in host
4145 lists is inconsistent and therefore confusing.
4151 PH/01 Two changes to the default runtime configuration:
4153 (1) Move the checks for relay_from_hosts and authenticated clients from
4154 after to before the (commented out) DNS black list checks.
4156 (2) Add control=submission to the relay_from_hosts and authenticated
4157 clients checks, on the grounds that messages accepted by these
4158 statements are most likely to be submissions.
4160 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
4162 (1) Generate an error if the third argument for the ${prvs expansion is
4165 (2) Treat a missing third argument of ${prvscheck as if it were an empty
4168 (3) Reset the variables that are obtained from the first argument of
4169 ${prvscheck and used in the second argument before leaving the code,
4170 because their memory is reclaimed, so using them afterwards may do
4173 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
4174 one (it's much easier than Tom thought :-).
4176 (5) Because of (4), we can now allow for the use of $prvscheck_result
4177 inside the third argument.
4179 PH/03 For some reason, the default setting of PATH when running a command from
4180 a pipe transport was just "/usr/bin". I have changed it to
4183 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
4184 anything to be listed in the output from -bV.
4186 PH/05 When a filter generated an autoreply, the entire To: header line was
4187 quoted in the delivery log line, like this:
4189 => >A.N.Other <ano@some.domain> <original@ddress> ...
4191 This has been changed so that it extracts the operative address. There
4192 may be more than one such address. If so, they are comma-separated, like
4195 => >ano@some.domain,ona@other.domain <original@ddress> ...
4197 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
4198 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
4199 not being looked up in the rDNS to get a host name, Exim was showing the
4200 IP address twice in Received: lines, even though the IP addresses were
4201 identical. For example:
4203 Received: from [1.2.3.4] (helo=[1.2.3.4])
4205 However, if the real host name was known, it was omitting the HELO data
4206 if it matched the actual IP address. This has been tidied up so that it
4207 doesn't show the same IP address twice.
4209 PH/07 When both +timestamp and +memory debugging was on, the value given by
4210 $tod_xxx expansions could be wrong, because the tod_stamp() function was
4211 called by the debug printing, thereby overwriting the timestamp buffer.
4212 Debugging no longer uses the tod_stamp() function when +timestamp is set.
4214 PH/08 When the original message was included in an autoreply transport, it
4215 always said "this is a copy of the message, including all the headers",
4216 even if body_only or headers_only was set. It now gives an appropriate
4219 PH/09 Applied a patch from the Sieve maintainer which:
4221 o fixes some comments
4222 o adds the (disabled) notify extension core
4223 o adds some debug output for the result of if/elsif tests
4224 o points to the current vacation draft in the documentation
4225 and documents the missing references header update
4229 o fixes a bug in processing the envelope test (when testing
4230 multiple envelope elements, the last element determined the
4233 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
4234 Electronic Mail") by including:
4236 Auto-submitted: auto-generated
4238 in the messages that it generates (bounce messages and others, such as
4239 warnings). In the case of bounce messages for non-SMTP messages, there was
4240 also a typo: it was using "Auto_submitted" (underscore instead of
4241 hyphen). Since every message generated by Exim is necessarily in response
4242 to another message, thes have all been changed to:
4244 Auto-Submitted: auto-replied
4246 in accordance with these statements in the RFC:
4248 The auto-replied keyword:
4250 - SHOULD be used on messages sent in direct response to another
4251 message by an automatic process,
4253 - MUST NOT be used on manually-generated messages,
4255 - MAY be used on Delivery Status Notifications (DSNs) and Message
4256 Disposition Notifications (MDNs),
4258 - MUST NOT be used on messages generated by automatic or periodic
4259 processes, except for messages which are automatic responses to
4262 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
4263 to the default Received: header definition.
4265 PH/12 Added log selector acl_warn_skipped (default on).
4267 PH/13 After a successful wildlsearch lookup, discard the values of numeric
4268 variables because (a) they are in the wrong storage pool and (b) even if
4269 they were copied, it wouldn't work properly because of the caching.
4271 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
4272 checking when decoding. Apparently there are clients that generate
4273 overlong encoded strings. Why am I not surprised?
4275 PH/15 If the first argument of "${if match_address" was not empty, but did not
4276 contain an "@" character, Exim crashed. Now it writes a panic log message
4277 and treats the condition as false.
4279 PH/16 In autoreply, treat an empty string for "once" the same as unset.
4281 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
4282 extension "envelope-auth". The code is finished and in agreement with
4283 other implementations, but there is no documentation so far and in fact,
4284 nobody wrote the draft yet. This extension is currently #undef'ed, thus
4285 not changing the active code.
4287 Print executed "if" and "elsif" statements when debugging is used. This
4288 helps a great deal to understand what a filter does.
4290 Document more things not specified clearly in RFC3028. I had all this
4291 sorted out, when out of a sudden new issues came to my mind. Oops."
4293 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
4296 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
4297 canonical form (as documented). However, after a host name lookup from
4298 the IP address, check_host() was doing a simple string comparison with
4299 addresses acquired from the DNS when checking that the found name did
4300 have the original IP as one of its addresses. Since any found IPv6
4301 addresses are likely to be in abbreviated form, the comparison could
4302 fail. Luckily, there already exists a function for doing the comparison
4303 by converting both addresses to binary, so now that is used instead of
4304 the text comparison.
4306 PH/20 There was another similar case to PH/19, when a complete host name was
4307 given in a host list; looking up its IP address could give an abbreviated
4308 form, whereas the current host's name might or might not be abbreviated.
4309 The same fix has been applied.
4315 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
4316 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
4319 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
4320 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
4322 PH/03 Typo: missing ".o" in src/pcre/Makefile.
4324 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
4325 header line, restrict the check to what is listed in RFCs 2369 and 2929.
4326 Also, for "Auto-Submitted", treat anything other than "no" as
4327 non-personal, in accordance with RFC 3834. (Previously it treated
4328 anything starting "auto-" as non-personal.)
4330 TF/01 The control=submission/name=... option had a problem with syntax
4331 errors if the name included a slash character. The /name= option
4332 now slurps the rest of the string, so it can include any characters
4333 but it must come last in the list of options (after /sender_retain
4336 PH/05 Some modifications to the interface to the fake nameserver for the new
4344 TK/01 Added the "success_on_redirect" address verification option. See
4345 NewStuff for rationale and an example.
4347 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
4349 PH/02 Patch to exigrep to allow it to work on syslog lines.
4351 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
4352 fread() to skip over the body file's header line, because in Cygwin the
4353 header line is locked and is inaccessible.
4355 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
4356 co-exist for some time) to make it clear that it is the Exim ID that is
4357 referenced, not the Message-ID: header line.
4359 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
4360 string_format() function, because snprintf() does not exist on all
4363 PH/06 The use of forbid_filter_existstest now also locks out the use of the
4364 ${stat: expansion item.
4366 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
4367 protocol synchronization error", to keep the pedants happy.
4369 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
4370 well as for IRIX systems, when gcc is being used. See the host.c source
4373 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
4375 PH/10 Named domain lists were not working if used in a queue_smtp_domains
4378 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
4379 transport and to the smtp transport in LMTP mode.
4381 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
4383 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
4384 run a filter in a subprocess. This could lead to confusion in subsequent
4385 lookups in the parent process. There should also be a search_tidyup() at
4386 the end of the subprocess.
4388 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
4389 only if the host matched helo_try_verify_hosts, which caused the
4390 verification to occur when the EHLO/HELO command was issued. The ACL just
4391 tested the remembered result. Now, if a previous verification attempt has
4392 not happened, "verify = helo" does it there and then.
4394 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
4396 TK/03 Fix log output including CR from clamd.
4398 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
4399 exist provoked a memory error which could cause a segfault.
4401 PH/15 Installed PCRE 6.2
4403 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
4405 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
4406 of the problem. Specifically, suggested +O2 rather than +O1 for the
4409 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
4411 PH/20 If a delivery was routed to a non-standard port by means of an SRV
4412 record, the port was not correctly logged when the outgoing_port log
4413 selector was set (it logged the transort's default port).
4415 PH/21 Added support for host-specific ports to manualroute, queryprogram,
4416 fallback_hosts, and "hosts" in the smtp transport.
4418 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
4419 host errors such as "Connection refused".
4421 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
4422 authentication with radiusclient 0.4.9:
4424 - Error returned from rc_read_config was caught wrongly
4425 - Username/password not passed on to radius server due to wrong length.
4427 The presumption is that some radiusclient API changes for 4.51/PH/17
4428 were not taken care of correctly. The code is still untested by me (my
4429 Linux distribution still has 0.3.2 of radiusclient), but it was
4430 contributed by a Radius user.
4432 PH/24 When doing a callout, the value of $domain wasn't set correctly when
4433 expanding the "port" option of the smtp transport.
4435 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
4436 while reading a MIME header. Thanks to Tom Hughes for a patch.
4438 PH/24 Include config.h inside local_scan.h so that configuration settings are
4441 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
4442 that in an ACL for RCPT (for example), you can examine exactly what was
4445 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
4446 commands, but it was not correctly comparing the address with the actual
4447 client host address. Thus, it would show the EHLO address in Received:
4448 header lines when this was not necessary.
4450 PH/27 Added the % operator to ${eval:}.
4452 PH/28 Exim tries to create and chdir to its spool directory when it starts;
4453 it should be ignoring failures (because with -C, for example, it has lost
4454 privilege). It wasn't ignoring creation failures other than "already
4457 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
4460 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
4461 because an input line was too long, either on its own, or by virtue of
4462 too many continuations, the temporary file was not being removed, and the
4463 return code was incorrect.
4465 PH/31 Missing "BOOL" in function definition in filtertest.c.
4467 PH/32 Applied Sieve patches from the maintainer.
4469 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
4471 PH/33 Added "verify = not_blind".
4473 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
4474 Local/Makefile (with some defaults set). These are used in built scripts
4475 such as exicyclog, but they have never been used in the exim_install
4476 script (though there are many overriding facilities there). I have
4477 arranged that the exim_install script now takes note of these two
4480 PH/35 Installed configuration files for Dragonfly.
4482 PH/36 When a locally submitted message by a trusted user did not contain a
4483 From: header, and the sender address was obtained from -f or from an SMTP
4484 MAIL command, and the trusted user did not use -F to supply a sender
4485 name, $originator_name was incorrectly used when constructing a From:
4486 header. Furthermore, $originator_name was used for submission mode
4487 messages from external hosts without From: headers in a similar way,
4488 which is clearly wrong.
4490 PH/37 Added control=suppress_local_fixups.
4492 PH/38 When log_selector = +received_sender was set, and the addition of the
4493 sender made the log line's construction buffer exactly full, or one byte
4494 less than full, an overflow happened when the terminating "\n" was
4497 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
4498 when the result of a list match is failure because a DNS lookup failed.
4500 PH/40 RM_COMMAND is now used in the building process.
4502 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
4503 the "build-* directories that it finds.
4505 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
4506 address in a domain literal was a prefix of an interface address.
4508 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
4509 when verifying a sender address, unless rewrite_headers is false.
4511 PH/44 Wrote a long comment about why errors_to addresses are verified as
4512 recipients, not senders.
4514 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
4515 the ratelimit ACL was added.
4517 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
4519 PH/46 Added extra information about PostgreSQL errors to the error string.
4521 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
4522 avoiding the need to install special zones in a real server. This is
4523 backwards compatible; if it can't find the fake resolver, it drops back.
4524 Thus, both old and new test suites can be run.
4526 TF/02 Added util/ratelimit.pl
4528 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
4529 clock is set back in time.
4531 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
4532 Candler <B.Candler@pobox.com>.
4534 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
4535 OK for addresses that are the result of redirections.
4537 PH/48 A number of further additions for the benefit of the new test suite,
4538 including a fake gethostbyname() that interfaces to the fake DNS resolver
4541 TF/06 The fix for widen_domains has also been applied to qualify_single and
4542 search_parents which are the other dnslookup options that can cause
4545 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
4548 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
4550 TF/07 Exim produced the error message "an SRV record indicated no SMTP
4551 service" if it encountered an MX record with an empty target hostname.
4552 The message is now "an MX or SRV record indicated no SMTP service".
4554 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
4555 if the DNS of the sending site is misconfigured. This is quite a
4556 common situation. This change restores the behaviour of treating a
4557 helo verification defer as a failure.
4559 PH/51 If self=fail was set on a router, the bounce message did not include the
4560 actual error message.
4566 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
4568 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4569 command itself ended in error, the underlying message about the transport
4570 filter timeout was being overwritten with the pipe command error. Now the
4571 underlying error message should be appended to the second error message.
4573 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4575 PH/02 Reduce the amount of output that "make" produces by default. Full output
4576 can still be requested.
4578 PH/03 The warning log line about a condition test deferring for a "warn" verb
4579 was being output only once per connection, rather than after each
4580 occurrence (because it was using the same function as for successful
4581 "warn" verbs). This seems wrong, so I have changed it.
4583 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4584 it should not have, which might have caused a crash in the right
4585 circumstances, but probably never did.
4587 PH/04 Installed a modified version of Tony Finch's patch to make submission
4588 mode fix the return path as well as the Sender: header line, and to
4589 add a /name= option so that you can make the user's friendly name appear
4592 TF/03 Added the control = fakedefer ACL modifier.
4594 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4595 Mark Lowes for thorough testing.
4597 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4599 TK/03 Merged latest SRS patch from Miles Wilton.
4601 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4602 with the definition in sysexits.h (which is #included earlier).
4603 Fortunately, Exim does not actually use EX_OK. The code used to try to
4604 preserve the sysexits.h value, by assuming that macro definitions were
4605 scanned for macro replacements. I have been disabused of this notion,
4606 so now the code just undefines EX_OK before #including unistd.h.
4608 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4609 in the smtp transport. When a block could not be written in a single
4610 write() function, the timeout was being re-applied to each part-write.
4611 This seems wrong - if the receiver was accepting one byte at a time it
4612 would take for ever. The timeout is now adjusted when this happens. It
4613 doesn't have to be particularly precise.
4615 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4616 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4618 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4619 without a domain if the check to <postmaster@domain> fails.
4621 SC/01 Eximstats: added -xls and the ability to specify output files
4622 (patch written by Frank Heydlauf).
4624 SC/02 Eximstats: use FileHandles for outputting results.
4626 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4628 SC/04 Eximstats: fixed display of large numbers with -nvr option
4630 SC/05 Eximstats: fixed merging of reports with empty tables.
4632 SC/06 Eximstats: added the -include_original_destination flag
4634 SC/07 Eximstats: removed tabs and trailing whitespace.
4636 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4638 TK/06 MBOX spool code: Add real "From " MBOX separator line
4639 so the .eml file is really in mbox format (even though
4640 most programs do not really care). Patch from Alex Miller.
4642 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4643 The latter is generated from $received_to and is only set if the
4644 message has one envelope recipient. SA can use these headers,
4645 obviously out-of-the-box. Patch from Alex Miller.
4647 PH/08 The ${def test on a variable was returning false if the variable's
4648 value was "0", contrary to what the specification has always said!
4649 The result should be true unless the variable is empty.
4651 PH/09 The syntax error of a character other than { following "${if
4652 def:variable_name" (after optional whitespace) was not being diagnosed.
4653 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4654 accidental colon was present, for example, could give incorrect results.
4656 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4657 result is used (not including appendfile, where other changes are about
4660 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4661 This involved changing a lot of size variables from int to off_t. It
4662 should work with maildirs and everything.
4664 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4665 spamd dying while we are connected to it.
4667 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4670 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4671 function for BDB 4.3.
4673 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4675 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4676 the sources for PCRE have been re-arranged and more files are now
4679 PH/15 The code I had for printing potentially long long variables in PH/11
4680 above was not the best (it lost precision). The length of off_t variables
4681 is now inspected at build time, and an appropriate printing format (%ld
4682 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4683 to be "long long int" or "long int". This is needed for the internal
4684 formatting function string_vformat().
4686 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4687 the configuration file to be ":syslog", then the script "guesses" where
4688 the logs files are, rather than using the compiled in default. In our
4689 case the guess is not the same as the compiled default, so the script
4690 suddenly stopped working when I started to use syslog. The patch checks
4691 to see if log_file_path is "". If so, it attempts to read it from exim
4692 with no configuration file to get the compiled in version, before it
4693 falls back to the previous guessing code."
4695 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4696 implementing BATV in an Exim configuration. See NewStuff for the gory
4699 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4700 Makefile that are specific to HP-UX.
4702 PH/18 If the "use_postmaster" option was set for a recipient callout together
4703 with the "random" option, the postmaster address was used as the MAIL
4704 FROM address for the random test, but not for the subsequent recipient
4705 test. It is now used for both.
4707 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4708 patch removes a few documentation additions to RFC 3028, because the
4709 latest draft now contains them. It adds the new en;ascii-case comparator
4710 and a new error check for 8bit text in MIME parts. Comparator and
4711 require names are now matched exactly. I enabled the subaddress
4712 extension, but it is not well tested yet (read: it works for me)."
4714 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4715 rework some of the code of TK/09 above to avoid the hardwired use of
4716 "%lld" and "long long". Replaced the call to snprintf() with a call to
4719 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4720 records point to non-existent hosts", "retry timeout exceeded", and
4721 "retry time not reached for any host after a long failure period".
4723 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4724 experimental DomainKeys support:
4726 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4727 (2) On an error such as an illegally used "control", the wrong name for
4728 the control was given.
4730 These problems did NOT occur unless DomainKeys support was compiled.
4732 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4734 PH/24 Added ${if match_ip condition.
4736 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4737 databases so that it will be absolutely obvious if a crash occurs in the
4738 DB library. This is a regular occurrence (often caused by mis-matched
4741 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4742 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4743 picked up on a user's system that detects such things. There doesn't seem
4744 to be a gcc warning option for this - only an attribute that has to be
4745 put on the function's prototype. It seems that in Fedora Core 4 they have
4746 set this on a number of new functions. No doubt there will be more in due
4749 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4750 specify a transport. However, if an address that was verified by such a
4751 router was the subject of a callout, Exim crashed because it tried to
4752 read the rcpt_include_affixes from the non-existent transport. Now it
4753 just assumes that the setting of that option is false. This bug was
4754 introduced by 4.51/PH/31.
4756 PH/28 Changed -d+all to exclude +memory, because that information is very
4757 rarely of interest, but it makes the output a lot bigger. People tend to
4758 do -d+all out of habit.
4760 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4761 code in os-type was giving problems when libc.so lives in lib64, like on
4764 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4765 aren't the modern standard, and it seems that some systems' include files
4766 don't always have them. Exim was already checking for some of the newer
4767 ones like T_AAAA, and defining it itself. I've added checks for all the
4768 record types that Exim uses.
4770 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4771 not automatically generating a new one, as it is supposed to. This
4772 prevented TLS from working. If the file did exist, but contained invalid
4773 data, a new version was generated, as expected. It was only the case of a
4774 non-existent file that was broken.
4776 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4777 with a change in libdomainkeys > 0.64.
4779 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4780 from DNS. If the selector record carries the flag, it now has
4781 precedence over the domain-wide flag.
4783 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4785 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4786 the use of a port name that isn't defined in /etc/services) occurred, the
4787 message was deferred as in a normal delivery, and thus remained on the
4788 spool, instead of being failed because of the mua_wrapper setting. This
4789 is now fixed, and I tidied up some of the mua_wrapper messages at the
4792 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4793 the messages in a hash of arrays rather than using individual hashes.
4794 This is a bit cleaner and results in dramatic memory savings, albeit
4795 at a slight CPU cost.
4797 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4798 as requested by Marc Sherman.
4800 SC/10 Eximstats: added histograms for user specified patterns as requested
4803 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4805 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4806 fopen() in the content-scanning modules that did not already have it.
4812 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4813 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4815 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4817 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4819 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4820 to test Sieve filters that use "vacation".
4822 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4823 that changes the way the GnuTLS parameters are stored in the cache file.
4824 The new format can be generated externally. For backward compatibility,
4825 if the data in the cache doesn't make sense, Exim assumes it has read an
4826 old-format file, and it generates new data and writes a new file. This
4827 means that you can't go back to an older release without removing the
4830 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4831 work if there are any delivery delays because "one_time" forces the
4832 parent to be marked "delivered", so its unseen clone is never tried
4833 again. For this reason, Exim now forbids the simultaneous setting of
4836 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4837 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4838 entry if you want to know the details. The fix, however, neglected to
4839 consider the case when local delivery batching is involved. The test for
4840 "previously delivered" was not happening when checking to see if an
4841 address could be batched with a previous (undelivered) one; under
4842 certain circumstances this could lead to multiple deliveries to the same
4845 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4846 in its include files, and this causes problems building Exim.
4848 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4849 header_syntax) but Exim was just ignoring anything given after a slash.
4850 In particular, this caused confusion with an attempt to use "verify =
4851 reverse_host_lookup/defer_ok". An error is now given when options are
4852 supplied for verify items that do not have them. (Maybe reverse_host_
4853 lookup should have a defer_ok option, but that's a different point.)
4855 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4856 defined by RFC 821) to 2048, because there were problems with some AUTH
4857 commands, and RFC 1869 says the size should be increased for extended
4858 SMTP commands that take arguments.
4860 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4863 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4864 "unknown" error; now it says that the functionality isn't in the binary.
4866 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4867 an address' error message when a string expansion fails (syntax or
4868 whatever). Otherwise the password may appear in the log. Following change
4869 PH/42 below, there is no longer a chance of it appearing in a bounce
4872 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4874 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4875 of its addresses was ever tried. (Bugzilla bug #2.)
4877 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4878 the result incorrectly in the debug output. (It correctly added a newline
4879 to what was transported.)
4881 TF/01 Added $received_time.
4883 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4884 commented out examples of how to interface to a virus scanner and to
4885 SpamAssassin. Also added commented examples of av_scanner and
4886 spamd_address settings.
4888 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4889 and controls are allowed in which ACLs. There were a couple of minor
4890 errors. Some of the entries in the conditions table (which is a table of
4891 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4892 negation of where the condition IS allowed.
4894 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4896 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4897 header file does not have a version number, so I've had to invent a new
4898 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4899 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4900 radiusclient), but it was contributed by a Radius user.
4902 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4903 files or directories, for OpenSSL.
4905 PH/19 When an Exim process that is running as root has to create an Exim log
4906 file, it does so in a subprocess that runs as exim:exim so as to get the
4907 ownership right at creation (otherwise, other Exim processes might see
4908 the file with the wrong ownership). There was no test for failure of this
4909 fork() call, which would lead to the process getting stuck as it waited
4910 for a non-existent subprocess. Forks do occasionally fail when resources
4911 run out. I reviewed all the other calls to fork(); they all seem to check
4914 PH/20 When checking for unexpected SMTP input at connect time (before writing
4915 the banner), Exim was not dealing correctly with a non-positive return
4916 from the read() function. If the client had disconnected by this time,
4917 the result was a log entry for a synchronization error with an empty
4918 string after "input=" when read() returned zero. If read() returned -1
4919 (an event I could not check), uninitialized data bytes were printed.
4920 There were reports of junk text (parts of files, etc) appearing after
4923 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4925 PH/22 Added support for macro redefinition, and (re)definition in between
4926 driver and ACL definitions.
4928 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4929 forgetting to use the resulting value; it was using the unexpanded value.
4931 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4932 hadn't been configured. The fix is from Juergen Kreileder, who
4933 understands it better than I do:
4935 "Here's what I see happening with three configured cyrus_sasl
4936 authenticators configured (plain, login, cram-md5):
4938 On startup auth_cyrus_sasl_init() gets called for each of these.
4939 This means three calls to sasl_listmech() without a specified mech_list.
4940 => SASL tests which mechs of all available mechs actually work
4941 => three warnings about OTP not working
4942 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4944 With the patch, sasl_listmech() also gets called three times. But now
4945 SASL's mech_list option is set to the server_mech specified in the the
4946 authenticator. Or in other words, the answer from sasl_listmech()
4947 gets limited to just the mech you're testing for (which is different
4949 => the return list contains just 'plain' or 'login', 'cram-md5' or
4950 nothing depending on the value of ob->server_mech.
4952 I've just tested the patch: Authentication still works fine,
4953 unavailable mechs specified in the exim configuration are still
4954 caught, and the auth.log warnings about OTP are gone."
4956 PH/25 When debugging is enabled, the contents of the command line are added
4957 to the debugging output, even when log_selector=+arguments is not
4960 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4961 answer is "GNU", and only if the return is "GNU/something" is the answer
4964 PH/27 $acl_verify_message is now set immediately after the failure of a
4965 verification in an ACL, and so is available in subsequent modifiers. In
4966 particular, the message can be preserved by coding like this:
4968 warn !verify = sender
4969 set acl_m0 = $acl_verify_message
4971 Previously, $acl_verify_message was set only while expanding "message"
4972 and "log_message" when a very denied access.
4974 PH/28 Modified OS/os.c-Linux with
4976 -#ifndef OS_LOAD_AVERAGE
4977 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4979 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4980 nomenclature these days.)
4982 PH/29 Installed patch from the Sieve maintainer that adds the options
4983 sieve_useraddress and sieve_subaddress to the redirect router.
4985 PH/30 In these circumstances:
4986 . Two addresses routed to the same list of hosts;
4987 . First host does not offer TLS;
4988 . First host accepts first address;
4989 . First host gives temporary error to second address;
4990 . Second host offers TLS and a TLS session is established;
4991 . Second host accepts second address.
4992 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4993 and peerdn, if requested) that were in fact used only for the second
4996 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4997 attention to any local part prefix or suffix that was matched by the
4998 router that accepted the address. It now behaves in the same way as it
4999 does for delivery: the affixes are removed from the local part unless
5000 rcpt_include_affixes is set on the transport.
5002 PH/32 Add the sender address, as F=<...>, to the log line when logging a
5003 timeout during the DATA phase of an incoming message.
5005 PH/33 Sieve envelope tests were broken for match types other than :is. I have
5006 applied a patch sanctioned by the Sieve maintainer.
5008 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
5009 the uid or gid is negative. A case of a negative gid caused this to be
5010 noticed. The fix allows for either to be negative.
5012 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
5013 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
5016 PH/36 The change PH/12 above was broken. Fixed it.
5018 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
5019 the grounds that routing the same address twice would always produce the
5020 same answer. This might have been true once, but it is certainly no
5021 longer true now. Routing a child address may depend on the previous
5022 routing that produced that child. Some complicated redirection strategies
5023 went wrong when messages had multiple recipients, and made Exim's
5024 behaviour dependent on the order in which the addresses were given.
5026 I have moved the duplicate checking until after the routing is complete.
5027 Exim scans the addresses that are assigned to local and remote
5028 transports, and removes any duplicates. This means that more work will be
5029 done, as duplicates will always all be routed, but duplicates are
5030 presumably rare, so I don't expect this is of any significance.
5032 For deliveries to pipes, files, and autoreplies, the duplicate checking
5033 still happens during the routing process, since they are not going to be
5036 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
5037 It corrects a timeout issue with spamd. This is Ian's comment: "The
5038 background is that sometimes spamd either never reads data from a
5039 connection it has accepted, or it never writes response data. The exiscan
5040 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
5041 blindly assumes that writes won't block so it may never time out."
5043 PH/39 Allow G after quota size as well as K and M.
5045 PH/40 The value set for $authenticated_id in an authenticator may not contain
5046 binary zeroes or newlines because the value is written to log lines and
5047 to spool files. There was no check on this. Now the value is run through
5048 the string_printing() function so that such characters are converted to
5049 printable escape sequences.
5051 PH/41 $message_linecount is a new variable that contains the total number of
5052 lines in the message. Compare $body_linecount, which is the count for the
5055 PH/42 Exim no longer gives details of delivery errors for specific addresses in
5056 bounce and delay warning messages, except in certain special cases, which
5059 (a) An SMTP error message from a remote host;
5060 (b) A message specified in a :fail: redirection;
5061 (c) A message specified in a "fail" command in a system filter;
5062 (d) A message specified in a FAIL return from the queryprogram router;
5063 (e) A message specified by the cannot_route_message router option.
5065 In these cases only, Exim does include the error details in bounce and
5066 warning messages. There are also a few cases where bland messages such
5067 as "unrouteable address" or "local delivery error" are given.
5069 PH/43 $value is now also set for the "else" part of a ${run expansion.
5071 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
5072 being worked on, but at least Exim now implements the latest version to
5075 PH/45 In a pipe transport, although a timeout while waiting for the pipe
5076 process to complete was treated as a delivery failure, a timeout while
5077 writing the message to the pipe was logged, but erroneously treated as a
5078 successful delivery. Such timeouts include transport filter timeouts. For
5079 consistency with the overall process timeout, these timeouts are now
5080 treated as errors, giving rise to delivery failures by default. However,
5081 there is now a new Boolean option for the pipe transport called
5082 timeout_defer, which, if set TRUE, converts the failures into defers for
5083 both kinds of timeout. A transport filter timeout is now identified in
5086 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
5087 systems where "make" and "gmake" are different, calling "gmake" at top
5088 level broke things. I've arranged for the value of $(MAKE) to be passed
5089 from the Makefile to this script so that it can call the same version of
5093 A note about Exim versions 4.44 and 4.50
5094 ----------------------------------------
5096 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
5097 changes of various kinds. As a consequence, a big documentation update was
5098 needed. This delayed the release for rather longer than seemed good, especially
5099 in the light of a couple of (minor) security issues. Therefore, the changes
5100 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
5101 release. So 4.44 and 4.50 are in effect two different branches that both start
5104 I have left the 4.50 change log unchanged; it contains all the changes since
5105 4.43. The change log for 4.44 is below; many of its items are identical to
5106 those for 4.50. This seems to be the most sensible way to preserve the
5107 historical information.
5113 1. Minor wording change to the doc/README.SIEVE file.
5115 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
5116 computation of the current number of files was incorrect.
5118 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5119 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5122 4. Give more explanation in the error message when the command for a transport
5123 filter fails to execute.
5125 5. There are several places where Exim runs a non-Exim command in a
5126 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5127 was being done only for the command run by the queryprogram router. It is
5128 now done for all such subprocesses. The other cases are: ${run, transport
5129 filters, and the commands run by the lmtp and pipe transports.
5131 6. Added CONFIGURE_GROUP build-time option.
5133 7. Some older OS have a limit of 256 on the maximum number of file
5134 descriptors. Exim was using setrlimit() to set 1000 as a large value
5135 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5136 systems. I've change it so that if it can't get 1000, it tries for 256.
5138 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5139 was an oversight, and furthermore, ever since the addition of extra
5140 controls (e.g. 4.43/32), the checks on when to allow different forms of
5141 "control" were broken. There should now be diagnostics for all cases when a
5142 control that does not make sense is encountered.
5144 9. Added the /retain_sender option to "control=submission".
5146 10. $recipients is now available in the predata ACL (oversight).
5148 11. Tidy the search cache before the fork to do a delivery from a message
5149 received from the command line. Otherwise the child will trigger a lookup
5150 failure and thereby defer the delivery if it tries to use (for example) a
5151 cached ldap connection that the parent has called unbind on.
5153 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5154 of $address_data from the recipient verification was clobbered by the
5155 sender verification.
5157 13. The value of address_data from a sender verification is now available in
5158 $sender_address_data in subsequent conditions in the ACL statement.
5160 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
5162 15. Added a new option "connect=<time>" to callout options, to set a different
5165 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5166 was its contents. (It was OK if the option was not defined at all.)
5168 17. A "Completed" log line is now written for messages that are removed from
5169 the spool by the -Mrm option.
5171 18. New variables $sender_verify_failure and $recipient_verify_failure contain
5172 information about exactly what failed.
5174 19. Added -dd to debug only the daemon process.
5176 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
5177 handles timeouts, both on the server side and network timeouts. Renamed the
5178 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
5180 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
5181 It is now set to "smtps".
5183 22. $host_address is now set to the target address during the checking of
5184 ignore_target_hosts.
5186 23. When checking ignore_target_hosts for an ipliteral router, no host name was
5187 being passed; this would have caused $sender_host_name to have been used if
5188 matching the list had actually called for a host name (not very likely,
5189 since this list is usually IP addresses). A host name is now passed as
5192 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5193 code that specifies a non-restarting handler (typically sigaction() in
5194 modern systems) in an attempt to fix a rare and obscure crash bug.
5196 25. Narrowed the window for a race in the daemon that could cause it to ignore
5197 SIGCHLD signals. This is not a major problem, because they are used only to
5198 wake it up if nothing else does.
5200 26. A malformed maildirsize file could cause Exim to calculate negative values
5201 for the mailbox size or file count. Odd effects could occur as a result.
5202 The maildirsize information is now recalculated if the size or filecount
5205 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5206 support for a long time. Removed HAVE_SYS_VFS_H.
5208 28. Installed the latest version of exipick from John Jetmore.
5210 29. In an address list, if the pattern was not a regular expression, an empty
5211 subject address (from a bounce message) matched only if the pattern was an
5212 empty string. Non-empty patterns were not even tested. This was the wrong
5213 because it is perfectly reasonable to use an empty address as part of a
5214 database query. An empty address is now tested by patterns that are
5215 lookups. However, all the other forms of pattern expect the subject to
5216 contain a local part and a domain, and therefore, for them, an empty
5217 address still always fails if the pattern is not itself empty.
5219 30. Exim went into a mad DNS loop when attempting to do a callout where the
5220 host was specified on an smtp transport, and looking it up yielded more
5221 than one IP address.
5223 31. Re-factored the code for checking spool and log partition space into a
5224 function that finds that data and another that does the check. The former
5225 is then used to implement four new variables: $spool_space, $log_space,
5226 $spool_inodes, and $log_inodes.
5228 32. The RFC2047 encoding function was originally intended for short strings
5229 such as real names; it was not keeping to the 75-character limit for
5230 encoded words that the RFC imposes. It now respects the limit, and
5231 generates multiple encoded words if necessary. To be on the safe side, I
5232 have increased the buffer size for the ${rfc2047: expansion operator from
5235 33. It is now permitted to omit both strings after an "if" condition; if the
5236 condition is true, the result is "true". As before, when the second string
5237 is omitted, a false condition yields an empty string. This makes it less
5238 cumbersome to write custom ACL and router conditions.
5240 34. Failure to deliver a bounce message always caused it to be frozen, even if
5241 there was an errors_to setting on the router. The errors_to setting is now
5244 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5245 canonical form (fully expanded) before being placed in
5246 $sender_host_address.
5248 36. The table in the code that translates DNS record types into text (T_A to
5249 "A" for instance) was missing entries for NS and CNAME. It is just possible
5250 that this could have caused confusion if both these types were looked up
5251 for the same domain, because the text type is used as part of Exim's
5252 per-process caching. But the chance of anyone hitting this buglet seems
5255 37. The dnsdb lookup has been extended in a number of ways.
5257 (1) There is a new type, "zns", which walks up the domain tree until it
5258 finds some nameserver records. It should be used with care.
5260 (2) There is a new type, "mxh", which is like "mx" except that it returns
5261 just the host names, not the priorities.
5263 (3) It is now possible to give a list of domains (or IP addresses) to be
5264 looked up. The behaviour when one of the lookups defers can be
5265 controlled by a keyword.
5267 (4) It is now possible to specify the separator character for use when
5268 multiple records are returned.
5270 38. The dnslists ACL condition has been extended: it is now possible to supply
5271 a list of IP addresses and/or domains to be looked up in a particular DNS
5274 39. Added log_selector=+queue_time_overall.
5276 40. When running the queue in the test harness, wait just a tad after forking a
5277 delivery process, to get repeatability of debugging output.
5279 41. Include certificate and key file names in error message when GnuTLS fails
5280 to set them up, because the GnuTLS error message doesn't include the name
5281 of the failing file when there is a problem reading it.
5283 42. Allow both -bf and -bF in the same test run.
5285 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
5287 44. The "Exiscan patch" is now merged into the mainline Exim source.
5289 45. Sometimes the final signoff response after QUIT could fail to get
5290 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5291 < 0 before doing a fflush(). This bug looks as though it goes back to the
5292 introduction of TLS in release 3.20, but "sometimes" must have been rare
5293 because the tests only now provoked it.
5295 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
5296 (this can affect the format of dates).
5298 47. exim_tidydb, when checking for the continued existence of a message for
5299 which it has found a message-specific retry record, was not finding
5300 messages that were in split spool directories. Consequently, it was
5301 deleting retry records that should have stayed in existence.
5303 48. Steve fixed some bugs in eximstats.
5305 49. The SPA authentication driver was not abandoning authentication and moving
5306 on to the next authenticator when an expansion was forced to fail,
5307 contradicting the general specification for all authenticators. Instead it
5308 was generating a temporary error. It now behaves as specified.
5310 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
5311 (the order specifies the preference for clients). The order is now AES256,
5312 AES128, 3DES, ARCFOUR128.
5314 51. Small patch to Sieve code - explicitly set From: when generating an
5317 52. Exim crashed if a remote delivery caused a very long error message to be
5318 recorded - for instance if somebody sent an entire SpamAssassin report back
5319 as a large number of 550 error lines. This bug was coincidentally fixed by
5320 increasing the size of one of Exim's internal buffers (big_buffer) that
5321 happened as part of the Exiscan merge. However, to be on the safe side, I
5322 have made the code more robust (and fixed the comments that describe what
5325 53. Now that there can be additional text after "Completed" in log lines (if
5326 the queue_time_overall log selector is set), a one-byte patch to exigrep
5327 was needed to allow it to recognize "Completed" as not the last thing in
5330 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
5331 patch that reportedly fixes this has been added. I am not expert enough to
5332 create a test for it. This is what the patch creator wrote:
5334 "I found a little strange behaviour of ldap code when working with
5335 Windows 2003 AD Domain, where users was placed in more than one
5336 Organization Units. When I tried to give exim partial DN, the exit code
5337 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
5338 But simultaneously result of request was absolutely normal ldap result,
5339 so I produce this patch..."
5341 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
5342 REFERENCE, so I have modified the code to exclude the patch when that macro
5345 55. Some experimental protocols are using DNS PTR records for new purposes. The
5346 keys for these records are domain names, not reversed IP addresses. The
5347 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5348 leaves it alone. Component reversal etc. now happens only for IP addresses.
5351 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
5353 57. Double the size of the debug message buffer (to 2048) so that more of very
5354 long debug lines gets shown.
5356 58. The exicyclog utility now does better if the number of log files to keep
5357 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
5359 59. Two changes related to the smtp_active_hostname option:
5361 (1) $smtp_active_hostname is now available as a variable.
5362 (2) The default for smtp_banner uses $smtp_active_hostname instead
5363 of $primary_hostname.
5365 60. The host_aton() function is supposed to be passed a string that is known
5366 to be a valid IP address. However, in the case of IPv6 addresses, it was
5367 not checking this. This is a hostage to fortune. Exim now panics and dies
5368 if the condition is not met. A case was found where this could be provoked
5369 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5370 components; fortuitously, this particular loophole had already been fixed
5371 by change 4.50/55 above.
5373 If there are any other similar loopholes, the new check in host_aton()
5374 itself should stop them being exploited. The report I received stated that
5375 data on the command line could provoke the exploit when Exim was running as
5376 exim, but did not say which command line option was involved. All I could
5377 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5378 running as the user.
5381 61. There was a buffer overflow vulnerability in the SPA authentication code
5382 (which came originally from the Samba project). I have added a test to the
5383 spa_base64_to_bits() function which I hope fixes it.
5386 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
5387 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
5389 63. The daemon start-up calls getloadavg() while still root for those OS that
5390 need the first call to be done as root, but it missed one case: when
5391 deliver_queue_load_max is set with deliver_drop_privilege. This is
5392 necessary for the benefit of the queue runner, because there is no re-exec
5393 when deliver_drop_privilege is set.
5395 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
5396 This has been fixed.
5398 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5399 was in use, was not putting the data itself into the right store pool;
5400 consequently, it could be overwritten for a subsequent message in the same
5401 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5404 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
5406 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
5407 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
5408 the function were treating the return as a boolean value, which happened to
5409 work because 0=false and not-0=true, but is not correct code.
5411 68. The host_aton() function was not handling scoped IPv6 addresses (those
5412 with, for example, "%eth0" on the end) correctly.
5414 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
5415 negated items (that is, ~something) in unsigned ints. Some compilers
5416 apparently mutter when there is no cast.
5418 70. If an address verification called from an ACL failed, and did not produce a
5419 user-specific message (i.e. there was only a "system" message), nothing was
5420 put in $acl_verify_message. In this situation, it now puts the system
5423 71. Change 4.23/11 added synchronization checking at the start of an SMTP
5424 session; change 4.31/43 added the unwanted input to the log line - except
5425 that it did not do this in the start of session case. It now does.
5427 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
5428 This is wrong and can cause the other end to generate a synchronization
5429 error if it is another Exim or anything else that does the synchronization
5430 check. A QUIT command is no longer sent after a timeout.
5432 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
5433 during host lookups.
5435 74. The defer_ok option of callout verification was not working if it was used
5436 when verifying addresses in header lines, that is, for this case:
5438 verify = header_sender/callout=defer_ok
5440 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
5441 those file descriptors could be used for SMTP connections. If anything
5442 wrote to stderr (the example that came up was "warn" in embedded Perl), it
5443 could be sent to the SMTP client, causing chaos. The daemon now opens
5444 stdin, stdout, and stderr to /dev/null when it puts itself into the
5447 76. Arrange for output from Perl's "warn" command to be written to Exim's main
5448 log by default. The user can override this with suitable Perl magic.
5450 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
5451 the log message when discard triggers, was not working for the DATA ACL or
5452 for the non-SMTP ACL.
5454 78. Error message wording change in sieve.c.
5456 79. If smtp_accept_max_per_host was set, the number of connections could be
5457 restricted to fewer than expected, because the daemon was trying to set up
5458 a new connection before checking whether the processes handling previous
5459 connections had finished. The check for completed processes is now done
5460 earlier. On busy systems, this bug wouldn't be noticed because something
5461 else would have woken the daemon, and it would have reaped the completed
5464 80. If a message was submitted locally by a user whose login name contained one
5465 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
5466 It caused a spool format error. I have fixed the spool reading code. A
5467 related problem was that the "from" clause in the Received: line became
5468 illegal because of the space(s). It is now covered by ${quote_local_part.
5470 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
5473 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
5475 83. Patch from Sieve maintainer to fix unterminated string problem in
5476 "vacation" handling.
5478 84. Some minor changes to the Linux configuration files to help with other
5479 OS variants using glibc.
5481 85. One more patch for Sieve to update vacation handling to latest spec.
5484 ----------------------------------------------------
5485 See the note above about the 4.44 and 4.50 releases.
5486 ----------------------------------------------------
5492 1. Change 4.43/35 introduced a bug that caused file counts to be
5493 incorrectly computed when quota_filecount was set in an appendfile
5496 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5497 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5500 3. Give more explanation in the error message when the command for a transport
5501 filter fails to execute.
5503 4. There are several places where Exim runs a non-Exim command in a
5504 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5505 was being done only for the command run by the queryprogram router. It is
5506 now done for all such subprocesses. The other cases are: ${run, transport
5507 filters, and the commands run by the lmtp and pipe transports.
5509 5. Some older OS have a limit of 256 on the maximum number of file
5510 descriptors. Exim was using setrlimit() to set 1000 as a large value
5511 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5512 systems. I've change it so that if it can't get 1000, it tries for 256.
5514 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5515 was an oversight, and furthermore, ever since the addition of extra
5516 controls (e.g. 4.43/32), the checks on when to allow different forms of
5517 "control" were broken. There should now be diagnostics for all cases when a
5518 control that does not make sense is encountered.
5520 7. $recipients is now available in the predata ACL (oversight).
5522 8. Tidy the search cache before the fork to do a delivery from a message
5523 received from the command line. Otherwise the child will trigger a lookup
5524 failure and thereby defer the delivery if it tries to use (for example) a
5525 cached ldap connection that the parent has called unbind on.
5527 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5528 of $address_data from the recipient verification was clobbered by the
5529 sender verification.
5531 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5532 was its contents. (It was OK if the option was not defined at all.)
5534 11. A "Completed" log line is now written for messages that are removed from
5535 the spool by the -Mrm option.
5537 12. $host_address is now set to the target address during the checking of
5538 ignore_target_hosts.
5540 13. When checking ignore_target_hosts for an ipliteral router, no host name was
5541 being passed; this would have caused $sender_host_name to have been used if
5542 matching the list had actually called for a host name (not very likely,
5543 since this list is usually IP addresses). A host name is now passed as
5546 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5547 code that specifies a non-restarting handler (typically sigaction() in
5548 modern systems) in an attempt to fix a rare and obscure crash bug.
5550 15. Narrowed the window for a race in the daemon that could cause it to ignore
5551 SIGCHLD signals. This is not a major problem, because they are used only to
5552 wake it up if nothing else does.
5554 16. A malformed maildirsize file could cause Exim to calculate negative values
5555 for the mailbox size or file count. Odd effects could occur as a result.
5556 The maildirsize information is now recalculated if the size or filecount
5559 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5560 support for a long time. Removed HAVE_SYS_VFS_H.
5562 18. Updated exipick to current release from John Jetmore.
5564 19. Allow an empty sender to be matched against a lookup in an address list.
5565 Previously the only cases considered were a regular expression, or an
5568 20. Exim went into a mad DNS lookup loop when doing a callout where the
5569 host was specified on the transport, if the DNS lookup yielded more than
5572 21. The RFC2047 encoding function was originally intended for short strings
5573 such as real names; it was not keeping to the 75-character limit for
5574 encoded words that the RFC imposes. It now respects the limit, and
5575 generates multiple encoded words if necessary. To be on the safe side, I
5576 have increased the buffer size for the ${rfc2047: expansion operator from
5579 22. Failure to deliver a bounce message always caused it to be frozen, even if
5580 there was an errors_to setting on the router. The errors_to setting is now
5583 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5584 canonical form (fully expanded) before being placed in
5585 $sender_host_address.
5587 24. Updated eximstats to version 1.33
5589 25. Include certificate and key file names in error message when GnuTLS fails
5590 to set them up, because the GnuTLS error message doesn't include the name
5591 of the failing file when there is a problem reading it.
5593 26. Expand error message when OpenSSL has problems setting up cert/key files.
5596 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5597 (this can affect the format of dates).
5599 28. exim_tidydb, when checking for the continued existence of a message for
5600 which it has found a message-specific retry record, was not finding
5601 messages that were in split spool directories. Consequently, it was
5602 deleting retry records that should have stayed in existence.
5604 29. eximstats updated to version 1.35
5605 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5606 1.35 - bugfix such that pie charts by volume are generated correctly
5608 30. The SPA authentication driver was not abandoning authentication and moving
5609 on to the next authenticator when an expansion was forced to fail,
5610 contradicting the general specification for all authenticators. Instead it
5611 was generating a temporary error. It now behaves as specified.
5613 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5614 (the order specifies the preference for clients). The order is now AES256,
5615 AES128, 3DES, ARCFOUR128.
5617 31. Small patch to Sieve code - explicitly set From: when generating an
5620 32. Exim crashed if a remote delivery caused a very long error message to be
5621 recorded - for instance if somebody sent an entire SpamAssassin report back
5622 as a large number of 550 error lines. This bug was coincidentally fixed by
5623 increasing the size of one of Exim's internal buffers (big_buffer) that
5624 happened as part of the Exiscan merge. However, to be on the safe side, I
5625 have made the code more robust (and fixed the comments that describe what
5628 33. Some experimental protocols are using DNS PTR records for new purposes. The
5629 keys for these records are domain names, not reversed IP addresses. The
5630 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5631 leaves it alone. Component reversal etc. now happens only for IP addresses.
5634 34. The host_aton() function is supposed to be passed a string that is known
5635 to be a valid IP address. However, in the case of IPv6 addresses, it was
5636 not checking this. This is a hostage to fortune. Exim now panics and dies
5637 if the condition is not met. A case was found where this could be provoked
5638 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5639 components; fortuitously, this particular loophole had already been fixed
5640 by change 4.50/55 or 4.44/33 above.
5642 If there are any other similar loopholes, the new check in host_aton()
5643 itself should stop them being exploited. The report I received stated that
5644 data on the command line could provoke the exploit when Exim was running as
5645 exim, but did not say which command line option was involved. All I could
5646 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5647 running as the user.
5650 35. There was a buffer overflow vulnerability in the SPA authentication code
5651 (which came originally from the Samba project). I have added a test to the
5652 spa_base64_to_bits() function which I hope fixes it.
5655 36. The daemon start-up calls getloadavg() while still root for those OS that
5656 need the first call to be done as root, but it missed one case: when
5657 deliver_queue_load_max is set with deliver_drop_privilege. This is
5658 necessary for the benefit of the queue runner, because there is no re-exec
5659 when deliver_drop_privilege is set.
5661 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5662 was in use, was not putting the data itself into the right store pool;
5663 consequently, it could be overwritten for a subsequent message in the same
5664 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5667 38. Sometimes the final signoff response after QUIT could fail to get
5668 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5669 < 0 before doing a fflush(). This bug looks as though it goes back to the
5670 introduction of TLS in release 3.20, but "sometimes" must have been rare
5671 because the tests only now provoked it.
5677 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5678 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5679 Now it returns an integer. A number of calls were still expecting a T/F
5680 return. Fortuitously, in all cases, the tests worked in OK situations,
5681 which is the norm. However, things would have gone wrong on any write
5682 failures on the smtp file descriptor. This function is used when sending
5683 messages over SMTP and also when doing verify callouts.
5685 2. When Exim is called to do synchronous delivery of a locally submitted
5686 message (the -odf or -odi options), it no longer closes stderr before doing
5689 3. Implemented the mua_wrapper option.
5691 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5693 5. Implemented the functions header_remove(), header_testname(),
5694 header_add_at_position(), and receive_remove_recipient(), and exported them
5697 6. If an ACL "warn" statement specified the addition of headers, Exim already
5698 inserted X-ACL-Warn: at the start if there was no header name. However, it
5699 was not making this test for the second and subsequent header lines if
5700 there were newlines in the string. This meant that an invalid header could
5701 be inserted if Exim was badly configured.
5703 7. Allow an ACL "warn" statement to add header lines at the start or after all
5704 the Received: headers, as well as at the end.
5706 8. Added the rcpt_4xx retry error code.
5708 9. Added postmaster_mailfrom=xxx to callout verification option.
5710 10. Added mailfrom=xxxx to the callout verification option, for verify=
5713 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5714 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5716 12. Inserted some casts to stop certain compilers warning when using pointer
5717 differences as field lengths or precisions in printf-type calls (mostly
5718 affecting debugging statements).
5720 13. Added optional readline() support for -be (dynamically loaded).
5722 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5723 same clock tick as a message's arrival, so that its received time was the
5724 same as the "first fail" time on the retry record, and that message
5725 remained on the queue past the ultimate address timeout, every queue runner
5726 would try a delivery (because it was past the ultimate address timeout) but
5727 after another failure, the ultimate address timeout, which should have then
5728 bounced the address, did not kick in. This was a "< instead of <=" error;
5729 in most cases the first failure would have been in the next clock tick
5730 after the received time, and all would be well.
5732 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5733 being recognized when the domain list was tested by the match_domain
5734 condition in an expansion string.
5736 16. Added the ${str2b64: operator.
5738 17. Exim was always calling setrlimit() to set a large limit for the number of
5739 processes, without checking whether the existing limit was already
5740 adequate. (It did check for the limit on file descriptors.) Furthermore,
5741 errors from getrlimit() and setrlimit() were being ignored. Now they are
5742 logged to the main and panic logs, but Exim does carry on, to try to do its
5743 job under whatever limits there are.
5745 18. Imported PCRE 5.0.
5747 19. Trivial typo in log message " temporarily refused connection" (the leading
5750 20. If the log selector return_path_on_delivery was set and an address was
5751 redirected to /dev/null, the delivery process crashed because it assumed
5752 that a return path would always be set for a "successful" delivery. In this
5753 case, the whole delivery is bypassed as an optimization, and therefore no
5756 21. Internal re-arrangement: the function for sending a challenge and reading
5757 a response while authentication was assuming a zero-terminated challenge
5758 string. It's now changed to take a pointer and a length, to allow for
5759 binary data in such strings.
5761 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5763 23. Exim was not respecting finduser_retries when seeking the login of the
5764 uid under which it was called; it was always trying 10 times. (The default
5765 setting of finduser_retries is zero.) Also, it was sleeping after the final
5766 failure, which is pointless.
5768 24. Implemented tls_on_connect_ports.
5770 25. Implemented acl_smtp_predata.
5772 26. If the domain in control=submission is set empty, Exim assumes that the
5773 authenticated id is a complete email address when it generates From: or
5774 Sender: header lines.
5776 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5777 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5778 chown and chgrp in /bin and hostname in /usr/bin.
5780 28. Exim was keeping the "process log" file open after each use, just as it
5781 does for the main log. This opens the possibility of it remaining open for
5782 long periods when the USR1 signal hits a daemon. Occasional processlog
5783 errors were reported, that could have been caused by this. Anyway, it seems
5784 much more sensible not to leave this file open at all, so that is what now
5787 29. The long-running daemon process does not normally write to the log once it
5788 has entered its main loop, and it closes the log before doing so. This is
5789 so that log files can straightforwardly be renamed and moved. However,
5790 there are a couple of unusual error situations where the daemon does write
5791 log entries, and I had neglected to close the log afterwards.
5793 30. The text of an SMTP error response that was received during a remote
5794 delivery was being truncated at 512 bytes. This is too short for some of
5795 the long messages that one sometimes sees. I've increased the limit to
5798 31. It is now possible to make retry rules that apply only when a message has a
5799 specific sender, in particular, an empty sender.
5801 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5802 it possible to be selective about when SMTP synchronization is enforced.
5804 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5806 32. Implemented hosts_connection_nolog.
5808 33. Added an ACL for QUIT.
5810 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5813 35. Added mailbox_size and mailbox_filecount to appendfile.
5815 36. Added control = no_multiline_responses to ACLs.
5817 37. There was a bug in the logic of the code that waits for the clock to tick
5818 in the case where the clock went backwards by a substantial amount such
5819 that the microsecond fraction of "now" was more than the microsecond
5820 fraction of "then" (but the whole seconds number was less).
5822 38. Added support for the libradius Radius client library this is found on
5823 FreeBSD (previously only the radiusclient library was supported).
5829 1. When certain lookups returned multiple values in the form name=value, the
5830 quoting of the values was not always being done properly. Specifically:
5831 (a) If the value started with a double quote, but contained no whitespace,
5833 (b) If the value contained whitespace other than a space character (i.e.
5834 tabs or newlines or carriage returns) it was not quoted.
5835 This fix has been applied to the mysql and pgsql lookups by writing a
5836 separate quoting function and calling it from the lookup code. The fix
5837 should probably also be applied to nisplus, ibase and oracle lookups, but
5838 since I cannot test any of those, I have not disturbed their existing code.
5840 2. A hit in the callout cache for a specific address caused a log line with no
5841 reason for rejecting RCPT. Now it says "Previous (cached) callout
5842 verification failure".
5844 3. There was an off-by-one bug in the queryprogram router. An over-long
5845 return line was truncated at 256 instead of 255 characters, thereby
5846 overflowing its buffer with the terminating zero. As well as fixing this, I
5847 have increased the buffer size to 1024 (and made a note to document this).
5849 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5850 when Exim is waiting for an SMTP response from a remote server, Exim
5851 restarts its select() call on the socket, thereby resetting its timeout.
5852 This is not a problem when such interrupts are rare. Somebody set up a cron
5853 job to run exiwhat every 2 minutes, which is less than the normal select()
5854 timeout (5 or 10 minutes). This meant that the select() timeout never
5855 kicked in because it was always reset. I have fixed this by comparing the
5856 time when an interrupt arrives with the time at the start of the first call
5857 to select(). If more time than the timeout has elapsed, the interrupt is
5858 treated as a timeout.
5860 5. Some internal re-factoring in preparation for the addition of Sieve
5861 extensions (by MH). In particular, the "personal" test is moved to a
5862 separate function, and given an option for scanning Cc: and Bcc: (which is
5863 not set for Exim filters).
5865 6. When Exim created an email address using the login of the caller as the
5866 local part (e.g. when creating a From: or Sender: header line), it was not
5867 quoting the local part when it contained special characters such as @.
5869 7. Installed new OpenBSD configuration files.
5871 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5872 try to make them clearer.
5874 9. Callout options, other than the timeout value, were being ignored when
5875 verifying sender addresses in header lines. For example, when using
5877 verify = header_sender/callout=no_cache
5879 the cache was (incorrectly) being used.
5881 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5882 only the Cygwin environment.
5884 11. When return_path_on_delivery was set as a log selector, if different remote
5885 addresses in the same message used different return paths and parallel
5886 remote delivery occurred, the wrong values would sometimes be logged.
5887 (Whenever a remote delivery process finished, the return path value from
5888 the most recently started remote delivery process was logged.)
5890 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5891 lines when AUTH and/or TLS are in use. This is the "received protocol"
5892 field. Exim used to use "asmtp" for authenticated SMTP, without any
5893 indication (in the protocol name) for TLS use. Now it follows the RFC and
5894 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5895 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5896 names appear in log lines as well as in Received: header lines.
5898 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5899 extensions, and comparison tests, and to fix some bugs.
5901 14. Changes to the "personal" filter test:
5903 (1) The test was buggy in that it was just doing the equivalent of
5904 "contains" tests on header lines. For example, if a user's address was
5905 anne@some.where, the "personal" test would incorrectly be true for
5907 To: susanne@some.where
5909 This test is now done by extracting each address from the header in turn,
5910 and checking the entire address. Other tests that are part of "personal"
5911 are now done using regular expressions (for example, to check local parts
5912 of addresses in From: header lines).
5914 (2) The list of non-personal local parts in From: addresses has been
5915 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5916 taken from the Sieve specification recommendations.
5918 (3) If the message contains any header line starting with "List-" it is
5919 treated as non-personal.
5921 (4) The test for "circular" in the Subject: header line has been removed
5922 because it now seems ill-conceived.
5924 15. Minor typos in src/EDITME comments corrected.
5926 16. Installed latest exipick from John Jetmore.
5928 17. If headers_add on a router specified a text string that was too long for
5929 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5930 of string_sprintf() is now avoided.
5932 18. $message_body_size was not set (it was always zero) when running the DATA
5933 ACL and the local_scan() function.
5935 19. For the "mail" command in an Exim filter, no default was being set for
5936 the once_repeat time, causing a random time value to be used if "once" was
5937 specified. (If the value happened to be <= 0, no repeat happened.) The
5938 default is now 0s, meaning "never repeat". The "vacation" command was OK
5939 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5940 (I found it when inspecting the code).
5942 20. There is now an overall timeout for performing a callout verification. It
5943 defaults to 4 times the callout timeout, which applies to individual SMTP
5944 commands during the callout. The overall timeout applies when there is more
5945 than one host that can be tried. The timeout is checked before trying the
5946 next host. This prevents very long delays if there are a large number of
5947 hosts and all are timing out (e.g. when the network connections are timing
5948 out). The value of the overall timeout can be changed by specifying an
5949 additional sub-option for "callout", called "maxwait". For example:
5951 verify = sender/callout=5s,maxwait=20s
5953 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5954 to the end before writing, but this should make it even safer).
5956 22. Exim was forgetting that it had advertised PIPELINING for the second and
5957 subsequent messages on an SMTP connection. It was also not resetting its
5958 memory on STARTTLS and an internal HELO.
5960 23. When Exim logs an SMTP synchronization error within a session, it now
5961 records whether PIPELINING has been advertised or not.
5963 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5964 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5965 rather than long int.
5967 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5969 26. Added the never_mail option to autoreply.
5975 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5976 crash if the getsockname() call failed; this can happen if a connection is
5977 closed very soon after it is established. The problem was simply in the
5978 order in which certain operations were done, causing Exim to try to write
5979 to the SMTP stream before it had set up the file descriptor. The bug has
5980 been fixed by making things happen in the correct order.
5986 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5987 before the connection was closed, thus losing the rejection response.
5989 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5990 some early Solaris releases, but causes trouble in current releases where
5991 socklen_t is defined.
5993 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5996 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5999 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
6000 that releasing the top of it at the end releases what was used for sub-
6001 expansions (unless the block got too big). However, discard this block if
6002 the first thing is a variable or header, so that we can use its block when
6003 it is dynamic (useful for very large $message_headers, for example).
6005 6. Lookups now cache *every* query, not just the most recent. A new, separate
6006 store pool is used for this. It can be recovered when all lookup caches are
6007 flushed. Lookups now release memory at the end of their result strings.
6008 This has involved some general refactoring of the lookup sources.
6010 7. Some code has been added to the store_xxx() functions to reduce the amount
6011 of flapping under certain conditions.
6013 8. log_incoming_interface used to affect only the <= reception log lines. Now
6014 it causes the local interface and port to be added to several more SMTP log
6015 lines, for example "SMTP connection from", and rejection lines.
6017 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
6019 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
6021 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
6022 could be overwritten at the end of the current message (or the start of a
6023 new message if it was set in a HELO ACL). The value is now preserved for
6024 the duration of the SMTP connection.
6026 12. If a transport had a headers_rewrite setting, and a matching header line
6027 contained an unqualified address, that address was qualified, even if it
6028 did not match any rewriting rules. The underlying bug was that the values
6029 of the flags that permit the existence of unqualified sender and recipient
6030 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
6031 non-local messages, and by -bnq for local messages) were not being
6032 preserved with the message after it was received.
6034 13. When Exim was logging an SMTP synchronization error, it could sometimes log
6035 "next input=" as part of the text comprising the host identity instead of
6036 the correct text. The code was using the same buffer for two different
6037 strings. However, depending on which order the printing function evaluated
6038 its arguments, the bug did not always show up. Under Linux, for example, my
6039 test suite worked just fine.
6041 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
6042 doesn't work with some older versions of Perl. It has been changed to "my",
6043 which in any case is probably the better facility to use.
6045 15. A really picky compiler found some instances of statements for creating
6046 error messages that either had too many or two few arguments for the format
6049 16. The size of the buffer for calls to the DNS resolver has been increased
6050 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
6051 for addresses that have a lot of PTR records. This alleviates a problem; it
6052 does not fully solve it.
6054 17. A dnsdb lookup for PTR records that receives more data than will fit in the
6055 buffer now truncates the list and logs the incident, which is the same
6056 action as happens when Exim is looking up a host name and its aliases.
6057 Previously in this situation something unpredictable would happen;
6058 sometimes it was "internal error: store_reset failed".
6060 18. If a server dropped the connection unexpectedly when an Exim client was
6061 using GnuTLS and trying to read a response, the client delivery process
6062 crashed while trying to generate an error log message.
6064 19. If a "warn" verb in an ACL added multiple headers to a message in a single
6065 string, for example:
6067 warn message = H1: something\nH2: something
6069 the text was added as a single header line from Exim's point of view
6070 though it ended up OK in the delivered message. However, searching for the
6071 second and subsequent header lines using $h_h2: did not work. This has been
6072 fixed. Similarly, if a system filter added multiple headers in this way,
6073 the routers could not see them.
6075 20. Expanded the error message when iplsearch is called with an invalid key to
6076 suggest using net-iplsearch in a host list.
6078 21. When running tests using -bh, any delays imposed by "delay" modifiers in
6079 ACLs are no longer actually imposed (and a message to that effect is
6082 22. If a "gecos" field in a passwd entry contained escaped characters, in
6083 particular, if it contained a \" sequence, Exim got it wrong when building
6084 a From: or a Sender: header from that name. A second bug also caused
6085 incorrect handling when an unquoted " was present following a character
6086 that needed quoting.
6088 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
6089 was not being matched caselessly.
6091 24. Arranged for all hyphens in the exim.8 source to be escaped with
6094 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
6095 properly. Recipient callout cache records were still being keyed to include
6096 the sender, even when use_sender was set false. This led to far more
6097 callouts that were necessary. The sender is no longer included in the key
6098 when use_sender is false.
6100 26. Added "control = submission" modifier to ACLs.
6102 27. Added the ${base62d: operator to decode base 62 numbers.
6104 28. dnsdb lookups can now access SRV records.
6106 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
6107 the configuration file.
6109 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
6110 (-v) mode. This makes the output for a verbose queue run more intelligible.
6112 31. Added a use_postmaster feature to recipient callouts.
6114 32. Added the $body_zerocount variable, containing the number of binary zero
6115 bytes in the message body.
6117 33. The time of last modification of the "new" subdirectory is now used as the
6118 "mailbox time last read" when there is a quota error for a maildir
6121 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
6123 35. Added +ignore_unknown as a special item in host lists.
6125 36. Code for decoding IPv6 addresses in host lists is now included, even if
6126 IPv6 support is not being compiled. This fixes a bug in which an IPv6
6127 address was recognized as an IP address, but was then not correctly decoded
6128 into binary, causing unexpected and incorrect effects when compared with
6135 1. Very minor rewording of debugging text in manualroute to say "list of
6136 hosts" instead of "hostlist".
6138 2. If verify=header_syntax was set, and a header line with an unqualified
6139 address (no domain) and a large number of spaces between the end of the
6140 name and the colon was received, the reception process suffered a buffer
6141 overflow, and (when I tested it) crashed. This was caused by some obsolete
6142 code that should have been removed. The fix is to remove it!
6144 3. When running in the test harness, delay a bit after writing a bounce
6145 message to get a bit more predictability in the log output.
6147 4. Added a call to search_tidyup() just before forking a reception process. In
6148 theory, someone could use a lookup in the expansion of smtp_accept_max_
6149 per_host which, without the tidyup, could leave open a database connection.
6151 5. Added the variables $recipient_data and $sender_data which get set from a
6152 lookup success in an ACL "recipients" or "senders" condition, or a router
6153 "senders" option, similar to $domain_data and $local_part_data.
6155 6. Moved the writing of debug_print from before to after the "senders" test
6158 7. Change 4.31/66 (moving the time when the Received: is generated) caused
6159 problems for message scanning, either using a data ACL, or using
6160 local_scan() because the Received: header was not generated till after they
6161 were called (in order to set the time as the time of reception completion).
6162 I have revised the way this works. The header is now generated after the
6163 body is received, but before the ACL or local_scan() are called. After they
6164 are run, the timestamp in the header is updated.
6170 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
6171 before starting a queue runner without re-exec. This happened only when
6172 deliver_drop_privilege was set or when the Exim user was set to root. The
6173 effect of the bug was that timeouts during subsequent deliveries caused
6174 crashes instead of being properly handled. The handler is now left at its
6175 default (and expected) setting.
6177 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
6178 message, again when deliver_drop_privilege is set or Exim is run as root.
6179 The bug described in (1) was not present in this case, but the tidying up
6180 of the other signals was missing. I have made the two cases consistent.
6182 3. The ignore_target_hosts setting on a manualroute router was being ignored
6183 for hosts that were looked up using the /MX notation.
6185 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
6188 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
6189 operated on the sender address. After changing the $sender_address to <>
6190 for the sender address verify, Exim was re-instated it as the original
6191 (before rewriting) address, but remembering that it had rewritten it, so it
6192 wasn't rewriting it again. This bug also had the effect of breaking the
6193 sender address verification caching when the sender address was rewritten.
6195 6. The ignore_target_hosts option was being ignored by the ipliteral router.
6196 This has been changed so that if the ip literal address matches
6197 ignore_target_hosts, the router declines.
6199 7. Added expansion conditions match_domain, match_address, and match_local_
6200 part (NOT match_host).
6202 8. The placeholder for the Received: header didn't have a length field set.
6204 9. Added code to Exim itself and to exim_lock to test for a specific race
6205 condition that could lead to file corruption when using MBX delivery. The
6206 issue is with the lockfile that is created in /tmp. If this file is removed
6207 after a process has opened it but before that process has acquired a lock,
6208 there is the potential for a second process to recreate the file and also
6209 acquire a lock. This could lead to two Exim processes writing to the file
6210 at the same time. The added code performs the same test as UW imapd; it
6211 checks after acquiring the lock that its file descriptor still refers to
6212 the same named file.
6214 10. The buffer for building added header lines was of fixed size, 8192 bytes.
6215 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
6218 11. Added the smtp_active_hostname option. If used, this will typically be made
6219 to depend on the incoming interface address. Because $interface_address is
6220 not set up until the daemon has forked a reception process, error responses
6221 that can happen earlier (such as "too many connections") no longer contain
6224 12. If an expansion in a condition on a "warn" statement fails because a lookup
6225 defers, the "warn" statement is abandoned, and the next ACL statement is
6226 processed. Previously this caused the whole ACL to be aborted.
6228 13. Added the iplsearch lookup type.
6230 14. Added ident_timeout as a log selector.
6232 15. Added tls_certificate_verified as a log selector.
6234 16. Added a global option tls_require_ciphers (compare the smtp transport
6235 option of the same name). This controls incoming TLS connections.
6237 17. I finally figured out how to make tls_require_ciphers do a similar thing
6238 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
6239 before starting the TLS session.
6241 18. Tabs are now shown as \t in -bP output.
6243 19. If the log selector return_path_on_delivery was set, Exim crashed when
6244 bouncing a message because it had too many Received: header lines.
6246 20. If two routers both had headers_remove settings, and the first one included
6247 a superfluous trailing colon, the final name in the first list and the
6248 first name in the second list were incorrectly joined into one item (with a
6249 colon in the middle).
6255 1. Added -C and -D options to the exinext utility, mainly to make it easier
6256 to include in the automated testing, but these could be helpful when
6257 multiple configurations are in use.
6259 2. The exinext utility was not formatting the output nicely when there was
6260 an alternate port involved in the retry record key, nor when there was a
6261 message id as well (for retries that were specific to a specific message
6262 and a specific host). It was also confused by IPv6 addresses, because of
6263 the additional colons they contain. I have fixed the IPv4 problem, and
6264 patched it up to do a reasonable job for IPv6.
6266 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
6267 delivery, the log line now contains "pipelined" if PIPELINING was used.
6269 4. An SMTP transport process used to panic and die if the bind() call to set
6270 an explicit outgoing interface failed. This has been changed; it is now
6271 treated in the same way as a connect() failure.
6273 5. A reference to $sender_host_name in the part of a conditional expansion
6274 that was being skipped was still causing a DNS lookup. This no longer
6277 6. The def: expansion condition was not recognizing references to header lines
6278 that used bh_ and bheader_.
6280 7. Added the _cache feature to named lists.
6282 8. The code for checking quota_filecount in the appendfile transport was
6283 allowing one more file than it should have been.
6285 9. For compatibility with Sendmail, the command line option
6293 and sets the incoming protocol and host name (for trusted callers). The
6294 host name and its colon can be omitted when only the protocol is to be set.
6295 Note the Exim already has two private options, -pd and -ps, that refer to
6296 embedded Perl. It is therefore impossible to set a protocol value of "d" or
6297 "s", but I don't think that's a major issue.
6299 10. A number of refactoring changes to the code, none of which should affect
6302 (a) The number of logging options was getting close to filling up the
6303 32-bit word that was used as a bit map. I have split them into two classes:
6304 those that are passed in the argument to log_write(), and those that are
6305 only ever tested independently outside of that function. These are now in
6306 separate 32-bit words, so there is plenty of room for expansion again.
6307 There is no change in the user interface or the logging behaviour.
6309 (b) When building, for example, log lines, the code previously used a
6310 macro that called string_cat() twice, in order to add two strings. This is
6311 not really sufficiently general. Furthermore, there was one instance where
6312 it was actually wrong because one of the argument was used twice, and in
6313 one call a function was used. (As it happened, calling the function twice
6314 did not affect the overall behaviour.) The macro has been replaced by a
6315 function that can join an arbitrary number of extra strings onto a growing
6318 (c) The code for expansion conditions now uses a table and a binary chop
6319 instead of a serial search (which was left over from when there were very
6320 few conditions). Also, it now recognizes conditions like "pam" even when
6321 the relevant support is not compiled in: a suitably worded error message is
6322 given if an attempt is made to use such a condition.
6324 11. Added ${time_interval:xxxxx}.
6326 12. A bug was causing one of the ddress fields not to be passed back correctly
6327 from remote delivery subprocesses. The field in question was not being
6328 subsequently used, so this caused to problems in practice.
6330 13. Added new log selectors queue_time and deliver_time.
6332 14. Might have fixed a bug in maildirsizefile handling that threw up
6333 "unexpected character" debug warnings, and recalculated the data
6334 unnecessarily. In any case, I expanded the warning message to give more
6337 15. Added the message "Restricted characters in address" to the statements in
6338 the default ACL that block characters like @ and % in local parts.
6340 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
6341 Three changes have been made:
6343 (a) There was a serious bug; a negative response to MAIL caused the whole
6344 recipient domain to be cached as invalid, thereby blocking all messages
6345 to all local parts at the same domain, from all senders. This bug has
6346 been fixed. The domain is no longer cached after a negative response to
6347 MAIL if the sender used is not empty.
6349 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
6352 (c) A new callout option, "use_sender" has been added for people who want
6353 the modified behaviour.
6359 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
6362 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
6363 indeed breaks things for older releases.
6365 3. Added additional logging to the case where there is a problem reading data
6366 from a filter that is running in a subprocess using a pipe, in order to
6367 try to track down a specific problem.
6369 4. Testing facility fudge: when running in the test harness and attempting
6370 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
6371 getting "No route to host". Convert this to a timeout.
6373 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
6376 6. Some OS don't have socklen_t but use size_t instead. This affects the
6377 fifth argument of getsockopt() amongst other things. This is now
6378 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
6379 can be set for individual OS. I have set it for SunOS5, OSF1, and
6380 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
6381 some earlier ones do not.
6383 7. Change 4.30/15 was not doing the test caselessly.
6385 8. The standard form for an IPv6 address literal was being rejected by address
6386 parsing in, for example, MAIL and RCPT commands. An example of this kind of
6387 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6388 this, as well as the form without the "IPv6" on the front (but only when
6389 address literals are enabled, of course).
6391 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
6393 10. Exim crashed if a message with an empty sender address specified by -f
6394 encountered a router with an errors_to setting. This could be provoked only
6395 by a command such as
6399 where an empty string was supplied; "<>" did not hit this bug.
6401 11. Installed PCRE release 4.5.
6403 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
6404 remained set. It is now erased.
6406 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
6407 times from message ids (which are base 36 rather than the normal 62).
6409 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
6410 were being counted as actual protocol errors, and logged if the log
6411 selector +smtp_protocol_error was set. One cannot be perfect in this test,
6412 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
6413 and DATA following a set of rejected RCPTs do not count as protocol errors.
6414 In other words, Exim assumes they were pipelined, though this may not
6415 actually be the case. Of course, in all cases the client gets an
6416 appropriate error code.
6418 15. If a lookup fails in an ACL condition, a message about the failure may
6419 be available; it is used if testing the ACL cannot continue, because most
6420 such messages specify what the cause of the deferral is. However, some
6421 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
6422 that caused an old message to be retained and used if a later statement
6423 caused a defer, replacing the real cause of the deferral.
6425 16. If an IP address had so many PTR records that the DNS lookup buffer
6426 was not large enough to hold them, Exim could crash while trying to process
6427 the truncated data. It now detects and logs this case.
6429 17. Further to 4.21/58, another change has been made: if (and only if) the
6430 first line of a message (the first header line) ends with CRLF, a bare LF
6431 in a subsequent header line has a space inserted after it, so as not to
6432 terminate the header.
6434 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
6435 unnecessarily, used atoi() instead of strtol(), and didn't check the
6436 termination when getting file sizes from file names by regex.
6438 19. Completely re-implemented the support for maildirsize files, in the light
6439 of a number of problems with the previous contributed implementation
6440 (4.30/29). In particular:
6442 . If the quota is zero, the maildirsize file is maintained, but no quota is
6445 . If the maildir directory does not exist, it is created before any attempt
6446 to write a maildirsize file.
6448 . The quota value in the file is just a cache; if the quota is changed in
6449 the transport, the new value overrides.
6451 . A regular expression is available for excluding directories from the
6454 20. The autoreply transport checks the characters in options that define the
6455 message's headers; it allows continued headers, but it was checking with
6456 isspace() after an embedded newline instead of explicitly looking for a
6459 21. If all the "regular" hosts to which an address was routed had passed their
6460 expiry times, and had not reached their retry times, the address was
6461 bounced, even if fallback hosts were defined. Now Exim should go on to try
6464 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
6465 equivalent code in the SMTP transport. Some hosts send humungous responses
6466 to HELO/EHLO, more than 1024 it seems.
6468 23. Refactoring: code in filter.c used (void *) for "any old type" but this
6469 gives compiler warnings in some environments. I've now done it "properly",
6472 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
6473 (because of problems with the built-in one) was declared to return uschar *
6474 instead of char *, causing compiler failure.
6476 25. Fixed a file descriptor leak when processing alias/forward files.
6478 26. Fixed a minor format string issue in dbfn.c.
6480 27. Typo in exim.c: ("dmbnz" for "dbmnz").
6482 28. If a filter file refered to $h_xxx or $message_headers, and the headers
6483 contained RFC 2047 "words", Exim's memory could, under certain conditions,
6486 29. When a sender address is verified, it is cached, to save repeating the test
6487 when there is more than one recipient in a message. However, when the
6488 verification involves a callout, it is possible for different callout
6489 options to be set for different recipients. It is too complicated to keep
6490 track of this in the cache, so now Exim always runs a verification when a
6491 callout is required, relying on the callout cache for the optimization.
6492 The overhead is duplication of the address routing, but this should not be
6495 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
6496 to be verified with callout=postmaster, and the main callout worked but the
6497 postmaster check failed, the verification correctly failed. However, if a
6498 subsequent RCPT command asked for sender verification *without* the
6499 postmaster check, incorrect caching caused this verification also to fail,
6502 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
6503 it was not caching the DNS options (qualify_single, search_parents) that
6504 were used when the lookup failed. A subsequent lookup with different
6505 options therefore always gave the same answer, though there were cases
6506 where it should not have. (Example: a "domains = !$mx_any" option on a
6507 dnslookup router: the "domains" option is always processed without any
6508 widening, but the router might have qualify_single set.) Now Exim uses the
6509 cached value only when the same options are set.
6511 32. Added John Jetmore's "exipick" utility to the distribution.
6513 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
6514 than a timeout (e.g. a certificate is required, and is not provided), an
6515 Exim server now closes the connection immediately. Previously it waited for
6516 the client to close - but if the client is SSL, it seems that they each
6517 wait for each other, leading to a delay before one of them times out.
6519 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
6520 maintained 0.8.x compatibility because I don't think many are using it, and
6521 it is clearly obsolete.
6523 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
6526 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
6527 client certificate was expired. A simple patch fixes this, though I don't
6528 understand the full logic of why the verify callback is called multiple
6531 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
6532 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
6533 which causes problems with some clients (such as the Certicom SSL Plus
6534 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
6535 disables the coutermeasure allowing Eudora to connect."
6537 38. Exim was not checking that a write() to a log file succeeded. This could
6538 lead to Bad Things if a log got too big, in particular if it hit a file
6539 size limit. Exim now panics and dies if it cannot write to a log file, just
6540 as it does if it cannot open a log file.
6542 39. Modified OS/Makefile-Linux so that it now contains
6544 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
6546 The two -D definitions ensure that Exim is compiled with large file
6547 support, which makes it possible to handle log files that are bigger than
6550 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
6551 instance) a domain was checked against a named list that involved a lookup,
6552 causing $domain_data to be set, then another domain was checked against the
6553 same list, then the first domain was re-checked, the value of $domain_data
6554 after the final check could be wrong. In particular, if the second check
6555 failed, it could be set empty. This bug probably also applied to
6558 41. The strip_trailing_dot option was not being applied to the address given
6559 with the -f command-line option.
6561 42. The code for reading a message's header from the spool was incrementing
6562 $received_count, but never initializing it. This meant that the value was
6563 incorrect (doubled) while delivering a message in the same process in which
6564 it was received. In the most common configuration of Exim, this never
6565 happens - a fresh exec is done - but it can happen when
6566 deliver_drop_privilege is set.
6568 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6569 it now includes up to 150 characters of the unexpected data in the log
6572 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6573 and building data strings. The size of both of these buffers was 10 000
6574 bytes - far larger than anybody would *ever* want, thought I. Needless to
6575 say, somebody hit the limit. I have increased the maximum line length to
6576 20 000 and the maximum data length of concatenated lines to 100 000. I have
6577 also fixed two bugs, because there was no checking on these buffers. Tsk,
6578 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6579 buffer is too small.
6581 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6582 lsearch lookups. Now it does.
6584 46. When parsing a route_list item in a manualroute router, a fixed-length
6585 buffer was used for the list of hosts. I made this 1024 bytes long,
6586 thinking that nobody would ever have a list of hosts that long. Wrong.
6587 Somebody had a whole pile of complicated expansion conditions, and the
6588 string was silently truncated, leading to an expansion error. It turns out
6589 that it is easier to change to an unlimited length (owing to other changes
6590 that have happened since this code was originally written) than to build
6591 structure for giving a limitation error. The length of the item that
6592 expands into the list of hosts is now unlimited.
6594 47. The lsearch lookup could not handle data where the length of text line was
6595 more than 4095 characters. Such lines were truncated, leading to shortened
6596 data being returned. It should now handle lines of any length.
6598 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6599 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6602 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6603 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6604 no longer happens in comment lines. A list of replacements is now placed
6605 at the head of all of the source files, except those whose only change is
6606 to replace PERL_COMMAND in the very first #! line.
6608 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6609 messages on the queue, with a bottom-up merge sort, using code contributed
6610 by Michael Haardt. This should make operations like -bp somewhat faster on
6611 large queues. It won't affect queue runners, except when queue_run_in_order
6614 51. Installed eximstats 1.31 in the distribution.
6616 52. Added support for SRV lookups to the dnslookup router.
6618 53. If an ACL referred to $message_body or $message_body_end, the value was not
6619 reset for any messages that followed in the same SMTP session.
6621 54. The store-handling optimization for building very long strings was not
6622 differentiating between the different store pools. I don't think this
6623 actually made any difference in practice, but I've tidied it.
6625 55. While running the routers to verify a sender address, $sender_address
6626 was still set to the sender address. This is wrong, because when routing to
6627 send a bounce to the sender, it would be empty. Therefore, I have changed
6628 it so that, while verifying a sender address, $sender_address is set to <>.
6629 (There is no change to what happens when verifying a recipient address.)
6631 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6632 target A or AAAA records (if not already returned) without resetting the
6633 qualify_single or search_parents options of the DNS resolver. These are
6634 inappropriate in this case because the targets of MX and SRV records must
6635 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6636 target that, when qualified, matched something in the local domain. These
6637 two options are now turned off when doing these lookups.
6639 57. It seems that at least some releases of Reiserfs (which does not have the
6640 concept of a fixed number of inodes) returns zero and not -1 for the
6641 number of available inodes. This interacted badly with check_spool_inodes,
6642 which assumed that -1 was the "no such thing" setting. What I have done is
6643 to check that the total number of inodes is greater than zero before doing
6644 the test of how many are available.
6646 58. When a "warn" ACL statement has a log_message modifier, the message is
6647 remembered, and not repeated. This is to avoid a lot of repetition when a
6648 message has many recipients that cause the same warning to be written.
6649 However, Exim was preserving the list of already written lines for an
6650 entire SMTP session, which doesn't seem right. The memory is now reset if a
6651 new message is started.
6653 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6654 debugging output unless log_rewrite was also set.
6656 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6657 of (int)(handle) when we know that handle contains (void *)(-1).
6659 61. The Exim daemon panic-logs an error return when it closes the incoming
6660 connection. However "connection reset by peer" seems to be common, and
6661 isn't really an error worthy of noting specially, so that particular error
6664 62. When Exim is trying to find all the local interfaces, it used to panic and
6665 die if the ioctl to get the interface flags failed. However, it seems that
6666 on at least one OS (Solaris 9) it is possible to have an interface that is
6667 included in the list of interfaces, but for which you get a failure error
6668 for this call. This happens when the interface is not "plumbed" into a
6669 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6670 failure of the "get flags" call assumes that the interface is down.
6672 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6673 makes life easier for people who are doing arithmetic on fields extracted
6674 from dates, where you often get leading zeros that should not be
6675 interpreted as octal.
6677 64. Added qualify_domain to the redirect router, to override the global
6680 65. If a pathologically long header line contained very many addresses (the
6681 report of this problem mentioned 10 000) and each of them was rewritten,
6682 Exim could use up a very large amount of memory. (It kept on making new
6683 copies of the header line as it rewrote, and never released the old ones.)
6684 At the expense of a bit more processing, the header rewriting function has
6685 been changed so that it no longer eats memory in this way.
6687 66. The generation of the Received: header has been moved from the time that a
6688 message starts to be received, to the time that it finishes. The timestamp
6689 in the Received: header should now be very close to that of the <= log
6690 line. There are two side-effects of this change:
6692 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6693 logged header lines no longer include the local Received: line, because
6694 it has not yet been created. The same applies to a copy of the message
6695 that is returned to a non-SMTP sender when a message is rejected.
6697 (b) When a filter file is tested using -bf, no additional Received: header
6698 is added to the test message. After some thought, I decided that this
6701 This change does not affect the value of $received_for. It is still set
6702 after address rewriting, but before local_scan() is called.
6704 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6706 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6707 gave an unhelpful panic error message, and a defer error. I have managed to
6708 change this behaviour so that it now rejects any supplied certificate,
6709 which seems right, as the list of acceptable certificates is empty.
6711 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6712 gave an unhelpful defer error. I have not managed to make this reject any
6713 supplied certificates, but the error message it gives is "no certificate
6714 supplied", which is not helpful.
6716 70. exigrep's output now also includes lines that are not associated with any
6717 message, but which match the given pattern. Implemented by a patch from
6718 Martin Sluka, which also tidied up the Perl a bit.
6720 71. Recipient callout verification, like sender verification, was using <> in
6721 the MAIL FROM command. This isn't really the right thing, since the actual
6722 sender may affect whether the remote host accepts the recipient or not. I
6723 have changed it to use the actual sender in the callout; this means that
6724 the cache record is now keyed on a recipient/sender pair, not just the
6725 recipient address. There doesn't seem to be a real danger of callout loops,
6726 since a callout by the remote host to check the sender would use <>.
6727 [SEE ABOVE: changed after hitting problems.]
6729 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6730 temporary errors. However, in the case of such a code being given after
6731 the end of a data transmission (i.e. after ".") Exim was failing to write
6732 a retry record for the message. (Yes, there was some broken host that was
6733 actually sending 8xx at this point.)
6735 73. An unknown lookup type in a host list could cause Exim to panic-die when
6736 the list was checked. (An example that provoked this was putting <; in the
6737 middle of a list instead of at the start.) If this happened during a DATA
6738 ACL check, a -D file could be left lying around. This kind of configuration
6739 error no longer causes Exim to die; instead it causes a defer error. The
6740 incident is still logged to the main and panic logs.
6742 74. Buglet left over from Exim 3 conversion. The message "too many messages
6743 in one connection" was written to the rejectlog but not the mainlog, except
6744 when address rewriting (yes!) was being logged.
6746 75. Added write_rejectlog option.
6748 76. When a system filter was run not as root (that is, when system_filter_user
6749 was set), the values of the $n variables were not being returned to the
6750 main process; thus, they were not subsequently available in the $sn
6753 77. Added +return_path_on_delivery log selector.
6755 78. A connection timeout was being treated differently from recipients deferred
6756 when testing hosts_max_try with a message that was older than the host's
6757 retry timeout. (The host should not be counted, thus allowing all hosts to
6758 be tried at least once before bouncing.) This may have been the cause of an
6759 occasionally reported bug whereby a message would remain on the queue
6760 longer than the retry timeout, but would be bounced if a delivery was
6761 forced. I say "may" because I never totally pinned down the problem;
6762 setting up timeout/retry tests is difficult. See also the next item.
6764 79. The ultimate address timeout was not being applied to errors that involved
6765 a combination of host plus message (for example, a timeout on a MAIL
6766 command). When an address resolved to a number of possible hosts, and they
6767 were not all tried for each delivery (e.g. because of hosts_max_try), a
6768 message could remain on the queue longer than the retry timeout.
6770 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6773 81. Fixed an obscure SMTP outgoing bug which required at least the following
6774 conditions: (a) there was another message waiting for the same server;
6775 (b) the server returned 5xx to all RCPT commands in the first message so
6776 that the message was not completed; (c) the server dropped the connection
6777 or gave a negative response to the RSET that Exim sends to abort the
6778 transaction. The observed case was a dropped connection after DATA that had
6779 been sent in pipelining mode. That is, the server had advertised PIPELINING
6780 but was not implementing it correctly. The effect of the bug was incorrect
6781 behaviour, such as trying another host, and this could lead to a crash.
6787 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6788 and daemon.c were passed as pointers to ints; they should have been
6789 pointers to socklen_t variables (which are typically unsigned ints).
6791 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6794 3. Fixed a really odd bug that affected only the testing scheme; patching a
6795 certain fixed string in the binary changed the value of another string that
6796 happened to be identical to the end of the original first string.
6798 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6799 name", it returns that address as the IP address. On some operating
6800 systems (e.g. Solaris), it also passes back the IP address string as the
6801 "host name". However, on others (e.g. Linux), it passes back an empty
6802 string. Exim wasn't checking for this, and was changing the host name to an
6803 empty string, assuming it had been canonicalized.
6805 5. Although rare, it is permitted to have more than one PTR record for a given
6806 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6807 all the names associated with an address, because they do in Solaris.
6808 However, it seems that they do not in Linux for data that comes from the
6809 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6810 I found this out when I moved to a new Linux workstation and tried to run
6811 the Exim test suite.
6813 To get round this problem I have changed the code so that it now does its
6814 own call to the DNS to look up PTR records when searching for a host name.
6815 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6816 addresses that are only in /etc/hosts are still found.
6818 This behaviour is, however, controlled by an option called host_lookup_
6819 order, which defaults to "bydns:byaddr". If people want to use the other
6820 order, or indeed, just use one or the other means of lookup, they can
6821 specify it in this variable.
6823 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6824 some operating systems, this comes back from gethostbyaddr() as an empty
6825 string, and this is what Exim used to test for. However, it seems that in
6826 other systems, "." is yielded. Exim now tests for this case too.
6828 7. The values of check_spool_space and check_log_space are now held internally
6829 as a number of kilobytes instead of an absolute number of bytes. If a
6830 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6831 kilobyte. This means that much larger values can be stored.
6833 8. Exim monitor: an attempt to get the action menu when not actually pointing
6834 at a message produces an empty menu entitled "No message selected". This
6835 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6836 no entries in it ("Shell widget menu has zero width and/or height"). So I
6837 have added a single, blank menu entry in this case.
6839 9. Added ${quote_local_part.
6841 10. MIME decoding is now applied to the contents of Subject: header lines when
6844 11. Now that a reference to $sender_host_address automatically causes a reverse
6845 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6846 host lookup before query-style lookups in lists that might use this
6847 variable. This has therefore been abolished, and the "net-" prefix is no
6848 longer necessary for query-style lookups.
6850 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6851 have been a typo for LFLAGS, so it has been changed.
6853 13. The install script calls Exim with "-C /dev/null" in order to find the
6854 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6855 to be output. However, since Exim outputs its version number before the
6856 error, it didn't break the script. It just looked ugly. I fixed this by
6857 always allowing "-C /dev/null" if the caller is root.
6859 14. Ignore overlarge ACL variable number when reading spool file - insurance
6860 against a later release with more variables having written the file.
6862 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6863 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6864 this, as well as the form without the "IPv6" on the front.
6866 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6867 OS/Makefile-Darwin file.
6869 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6870 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6872 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6873 variable. However, some OS use ints for inodes. I've added cast to long int
6874 to get rid of the compiler warning.
6876 19. I had forgotten to lock out "/../" in configuration file names when
6877 ALT_CONFIG_PREFIX was set.
6879 20. Routers used for verification do not need to specify transports. However,
6880 if such a router generated a host list, and callout was configured, Exim
6881 crashed, because it could not find a port number from the (non-existent)
6882 transport. It now assumes port 25 in this circumstance.
6884 21. Added the -t option to exigrep.
6886 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6887 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6888 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6889 them is set, LOOKUP_LSEARCH is forced.
6891 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6893 24. Added sender and host information to the "rejected by local_scan()" log
6894 line; previously there was no indication of these.
6896 25. Added .include_if_exists.
6898 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6899 Linux. It turns out that not all file systems support this. Apparently some
6900 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6901 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6902 not supported on the file descriptor, is now ignored when Exim is trying to
6903 sync a directory. This applies only to Linux.
6905 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6907 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6908 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6911 29. Added support for maildirsize files from supplied patch (modified a bit).
6913 30. The use of :fail: followed by an empty string could lead Exim to respond to
6914 sender verification failures with (e.g.):
6916 550 Verification failed for <xxx>
6917 550 Sender verify failed
6919 where the first response line was missing the '-' that indicates it is not
6920 the final line of the response.
6922 31. The loop for finding the name of the user that called Exim had a hardwired
6923 limit of 10; it now uses the value of finduser_retries, which is used for
6924 all other user lookups.
6926 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6929 33. Exim was neglecting to zero errno before one call of strtol() when
6930 expanding a string and expecting an integer value. On some systems this
6931 resulted in spurious "integer overflow" errors. Also, it was casting the
6932 result into an int without checking.
6934 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6935 did not work. The code looks as if it has *never* worked, though it appears
6936 to have been documented since at least release 1.62. I have made it work.
6938 35. The "timeout_DNS" error in retry rules, also documented since at least
6939 1.62, also never worked. As it isn't clear exactly what this means, and
6940 clearly it isn't a major issue, I have abolished the feature by treating it
6941 as "timeout", and writing a warning to the main and panic logs.
6943 36. The display of retry rules for -brt wasn't always showing the error code
6946 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6947 timeout_connect_A, timeout_connect_MX.
6949 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6950 to the empty sender.
6952 39. The daemon was not analysing the content of -oX till after it had closed
6953 stderr and disconnected from the controlling terminal. This meant that any
6954 syntax errors were only noted on the panic log, and the return code from
6955 the command was 0. By re-arranging the code a little, I've made the
6956 decoding happen first, so such errors now appear on stderr, and the return
6957 code is 1. However, the actual setting up of the sockets still happens in
6958 the disconnected process, so errors there are still only recorded on the
6961 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6962 connections (as happens on some IP stacks) was logged at start up time as
6963 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6964 it from "IPv6 and IPv4", which means that two separate sockets are being
6967 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6968 says whether AF_INET or AF_INET6 was passed as an argument.
6970 42. Exiwhat output was messed up when time zones were included in log
6973 43. Exiwhat now gives more information about the daemon's listening ports,
6974 and whether -tls-on-connect was used.
6976 44. The "port" option of the smtp transport is now expanded.
6978 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6979 silently ignored. Now an error message is written to the main and panic
6982 46. There's a new ACL modifier called "logwrite" which writes to a log file
6983 as soon as it is encountered.
6985 47. Added $local_user_uid and $local_user_gid at routing time.
6987 48. Exim crashed when trying to verify a sender address that was being
6990 49. Exim was recognizing only a space character after ".include". It now also
6991 recognizes a tab character.
6993 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6994 extracting the relevant information from the specification. The man page no
6995 longer contains scrambled data for the -d option, and I've added a section
6996 at the front about calling Exim under different names.
6998 51. Added "extra_headers" argument to the "mail" command in filter files.
7000 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
7003 53. Installed eximstats 1.29.
7005 54. Added transport_filter_timeout as a generic transport option.
7007 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
7008 Cc: header lines. This was required by RFC 822, but it not required by RFC
7011 56. Exim used to add From:, Date:, and Message-Id: header lines to any
7012 incoming messages that did not have them. Now it does so only if the
7013 message originates locally, that is, if there is no associated remote host
7014 address. When Resent- header lines are present, this applies to the Resent-
7015 lines rather than the non-Resent- lines.
7017 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
7018 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
7020 58. Messages for configuration errors now include the name of the main
7021 configuration file - useful now that there may be more than one file in a
7022 list (.included file names were always shown).
7024 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
7025 for those rare installations that do not start the daemon as root or run it
7026 setuid root. I've cut out the call to initgroups() if the daemon is not
7029 60. The Exim user and group can now be bound into the binary as text strings
7030 that are looked up at the start of Exim's processing.
7032 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
7034 62. Added $mailstore_basename variable.
7036 63. Installed patch to sieve.c from Michael Haardt.
7038 64. When Exim failed to open the panic log after failing to open the main log,
7039 the original message it was trying to log was written to stderr and debug
7040 output, but if they were not available (the usual case in production), it
7041 was lost. Now it is written to syslog before the two lines that record the
7042 failures to open the logs.
7044 65. Users' Exim filters run in subprocesses under the user's uid. It is
7045 possible for a "deliver" command or an alias in a "personal" command to
7046 provoke an address rewrite. If logging of address rewriting is configured,
7047 this fails because the process is not running as root or exim. There may be
7048 a better way of dealing with this, but for the moment (because 4.30 needs
7049 to be released), I have disabled address rewrite logging when running a
7050 filter in a non-root, non-exim process.
7056 1. The buildconfig auxiliary program wasn't quoting the value set for
7057 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
7058 not defined. This bug was masked in 4.22 by the effect that was fixed in
7061 2. Some messages that were rejected after a message id was allocated were
7062 shown as "incomplete" by exigrep. It no longer does this for messages that
7063 are rejected by local_scan() or the DATA or non-SMTP ACLs.
7065 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
7066 have allow_domain_literals set, the ID did not get logged in the <= line.
7067 Domain literals are now always recognized in Message-ID: header lines.
7069 4. The first argument for a ${extract expansion item is the key name or field
7070 number. Leading and trailing spaces in this item were not being ignored,
7071 causing some misleading effects.
7073 5. When deliver_drop_privilege was set, single queue runner processes started
7074 manually (i.e. by the command "exim -q") or by the daemon (which uses the
7075 same command in the process it spins off) were not dropping privilege.
7077 6. When the daemon running as "exim" started a queue runner, it always
7078 re-executed Exim in the spun-off process. This is a waste of effort when
7079 deliver_drop_privilege is set. The new process now just calls the
7080 queue-runner function directly.
7086 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
7089 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
7090 ignored. Though the use of -oP was forcing the writing of a pid file, it
7091 was always written to the default place.
7093 3. If the message "no IP address found for host xxxx" is generated during
7094 incoming verification, it is now followed by identification of the incoming
7095 connection (so you can more easily find what provoked it).
7097 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
7099 5. Added some features to "harden" Exim a bit more against certain attacks:
7101 (a) There is now a build-time option called FIXED_NEVER_USERS that can
7102 be put in Local/Makefile. This is like the never_users runtime option,
7103 but it cannot be overridden. The default setting is "root".
7105 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
7106 prefix string with which any file named in a -C command line option
7109 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
7110 is retained for -C and -D only if the caller of Exim is root. Without
7111 it, the exim user may also use -C and -D and retain privilege.
7113 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
7114 command line option is disabled.
7116 6. Macro names set by the -D option must start with an upper case letter, just
7117 like macro names defined in the configuration file.
7119 7. Added "dereference=" facility to LDAP.
7121 8. Two instances of the typo "uknown" in the source files are fixed.
7123 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
7124 the Configure-Makefile script screwed up while processing it.
7126 10. Incorporated PCRE 4.4.
7128 11. The SMTP synchronization check was not operating right at the start of an
7129 SMTP session. For example, it could not catch a HELO sent before the client
7130 waited for the greeting. There is now a check for outstanding input at the
7131 point when the greeting is written. Because of the duplex, asynchronous
7132 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
7133 way, but not yet received, when the check is performed.
7135 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
7136 on TCP/IP sockets, because this apparently causes some broken clients to
7139 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
7140 unchanged) from the Cygwin maintainer.
7142 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
7143 was supported instead of testing for mbx. Effectively a typo.
7145 15. The spa authenticator server code was not checking that the input it
7146 received was valid base64.
7148 16. The debug output line for the "set" modifier in ACLs was not showing the
7149 name of the variable that was being set.
7151 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
7153 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
7154 lookup on its own. Something else was needed to trigger the lookup. For
7155 example, a match in host_lookup or the need for a host name in a host list.
7156 Now, if $sender_host_name is referenced and the host name has not yet been
7157 looked up, a lookup is performed. If the lookup fails, the variable remains
7158 empty, and $host_lookup_failed is set to "1".
7160 19. Added "eqi" as a case-independent comparison operator.
7162 20. The saslauthd authentication condition could segfault if neither service
7163 nor realm was specified.
7165 21. If an overflowing value such as "2048M" was set for message_size_limit, the
7166 error message that was logged was misleading, and incoming SMTP
7167 connections were dropped. The message is now more accurate, and temporary
7168 errors are given to SMTP connections.
7170 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
7171 (except RSET) with a 421 error, until QUIT is received. However, it was
7172 failing to send a response to QUIT.
7174 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
7175 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
7176 ACL is now run after the helo_try_verify_hosts code.
7178 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
7179 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
7180 case-independent, so other case variants are also recognized). Apparently
7181 some systems use these upper case variants.
7183 25. If more than two messages were waiting for the same host, and a transport
7184 filter was specified for the transport, Exim sent two messages over the
7185 same TCP/IP connection, and then failed with "socket operation on non-
7186 socket" when it tried to send the third.
7188 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
7190 27. The extern definition of crypt16() in expand.c was not being excluded when
7191 the OS had its own crypt16() function.
7193 28. Added bounce_return_body as a new option, and bounce_return_size_limit
7194 as a preferred synonym for return_size_limit, both as an option and as an
7197 29. Added LIBS=-liconv to OS/Makefile-OSF1.
7199 30. Changed the default configuration ACL to relax the local part checking rule
7200 for addresses that are not in any local domains. For these addresses,
7201 slashes and pipe symbols are allowed within local parts, but the sequence
7202 /../ is explicitly forbidden.
7204 31. SPA server authentication was not clearing the challenge buffer before
7207 32. log_message in a "warn" ACL statement was writing to the reject log as
7208 well as to the main log, which contradicts the documentation and doesn't
7209 seem right (because no rejection is happening). So I have stopped it.
7211 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
7212 However, I am unable to do any testing of this.
7214 34. Fixed an infelicity in the appendfile transport. When checking directories
7215 for a mailbox, to see if any needed to be created, it was accidentally
7216 using path names with one or more superfluous leading slashes; tracing
7217 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
7219 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
7220 contents are added to the log line that is written for every discarded
7221 recipient. (Previously a log_message setting was ignored.)
7223 36. The ${quote: operator now quotes the string if it is empty.
7225 37. The install script runs exim in order to find its version number. If for
7226 some reason other than non-existence or emptiness, which it checks, it
7227 could not run './exim', it was installing it with an empty version number,
7228 i.e. as "exim-". This error state is now caught, and the installation is
7231 38. An argument was missing from the function that creates an error message
7232 when Exim fails to connect to the socket for saslauthd authentication.
7233 This could cause Exim to crash, or give a corrupted message.
7235 39. Added isip, isip4, and isip6 to ${if conditions.
7237 40. The ACL variables $acl_xx are now saved with the message, and can be
7238 accessed later in routers, transports, and filters.
7240 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
7241 strings in the file are not string-expanded.
7243 42. If a MAIL command specified a SIZE value that was too large to fit into an
7244 int variable, the check against message_size_limit failed. Such values are
7245 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
7246 day this will have to be increased, but I don't think I want to be around
7247 when emails are that large.
7254 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
7255 iconv() is not standard in FreeBSD.
7257 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
7258 IPv6 enabled. The observed symptom was a segmentation fault on return from
7259 the function os_common_find_running_interfaces() in src/os.c.
7261 3. In the check_special_case() function in daemon.c I had used "errno" as an
7262 argument name, which causes warnings on some systems. This was basically a
7263 typo, since it was named "eno" in the comments!
7265 4. The code that waits for the clock to tick (at a resolution of some fraction
7266 of a second) so as to ensure message-id uniqueness was always waiting for
7267 at least one whole tick, when it could have waited for less. [This is
7268 almost certainly not relevant at current processor speeds, where it is
7269 unlikely to ever wait at all. But we try to future-proof.]
7271 5. The function that sleeps for a time interval that includes fractions of a
7272 second contained a race. It did not block SIGALRM between setting the
7273 timer, and suspending (a couple of lines later). If the interval was short
7274 and the sigsuspend() was delayed until after it had expired, the suspension
7275 never ended. On busy systems this could lead to processes getting stuck for
7278 6. Some uncommon configurations may cause a lookup to happen in a queue runner
7279 process, before it forks any delivery processes. The open lookup caching
7280 mechanism meant that the open file or database connection was passed into
7281 the delivery process. The problem was that delivery processes always tidy
7282 up cached lookup data. This could cause a problem for the next delivery
7283 process started by the queue runner, because the external queue runner
7284 process does not know about the closure. So the next delivery process
7285 still has data in the lookup cache. In the case of a file lookup, there was
7286 no problem because closing a file descriptor in a subprocess doesn't affect
7287 the parent. However, if the lookup was caching a connection to a database,
7288 the connection was closed, and the second delivery process was likely to
7289 see errors such as "PGSQL: query failed: server closed the connection
7290 unexpectedly". The problem has been fixed by closing all cached lookups
7291 in a queue runner before running a delivery process.
7293 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
7294 seem to have the "const" qualifier which it has on other OS. I've
7297 8. Change 4.21/2 was too strict. It is only if there are two authenticators
7298 *of the same type* (client or server) with the same public name that an
7299 error should be diagnosed.
7301 9. When Exim looked up a host name for an IP address, but failed to find the
7302 original IP address when looking up the host name (a safety check), it
7303 output the message "<ip address> does not match any IP for NULL", which was
7304 confusing, to say the least. The bug was that the host name should have
7305 appeared instead of "NULL".
7307 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
7308 user that is built into the binary, and the -C or -D options is used, root
7309 privilege is dropped before the configuration file is read. In addition,
7310 logging is switched to stderr instead of the normal log files. If the
7311 configuration then re-defines the Exim user, the unprivileged environment
7312 is probably not what is expected, so Exim logs a panic warning message (but
7315 However, if deliver_drop_privilege is set, the unprivileged state may well
7316 be exactly what is intended, so the warning has been cut out in that case,
7317 and Exim is allowed to try to write to its normal log files.
7323 1. smtp_return_error_details was not giving details for temporary sender
7324 or receiver verification errors.
7326 2. Diagnose a configuration error if two authenticators have the same public
7329 3. Exim used not to create the message log file for a message until the first
7330 delivery attempt. This could be confusing when incoming messages were held
7331 for policy or load reasons. The message log file is now created at the time
7332 the message is received, and an initial "Received" line is written to it.
7334 4. The automatically generated man page for command line options had a minor
7335 bug that caused no ill effects; however, a more serious problem was that
7336 the procedure for building the man page automatically didn't always
7337 operate. Consequently, release 4.20 contains an out-of-date version. This
7338 shouldn't happen again.
7340 5. When building Exim with embedded Perl support, the script that builds the
7341 Makefile was calling 'perl' to find its compile-time parameters, ignoring
7342 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
7344 6. The freeze_tell option was not being used for messages that were frozen on
7345 arrival, either by an ACL or by local_scan().
7347 7. Added the smtp_incomplete_transaction log selector.
7349 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
7350 was accepting AUTH without a new EHLO.
7352 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
7353 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
7356 10. Logging of TCP/IP connections (when configured) now happens in the main
7357 daemon process instead of the child process, so that the TCP/IP connection
7358 count is more accurate (but it can never be perfect).
7360 11. The use of "drop" in a nested ACL was not being handled correctly in the
7361 outer ACL. Now, if condition failure induced by the nested "drop" causes
7362 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
7363 or "require"), the connection is dropped.
7365 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
7366 that yield "discard" can now be used with an "accept" or a "discard" verb,
7367 but an error is generated for any others (because I can't see a useful way
7368 to define what should happen).
7370 13. When an ACL is read dynamically from a file (or anywhere else), the lines
7371 are now processed in the same way as lines in the Exim configuration file.
7372 In particular, continuation lines are supported.
7374 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
7376 15. Added -ti meaning -t -i.
7378 16. Check for letters, digits, hyphens, and dots in the names of dnslist
7379 domains, and warn by logging if others are found.
7381 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
7382 returned from GIFCONF when Exim is trying to find the list of interfaces on
7383 a host. The code in os.c has been modified to copy each ifreq to an aligned
7384 structure in all cases.
7386 Also, in some cases, the returned ifreq's were being copied to a 'struct
7387 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
7388 means the last couple of bytes of an IPv6 address could be chopped if the
7389 ifreq contained only a normal sockaddr (14 bytes storage).
7391 18. Named domain lists were not supported in the hosts_treat_as_local option.
7392 An entry such as +xxxx was not recognized, and was treated as a literal
7395 19. Ensure that header lines added by a DATA ACL are included in the reject log
7396 if the ACL subsequently rejects the message.
7398 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
7399 MD5 (which is deprecated).
7401 21. When testing a filter file using -bf, Exim was writing a message when it
7402 took the sender from a "From " line in the message, but it was not doing so
7403 when it took $return_path from a Return-Path: header line. It now does.
7405 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
7406 with a valid header line field name (a series of printing characters
7407 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
7409 23. Changed "disc" in the source to "disk" to conform to the documentation and
7410 the book and for uniformity.
7412 24. Ignore Sendmail's -Ooption=value command line item.
7414 25. When execve() failed while trying to run a command in a pipe transport,
7415 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
7416 could be confused with a return value of 69 from the command itself. This
7417 has been changed to 127, the value the shell returns if it is asked to run
7418 a non-existent command. The wording for the related log line suggests a
7419 non-existent command as the problem.
7421 26. If received_header_text expands to an empty string, do not add a Received:
7422 header line to the message. (Well, it adds a token one on the spool, but
7423 marks it "old" so that it doesn't get used or transmitted.)
7425 27. Installed eximstats 1.28 (addition of -nt option).
7427 28. There was no check for failure on the call to getsockname() in the daemon
7428 code. This can fail if there is a shortage of resources on the system, with
7429 ENOMEM, for example. A temporary error is now given on failure.
7431 29. Contrary to the C standard, it seems that in some environments, the
7432 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
7433 program. Exim now does this explicitly; it affects the formatting of
7434 timestamps using strftime().
7436 30. If exiqsumm was given junk data, it threw up some uninitialized variable
7437 complaints. I've now initialized all the variables, to avoid this.
7439 32. Header lines added by a system filter were not being "seen" during
7440 transport-time rewrites.
7442 33. The info_callback() function passed to OpenSSL is set up with type void
7443 (*)(SSL *, int, int), as described somewhere. However, when calling the
7444 function (actually a macro) that sets it up, the type void(*)() is
7445 expected. I've put in a cast to prevent warnings from picky compilers.
7447 34. If a DNS black list lookup found a CNAME record, but there were no A
7448 records associated with the domain it pointed at, Exim crashed.
7450 35. If a DNS black list lookup returned more than one A record, Exim ignored
7451 all but the first. It now scans all returned addresses if a particular IP
7452 value is being sought. In this situation, the contents of the
7453 $dnslist_value variable are a list of all the addresses, separated by a
7456 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
7457 to accept a host name and all its aliases if the forward lookup for any of
7458 them yielded the IP address of the incoming connection. Now it accepts only
7459 those names whose forward lookup yields the correct IP address. Any other
7460 names are discarded. This closes a loophole whereby a rogue DNS
7461 administrator could create reverse DNS records to break through a
7462 wildcarded host restriction in an ACL.
7464 37. If a user filter or a system filter that ran in a subprocess used any of
7465 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
7466 the wrong values were passed to the pipe command ($thisaddress had the
7467 value of $0, $0 had the value of $1, etc). This bug was introduced by
7468 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
7470 38. Improved the line breaking for long SMTP error messages from ACLs.
7471 Previously, if there was no break point between 40 and 75 characters, Exim
7472 left the rest of the message alone. Two changes have been made: (a) I've
7473 reduced the minimum length to 35 characters; (b) if it can't find a break
7474 point between 35 and 75 characters, it looks ahead and uses the first one
7475 that it finds. This may give the occasional overlong line, but at least the
7476 remaining text gets split now.
7478 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
7479 file descriptors might be low, and that setting 1000 would always raise it.
7480 It turns out that in some environments, the limit is already over 1000 and
7481 that lowering it causes trouble. So now Exim takes care not to decrease it.
7483 40. When delivering a message, the value of $return_path is set to $sender_
7484 address at the start of routing (routers may change the value). By an
7485 oversight, this default was not being set up when an address was tested by
7486 -bt or -bv, which affected the outcome if any router or filter referred to
7489 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
7490 log only when "message" or "log_message" are set. However, if one of the
7491 conditions was an address verification, or a call to a nested ACL, the
7492 messages generated by the underlying test were being passed through. This
7493 no longer happens. The underlying message is available in $acl_verify_
7494 message for both "message" and "log_message" expansions, so it can be
7495 passed through if needed.
7497 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
7498 new expansion $bh_ to give the encoded byte string without charset
7499 translation. Translation happens only if iconv() is available; HAVE_ICONV
7500 indicates this at build time. HEADERS_CHARSET gives the charset to
7501 translate to; headers_charset can change it in the configuration, and
7502 "headers charset" can change it in an individual filter file.
7504 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
7505 that creates RFC 2047 encoded "words" labels them as that charset instead
7506 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
7507 expansion operator; (ii) when Exim creates a From: line for a local
7508 message; (iii) when a header line is rewritten to include a "phrase" part.
7510 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
7511 buggy, causing it to skip the first lines of messages whose message ID
7512 ended in 'D'. This would not have bitten before Exim release 4.14, because
7513 message IDs were unlikely to end in 'D' before then. The effect was to have
7514 incorrect size information for certain domains.
7516 45. #include "config.h" was missing at the start of the crypt16.c module. This
7517 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
7520 46. If there was a timeout during a "random" callout check, Exim treated it as
7521 a failure of the random address, and carried on sending RSET and the real
7522 address. If the delay was just some slowness somewhere, the response to the
7523 original RCPT would be taken as a response to RSET and so on, causing
7524 mayhem of various kinds.
7526 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
7527 when I implemented it. It didn't allow for the fact that some option values
7528 may legitimately be negative (e.g. size_addition), and it didn't even do
7529 the right test for positive values.
7531 48. Domain names in DNS records are case-independent. Exim always looks them up
7532 in lower case. Some resolvers return domain names in exactly the case they
7533 appear in the zone file, that is, they may contain uppercase letters. Not
7534 all resolvers do this - some return always lower case. Exim was treating a
7535 change of case by a resolver as a change of domain, similar to a widening
7536 of a domain abbreviation. This triggered its re-routing code and so it was
7537 trying to route what was effectively the same domain again. This normally
7538 caused routing to fail (because the router wouldn't handle the domain
7539 twice). Now Exim checks for this case specially, and just changes the
7540 casing of the domain that it ultimately uses when it transmits the message
7543 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
7546 50. If a filter generated a file delivery with a non-absolute name (possible if
7547 no home directory exists for the router), the forbid_file option was not
7550 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
7551 the existing equality matching.
7553 52. Exim was using ints instead of ino_t variables in some places where it was
7554 dealing with inode numbers.
7556 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
7557 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
7558 called TMPDIR, and if it finds it is different, it changes its value.
7560 54. The smtp_printf() function is now made available to local_scan() so
7561 additional output lines can be written before returning. There is also an
7562 smtp_fflush() function to enable the detection of a dropped connection.
7563 The variables smtp_input and smtp_batched_input are exported to
7566 55. Changed the default runtime configuration: the message "Unknown user"
7567 has been removed from the ACL, and instead placed on the localuser router,
7568 using the cannot_route_message feature. This means that any verification
7569 failures that generate their own messages won't get overridden. Similarly,
7570 the "Unrouteable address" message that was in the ACL for unverifiable
7571 relay addresses has also been removed.
7573 56. Added hosts_avoid_esmtp to the smtp transport.
7575 57. The exicyclog script was not checking for the esoteric option
7576 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7577 will work only if exicyclog is run under the appropriate euid.
7579 58. Following a discussion on the list, the rules by which Exim recognises line
7580 endings on incoming messages have been changed. The -dropcr and drop_cr
7581 options are now no-ops, retained only for backwards compatibility. The
7582 following line terminators are recognized: LF CRLF CR. However, special
7583 processing applies to CR:
7585 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7586 nor a local message in the state where . is a terminator.
7588 (ii) If a bare CR is encountered in a header line, an extra space is added
7589 after the line terminator so as not to end the header. The reasoning
7590 behind this is that bare CRs in header lines are most likely either
7591 to be mistakes, or people trying to play silly games.
7593 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7594 was being incorrectly given as 18 bytes larger than it should have been.
7595 This is a VOB (very old bug).
7597 60. This may never have affected anything current, but just in case it has:
7598 When the local host is found other than at the start of a list of hosts,
7599 the local host, those with the same MX, and any that follow, are discarded.
7600 When the list in question was part of a longer list of hosts, the following
7601 hosts (not currently being processed) were also being discarded. This no
7602 longer happens. I'm not sure if this situation could ever has previously
7605 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7608 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7609 header. This is something that is recommended in a new Internet Draft, and
7610 is something that is documented as being done by Sendmail. There are two
7611 possible values. For messages generated by the autoreply transport, Exim
7614 Auto-Submitted: auto-replied
7616 whereas for all other generated messages (e.g. bounces) it adds
7618 Auto-Submitted: auto-generated
7620 63. The "personal" condition in filters now includes a test for the
7621 Auto-Submitted: header. If it contains the string "auto-" the message it
7622 not considered personal.
7624 64. Added rcpt_include_affixes as a generic transport option.
7626 65. Added queue_only_override (default true).
7628 66. Added the syslog_duplication option.
7630 67. If what should have been the first header line of a message consisted of
7631 a space followed by a colon, Exim was mis-interpreting it as a header line.
7632 It isn't of course - it is syntactically invalid and should therefore be
7633 treated as the start of the message body. The misbehaviour could have
7634 caused a number of strange effects, including loss of data in subsequent
7635 header lines, and spool format errors.
7637 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7638 client host had authenticated. This control can now be exercised by an ACL
7639 for more flexibility.
7641 69. By default, callouts do not happen when testing with -bh. There is now a
7642 variant, -bhc, which does actually run the callout code, including
7643 consulting and updating the callout cache.
7645 70. Added support for saslauthd authentication, courtesy of Alexander
7648 71. If statvfs() failed on the spool or log directories while checking their
7649 size for availability, Exim confusingly gave the error "space shortage".
7650 Furthermore, in debugging mode it crashed with a floating point exception.
7651 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7652 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7653 serious problem, Exim now writes to the main and panic logs when this
7654 happens, with details of the failure. It then refuses to accept the
7655 incoming message, giving the message "spool directory problem" or "log
7656 directory problem" with a 421 code for SMTP messages.
7658 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7659 0, 1, and 2 exist, because some OS complain for execs without them (see
7660 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7661 descriptors. However, the code omitted to check that the open succeeded,
7662 causing mysterious errors if for some reason the permissions on /dev/null
7663 got screwed. Now Exim writes a message to the main and panic logs, and
7664 bombs out if it can't open /dev/null.
7666 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7667 interact so that it is all more flexible. It is supposed to remain
7668 backwards compatible. Also added extra_local_interfaces.
7670 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7671 to bomb out with an assertion failure - to the client this appears as a
7672 connection drop. This problem occurs in the part of the code that was taken
7673 from the Samba project. Fortunately, the assertion is in a very simple
7674 function, so I have fixed this by reproducing the function inline in the
7675 one place where it is called, and arranging for authentication to fail
7676 instead of killing the process with assert().
7678 75. The SPA client code was not working when the server requested OEM rather
7679 than Unicode encoding.
7681 76. Added code to make require_files with a specific uid setting more usable in
7682 the case where statting the file as root fails - usually a non-root-mounted
7683 NFS file system. When this happens and the failure is EACCES, Exim now
7684 forks a subprocess and does the per-uid checking as the relevant uid.
7686 77. Added process_log_path.
7688 78. If log_file_path was not explicitly set, a setting of check_log_space or
7689 check_log_inodes was ignored.
7691 79. If a space check for the spool or log partitions fails, the incident is now
7692 logged. Of course, in the latter case the data may get lost...
7694 80. Added the %p formatting code to string_format() so that it can be used to
7695 print addresses in debug_print(). Adjusted all the address printing in the
7696 debugging in store.c to use %p rather than %d.
7698 81. There was a concern that a line of code in smtp_in.c could overflow a
7699 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7700 initially expressed, the concern was not well-founded, because trailing
7701 spaces are removed early. However, if the trailing spaces were followed by
7702 a NULL, they did not get removed, so the overflow was possible. Two fixes
7705 (a) I re-wrote the offending code in a cleaner fashion.
7706 (b) If an incoming SMTP command contains a NULL character, it is rejected
7709 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7710 runs initgroups(), so that if the Exim user is in any additional groups,
7711 they will be used during message reception.
7717 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob