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 Bug 1329: Fix format of Maildir-format filenames to match other mail-
10 related applications. Previously an "H" was used where available info
11 says that "M" should be, so change to match.
13 JH/02 Bug 2587: Fix pam expansion condition. Tainted values are commonly used
14 as arguments, so an implementation trying to copy these into a local
15 buffer was taking a taint-enforcement trap. Fix by using dynamically
16 created buffers. Similar fix for radius expansion condition.
18 JH/03 Bug 2586: Fix listcount expansion operator. Using tainted arguments is
19 reasonable, eg. to count headers. Fix by using dynamically created
20 buffers rather than a local. Do similar fixes for ACL actions "dcc",
21 "log_reject_target", "malware" and "spam"; the arguments are expanded
22 so could be handling tainted values.
24 JH/04 Bug 2590: Fix -bi (newaliases). A previous code rearrangement had
25 broken the (no-op) support for this sendmail command. Restore it
26 to doing nothing, silently, and returning good status.
28 JH/05 Bug 2593: Fix "vacation" in Exim filter. Previously, when a "once"
29 record path was given (or the default used) without a leading directory
30 path, an error occurred on trying to open it. Use the transport's working
33 JH/06 Bug 2594: Change the name used for certificate name checks in the smtp
34 transport. Previously it was the name on the DNS A-record; use instead
35 the head of the CNAME chain leading there (if there is one). This seems
36 to align better with RFC 6125.
38 JH/07 Bug 2597: Fix a resource leak. Using a lookup in obtaining a value for
39 smtp_accept_max_per_host allocated resources which were not released
40 when the limit was exceeded. This eventually crashed the daemon. Fix
41 by adding a relase action in that path.
43 JH/08 Bug 2598: Fix verify ACL condition. The options for the condition are
44 expanded; previously using tainted values was rejected. Fix by using
45 dynamically-created buffers.
47 JH/09 Relax restrictions on ACL verify condition needing access to message
48 headers. Previously they were only permitted in data and non-smtp ACLs;
49 permit also mime, dkim, prdr quit and notquit. Applies to header-syntax,
50 not_blind, header_sender and header_names_ascii verification.
52 JH/10 Bug 2603: Fix coding of string copying to only evaluate arguments once.
53 Previously a macro used one argument twice; when called with the
54 argument as an expression having side-effects, incorrect operation
55 resulted. Use an inlineable function.
57 JH/11 Bug 2604: Fix request to cutthrough-deliver when a connection is already
58 held open for a verify callout. Previously this wan not accounted for
59 and a corrupt onward SMTP conversation resulted.
61 JH/12 Bug 2607: Fix the ${srs_encode } expansion to handle quoted local_parts.
62 Previously they were embedded naively in the constructed address; when
63 needed, strip the quoting and quote the entire local_part.
64 Also make the inbound_srs expansion condition handle quoting.
66 JH/13 Fix dsearch "subdir" filter to ignore ".". Previously only ".." was
67 excluded, not matching the documentation.
69 JH/14 Bug 2606: Fix a segfault in sqlite lookups. When no, or a bad, filename
70 was given for the sqlite_dbfile a trap resulted.
72 JH/15 Bug 2620: Fix "spam" ACL condition. Previously, tainted values for the
73 "name" argument resulted in a trap. There is no reason to disallow such;
74 this was a coding error.
76 JH/16 Bug 2615: Fix pause during message reception, on systems that have been
77 suspended/resumed. The Linux CLOCK_MONOTONIC does not account for time
78 spent suspended, ignoring the Posix definition. Previously we assumed
79 it did and a constant offset from real time could be used as a correction.
80 Change to using the same clock source for the start-of-message and the
81 post-message next-tick-wait. Also change to using CLOCK_BOOTTIME if it
82 exists, just to get a clock slightly more aligned to reality.
84 JH/17 Bug 2295: Fix DKIM signing to always semicolon-terminate. Although the
85 RFC says it is optional some validators care. The missing char was not
86 intended but triggered by a line-wrap alignement. Discovery and fix by
87 Guillaume Outters, hacked on by JH.
89 JH/18 Bug 2617: Fix a taint trap in parse_fix_phrase(). Previously when the
90 name being quoted was tainted a trap would be taken. Fix by using
91 dynamicaly created buffers. The routine could have been called by a
92 rewrite with the "h" flag, by using the "-F" command-line option, or
93 by using a "name=" option on a control=submission ACL modifier.
95 JH/19 SPF: change the Authentication-Results expansion component to give
96 smtp.helo when the sender domain is empty. Previously it gave
99 JH/20 Bug 2631: ACL dnslist conditions now ignore and log any lookups returns
100 not in 127.0.0.0/8 to help in spotting list domains taken over by a
101 domain-parking registrar.
103 JH/21 Bug 2630: Fix eol-replacement string for the ${readsocket } expansion.
104 Previously when a whitespace character was specified it was not inserted
105 after removing the newline.
107 JH/22 Bug 2265: Force SNI usage for smtp transport DANE'd connections, to be
108 the domain part of the recipient address. This overrides any tls_sni
109 option set, which was previously used.
111 JH/23 Logging: with the +tls_sni log_selector, do not wrap the received SNI
114 JH/24 Bug 2634: Fix a taint trap seen on NetBSD: the testing coded for
115 is_tainted() had an off-by-one error in the overenthusiastic direction.
116 Find and fix by Gavan. Although NetBSD is not a supported platform for
117 4.94 this bug could affect other platforms.
119 PP/01 Fix default prime selection to be consistent.
120 One path used ike23 still, instead of exim.dev.20160529.3; now both
121 execution flows will use the same DH primes (currently
122 exim.dev.20160529.3).
124 JH/25 OpenSSL: Fix back-compatibility behaviour surrounding tls_certificates
125 option in smtp transport, to match the documentation. Previously
126 verification was not being done in some cases where it should have been.
128 JH/26 Bug 2646: fix a memory usage issue in ldap lookups. Previously, when more
129 than one server was defined and depending on the platform memory layout
130 details, an internal consistency trap could be hit while walking the list
133 JH/27 Bug 2648: fix the passing of an authenticator public-name through spool
134 files. The value is used by the authresults expansion item. Previously
135 if this was used in a router or transport, a crash could result.
137 JH/28 Fix spurious logging of select error. Some platforms, notably FreeBSD,
138 have a sufficient incidence of EINTR returns from select that an
139 interaction with other operations done by the main daemon loop exposed
140 a bug in the error-handling. This was benign apart from the log
143 JH/29 Bug 2675: add outgoing-interface I= element to deferred "==" log lines,
144 for consistency with delivered "=>" and failed "**" lines. While we're
145 there, handle PRX and TFO.
147 JH/30 Bug 2677: fix matching of long addresses. Since 4.93 a limit of 256 was
148 applied. This resulted, if any header-line rewrite rules were configured,
149 in a panic-log trigerrable by sending a message with a long address in
150 a header. Fix by increaing the arbitrary limit to larger than a single
151 (dewrapped) 5322 header line maximum size.
153 JH/31 The ESMTP option name advertised for the SUPPORT_EARLY_PIPE build option
154 is changed from X_PIPE_CONNECT to PIPE_CONNECT. This is in line with
155 RFC 6648 which deprecates X- options in protocols as a general practice.
156 Changeover between the implementations is handled by the mechanisms
159 JH/32 Bug 2599: fix delay of delivery to a local address where there is also
160 a remote which uses callout/hold. Previously the local was queued.
162 JH/33 Fix a taint trap in the ${listextract } expansion when the source data
165 JH/34 Fix the placement of a multiple-message delivery marker in the delivery
166 log line. The asterisk is now consistently appended to the remote IP
167 (and port, if given), and will also be provided on defer and fail log
168 lines. Previously it could be placed on the local IP if that was being
169 logged, and was only provided on delivery lines.
171 JH/35 Bug 2343: Harden exim_tidydb against corrupt wait- files.
173 JH/36 Bug 2687: Fix interpretation of multiple ^ chars in a plaintext
174 authenticator client_send option. Previously the next char, after a pair
175 was collapsed, was taken verbatim (so ^^^foo became ^^foo; ^^^^foo became
176 ^^\x00foo). Fixed to get ^\x00foo and ^^foo respectively to match the
177 documentation. There is still no way to get a leading ^ immediately
178 after a NUL (ie. for the password of a PLAIN method authenticator.
180 JH/37 Enforce the expected size, for fixed-size records read from hints-DB
181 files. For bad sizes read, delete the record and whine to paniclog.
183 JH/38 When logging an AUTH failure, as server, do not include sensitive
184 information. Previously, the credentials would be included if given
185 as part of the AUTH command line and an ACL denied authentidcation.
187 JH/39 Bug 2691: fix $local_part_data. When the matching list element
188 referred to a file, bad data was returned. This likely also affected
191 JH/40 The gsasl authenticator now supports caching of the salted password
192 generated by the client-side implementation. This required the addition
193 of a new variable: $auth4.
195 JH/41 Fix daemon SIGHUP on FreeBSD. Previously, a named socket for IPC was
196 left undeleted; the attempt to re-create it then failed - resulting in
197 the usual "SIGHUP tp have daemon reload configuration" to not work.
198 This affected any platform not supporting "abstract" Unix-domain
199 sockets (i.e. not Linux).
201 JH/42 Bug 2693: Harden against a peer which reneges on a 452 "too many
202 recipients" response to RCPT in a later response, with a 250. The
203 previous coding assumed this would not happen, and under PIPELINING
204 would result in both lost and duplicate recipients for a message.
206 JH/43 Bug 2694: Fix weighted distribution of work to multiple spamd servers.
207 Previously the weighting was incorrectly applied. Similar fix for socks
208 proxies. Found and fixed by Heiko Schlichting.
210 JH/44 Bug 2701: Fix list-expansion of dns_ipv4_lookup. Previously, it did
211 not handle sub-lists included using the +namedlist syntax. While
212 investigating, the same found for dns_trust_aa, dns_again_means_nonexist,
213 dnssec_require_domains, dnssec_request_domains, srv_fail_domains,
216 JH/45 Use a (new) separate store pool-pair for DKIM verify working data.
217 Previously the permanent pool was used, so the sore could not be freed.
218 This meant a connection with many messages would use continually-growing
221 JH/46 Use an exponentially-increasing block size when malloc'ing store. Do it
222 per-pool so as not to waste too much space. Previously a constant size
223 was used which resulted in O(n^2) behaviour; now we get O(n log n) making
224 DOS attacks harder. The cost is wasted memory use in the larger blocks.
226 JH/47 Use explicit alloc/free for DNS lookup workspace. This permits using the
227 same space repeatedly, and a smaller process footprint.
235 JH/01 Avoid costly startup code when not strictly needed. This reduces time
236 for some exim process initialisations. It does mean that the logging
237 of TLS configuration problems is only done for the daemon startup.
239 JH/02 Early-pipelining support code is now included unless disabled in Makefile.
241 JH/03 DKIM verification defaults no long accept sha1 hashes, to conform to
242 RFC 8301. They can still be enabled, using the dkim_verify_hashes main
245 JH/04 Support CHUNKING from an smtp transport using a transport_filter, when
246 DKIM signing is being done. Previously a transport_filter would always
247 disable CHUNKING, falling back to traditional DATA.
249 JH/05 Regard command-line recipients as tainted.
251 JH/06 Bug 340: Remove the daemon pid file on exit, when due to SIGTERM.
253 JH/07 Bug 2489: Fix crash in the "pam" expansion condition. It seems that the
254 PAM library frees one of the arguments given to it, despite the
255 documentation. Therefore a plain malloc must be used.
257 JH/08 Bug 2491: Use tainted buffers for the transport smtp context. Previously
258 on-stack buffers were used, resulting in a taint trap when DSN information
259 copied from a received message was written into the buffer.
261 JH/09 Bug 2493: Harden ARC verify against Outlook, whick has been seen to mix
262 the ordering of its ARC headers. This caused a crash.
264 JH/10 Bug 2492: Use tainted memory for retry record when needed. Previously when
265 a new record was being constructed with information from the peer, a trap
268 JH/11 Bug 2494: Unset the default for dmarc_tld_file. Previously a naiive
269 installation would get error messages from DMARC verify, when it hit the
270 nonexistent file indicated by the default. Distros wanting DMARC enabled
271 should both provide the file and set the option.
272 Also enforce no DMARC verification for command-line sourced messages.
274 JH/12 Fix an uninitialised flag in early-pipelining. Previously connections
275 could, depending on the platform, hang at the STARTTLS response.
277 JH/13 Bug 2498: Reset a counter used for ARC verify before handling another
278 message on a connection. Previously if one message had ARC headers and
279 the following one did not, a crash could result when adding an
280 Authentication-Results: header.
282 JH/14 Bug 2500: Rewind some of the common-coding in string handling between the
283 Exim main code and Exim-related utities. The introduction of taint
284 tracking also did many adjustments to string handling. Since then, eximon
285 frequently terminated with an assert failure.
287 JH/15 When PIPELINING, synch after every hundred or so RCPT commands sent and
288 check for 452 responses. This slightly helps the inefficieny of doing
289 a large alias-expansion into a recipient-limited target. The max_rcpt
290 transport option still applies (and at the current default, will override
291 the new feature). The check is done for either cause of synch, and forces
292 a fast-retry of all 452'd recipients using a new MAIL FROM on the same
293 connection. The new facility is not tunable at this time.
295 JH/16 Fix the variables set by the gsasl authenticator. Previously a pointer to
296 library live data was being used, so the results became garbage. Make
297 copies while it is still usable.
299 JH/17 Logging: when the deliver_time selector ise set, include the DT= field
300 on delivery deferred (==) and failed (**) lines (if a delivery was
301 attemtped). Previously it was only on completion (=>) lines.
303 JH/18 Authentication: the gsasl driver not provides the $authN variables in time
304 for the expansion of the server_scram_iter and server_scram_salt options.
306 WB/01 SPF: DNS lookups for the obsolete SPF RR type done by the libspf2 library
307 are now specifically given a NO_DATA response without hitting the system
308 resolver. The library goes on to do the now-standard TXT lookup.
309 Use of dnsdb lookups is not affected.
311 JH/19 Bug 2507: Modules: on handling a dynamic-module (lookups) open failure,
312 only retrieve the errormessage once. Previously two calls to dlerror()
313 were used, and the second one (for mainlog/paniclog) retrieved null
316 JH/20 Taint checking: disallow use of tainted data for
317 - the appendfile transport file and directory options
318 - the pipe transport command
319 - the autoreply transport file, log and once options
320 - file names used by the redirect router (including filter files)
322 - paths used by single-key lookups
323 Previously this was permitted.
325 JH/21 Bug 2501: Fix init call in the heimdal authenticator. Previously it
326 adjusted the size of a major service buffer; this failed because the
327 buffer was in use at the time. Change to a compile-time increase in the
328 buffer size, when this authenticator is compiled into exim.
330 JH/22 Taint-checking: move to safe-mode taint checking on all platforms. The
331 previous fast-mode was untenable in the face of glibs using mmap to
332 support larger malloc requests.
334 PP/01 Update the openssl_options possible values through OpenSSL 1.1.1c.
335 New values supported, if defined on system where compiled:
336 allow_no_dhe_kex, cryptopro_tlsext_bug, enable_middlebox_compat,
337 no_anti_replay, no_encrypt_then_mac, prioritize_chacha, tlsext_padding
339 JH/23 Performance improvement in the initial phase of a two-pass queue run. By
340 running a limited number of proceses in parallel, a benefit is gained. The
341 amount varies with the platform hardware and load. The use of the option
342 queue_run_in_order means we cannot do this, as ordering becomes
345 JH/24 Bug 2524: fix the cyrus_sasl auth driver gssapi usage. A previous fix
346 had introduced a string-copy (for ensuring NUL-termination) which was not
347 appropriate for that case, which can include embedded NUL bytes in the
348 block of data. Investigation showed the copy to actually be needless, the
349 data being length-specified.
351 JH/25 Fix use of concurrent TLS connections under GnuTLS. When a callout was
352 done during a receiving connection, and both used TLS, global info was
353 used rather than per-connection info for tracking the state of data
354 queued for transmission. This could result in a connection hang.
356 JH/26 Fix use of the SIZE parameter on MAIL commands, on continued connections.
357 Previously, when delivering serveral messages down a single connection
358 only the first would provide a SIZE. This was due to the size information
359 not being properly tracked.
361 JH/27 Bug 2530: When operating in a timezone with sub-minute offset, such as
362 TAI (at 37 seconds currently), pretend to be in UTC for time-related
363 expansion and logging. Previously, spurious values such as a future
364 minute could be seen.
366 JH/28 Bug 2533: Fix expansion of ${tr } item. When called in some situations
367 it could crash from a null-deref. This could also affect the
368 ${addresses: } operator and ${readsock } item.
370 JH/29 Bug 2537: Fix $mime_part_count. When a single connection had a non-mime
371 message following a mime one, the variable was not reset.
373 JH/30 When an pipelined-connect fails at the first response, assume incorrect
374 cached capability (perhaps the peer reneged?) and immediately retry in
377 JH/31 Fix spurious detection of timeout while writing to transport filter.
379 JH/32 Bug 2541: Fix segfault on bad cmdline -f (sender) argument. Previously
380 an attempt to copy the string was made before checking it.
382 JH/33 Fix the dsearch lookup to return an untainted result. Previously the
383 taint of the lookup key was maintained; we now regard the presence in the
384 filesystem as sufficient validation.
386 JH/34 Fix the readsocket expansion to not segfault when an empty "options"
387 argument is supplied.
389 JH/35 The dsearch lookup now requires that the directory is an absolute path.
390 Previously this was not checked, and nonempty relative paths made an
391 access under Exim's current working directory.
393 JH/36 Bug 2554: Fix msg:defer event for the hosts_max_try_hardlimit case.
394 Previously no event was raised.
396 JH/37 Bug 2552: Fix the check on spool space during reception to use the SIZE
397 parameter supplied by the sender MAIL FROM command. Previously it was
398 ignored, and only the check_spool_space option value for the required
401 JH/38 Fix $dkim_key_length. This should, after a DKIM verification, present
402 the size of the signing public-key. Previously it was instead giving
403 the size of the signature hash.
405 JH/39 DKIM verification: the RFC 8301 restriction on sizes of RSA keys is now
406 the default. See the (new) dkim_verify_min_keysizes option.
408 JH/40 Fix a memory-handling bug: when a connection carried multiple messages
409 and an ACL use a lookup for checking either the local_part or domain,
410 stale data could be accessed. Ensure that variable references are
411 dropped between messages.
413 JH/41 Bug 2571: Fix SPA authenticator. Running as a server, an offset supplied
414 by the client was not checked as pointing within response data before
415 being used. A malicious client could thus cause an out-of-bounds read and
416 possibly gain authentication. Fix by adding the check.
418 JH/42 Internationalisation: change the default for downconversion in the smtp
419 transport to be "if needed". Previously it was "as previously set" for
420 the message, which usually meant "if needed" for message-submission but
421 "no" for everything else. However, MTAs have been seen using SMTPUTF8
422 even when the envelope addresses did not need it, resulting in forwarding
423 failures to non-supporting MTAs. A downconvert in such cases will be
424 a no-op on the addresses, merely dropping the use of SMTPUTF8 by the
425 transport. The change does mean that addresses needing conversion will
426 be converted when previously a delivery failure would occur.
428 JH/43 Fix possible long line in DSN. Previously when a very long SMTP error
429 response was received it would be used unchecked in a fail-DSN, violating
430 standards on line-length limits. Truncate if needed.
432 HS/01 Remove parameters of the link to www.open-spf.org. The linked form
433 doesn't work. (Additionally add a new main config option to configure the
440 JH/01 OpenSSL: With debug enabled output keying information sufficient, server
441 side, to decode a TLS 1.3 packet capture.
443 JH/02 OpenSSL: Suppress the sending of (stateful) TLS1.3 session tickets.
444 Previously the default library behaviour applied, sending two, each in
447 JH/03 Debug output for ACL now gives the config file name and line number for
450 JH/04 The default received_header_text now uses the RFC 8314 tls cipher clause.
452 JH/05 DKIM: ensure that dkim_domain elements are lowercased before use.
454 JH/06 Fix buggy handling of autoreply bounce_return_size_limit, and a possible
455 buffer overrun for (non-chunking) other transports.
457 JH/07 GnuTLS: Our use of late (post-handshake) certificate verification, under
458 TLS1.3, means that a server rejecting a client certificate is not visible
459 to the client until the first read of encrypted data (typically the
460 response to EHLO). Add detection for that case and treat it as a failed
461 TLS connection attempt, so that the normal retry-in-clear can work (if
462 suitably configured).
464 JB/01 Bug 2375: fix expansions of 822 addresses having comments in local-part
465 and/or domain. Found and fixed by Jason Betts.
467 JH/08 Add hardening against SRV & TLSA lookups the hit CNAMEs (a nonvalid
468 configuration). If a CNAME target was not a wellformed name pattern, a
471 JH/09 Logging: Fix initial listening-on line for multiple ports for an IP when
472 the OS reports them interleaved with other addresses.
474 JH/10 OpenSSL: Fix aggregation of messages. Previously, when PIPELINING was
475 used both for input and for a verify callout, both encrypted, SMTP
476 responses being sent by the server could be lost. This resulted in
477 dropped connections and sometimes bounces generated by a peer sending
480 JH/11 Harden plaintext authenticator against a badly misconfigured client-send
481 string. Previously it was possible to cause undefined behaviour in a
482 library routine (usually a crash). Found by "zerons".
484 JH/12 Bug 2384: fix "-bP smtp_receive_timeout". Previously it returned no
487 JH/13 Bug 2386: Fix builds with Dane under LibreSSL 2.9.0 onward. Some old
488 API was removed, so update to use the newer ones.
490 JH/14 Bug 1891: Close the log file if receiving a non-smtp message, without
491 any timeout set, is taking a long time. Previously we would hang on to a
492 rotated logfile "forever" if the input was arriving with long gaps
493 (a previous attempt to fix addressed lack, for a long time, of initial
496 HS/01 Bug 2390: Use message_id for tempfile creation to avoid races in a
497 shared (NFS) environment. The length of the tempfile name is now
498 4 + 16 ("hdr.$message_exim_id") which might break on file
499 systems which restrict the file name length to lower values.
502 HS/02 Bug 2390: Use message_id for tempfile creation to avoid races in a
503 shared (NFS) environment.
505 HS/03 Bug 2392: exigrep does case sensitive *option* processing (as it
506 did for all versions <4.90). Notably -M, -m, --invert, -I may be
509 JH/15 Use unsigned when creating bitmasks in macros, to avoid build errors
510 on some platforms for bit 31.
512 JH/16 GnuTLS: rework ciphersuite strings under recent library versions. Thanks
513 to changes apparently associated with TLS1.3 handling some of the APIs
514 previously used were either nonfunctional or inappropriate. Strings
515 like TLS1.3:ECDHE_SECP256R1__RSA_PSS_RSAE_SHA256__AES_256_GCM__AEAD:256
516 and TLS1.2:ECDHE_SECP256R1__RSA_SHA256__AES_128_CBC__SHA256:128 replace
517 the previous TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256 .
518 This affects log line X= elements, the $tls_{in,out}_cipher variables,
519 and the use of specific cipher names in the encrypted= ACL condition.
521 JH/17 OpenSSL: the default openssl_options now disables ssl_v3.
523 JH/18 GnuTLS: fix $tls_out_ocsp under hosts_request_ocsp. Previously the
524 verification result was not updated unless hosts_require_ocsp applied.
526 JH/19 Bug 2398: fix listing of a named-queue. Previously, even with the option
527 queue_list_requires_admin set to false, non-admin users were denied the
530 JH/20 Bug 2389: fix server advertising of usable certificates, under GnuTLS in
531 directory-of-certs mode. Previously they were advertised despite the
534 JH/21 The smtp transport option "hosts_noproxy_tls" is now unset by default.
535 A single TCP connection by a client will now hold a TLS connection open
536 for multiple message deliveries, by default. Previously the default was to
539 JH/22 The smtp transport option "hosts_try_dane" now enables all hosts by
540 default. If built with the facility, DANE will be used. The facility
541 SUPPORT_DANE is now enabled in the prototype build Makefile "EDITME".
543 JH/23 The build default is now for TLS to be included; the SUPPORT_TLS define
544 is replaced with DISABLE_TLS. Either USE_GNUTLS or (the new) USE_OPENSSL
545 must be defined and you must still, unless you define DISABLE_TLS, manage
546 the the include-dir and library-file requirements that go with that
547 choice. Non-TLS builds are still supported.
549 JH/24 Fix duplicated logging of peer name/address, on a transport connection-
552 JH/25 The smtp transport option "hosts_try_fastopen" now enables all hosts by
553 default. If the platform supports and has the facility enabled, it will
554 be requested on all coneections.
556 JH/26 The PIPE_CONNECT facility is promoted from experimental status and is now
557 controlled by the build-time option SUPPORT_PIPE_CONNECT.
559 PP/01 Unbreak heimdal_gssapi, broken in 4.92.
561 JH/27 Bug 2404: Use the main-section configuration option "dsn_from" for
562 success-DSN messages. Previously the From: header was always the default
563 one for these; the option was ignored.
565 JH/28 Fix the timeout on smtp response to apply to the whole response.
566 Previously it was reset for every read, so a teergrubing peer sending
567 single bytes within the time limit could extend the connection for a
568 long time. Credit to Qualsys Security Advisory Team for the discovery.
570 JH/29 Fix DSN Final-Recipient: field. Previously it was the post-routing
571 delivery address, which leaked information of the results of local
572 forwarding. Change to the original envelope recipient address, per
575 JH/30 Bug 2411: Fix DSN generation when RFC 3461 failure notification is
576 requested. Previously not bounce was generated and a log entry of
577 error ignored was made.
579 JH/31 Avoid re-expansion in ${sort } expansion. (CVE-2019-13917)
581 JH/32 Introduce a general tainting mechanism for values read from the input
582 channel, and values derived from them. Refuse to expand any tainted
583 values, to catch one form of exploit.
585 JH/33 Bug 2413: Fix dkim_strict option. Previously the expansion result
586 was unused and the unexpanded text used for the test. Found and
587 fixed by Ruben Jenster.
589 JH/34 Fix crash after TLS shutdown. When the TCP/SMTP channel was left open,
590 an attempt to use a TLS library read routine dereffed a nul pointer,
593 JH/35 Bug 2409: filter out-of-spec chars from callout response before using
594 them in our smtp response.
596 JH/36 Have the general router option retry_use_local_part default to true when
597 any of the restrictive preconditions are set (to anything). Previously it
598 was only for check_local user. The change removes one item of manual
599 configuration which is required for proper retries when a remote router
600 handles a subset of addresses for a domain.
602 JH/37 Appendfile: when evaluating quota use (non-quota_size_regex) take the file
603 link count into consideration.
605 HS/04 Fix handling of very log lines in -H files. If a -<key> <value> line
606 caused the extension of big_buffer, the following lines were ignored.
608 JH/38 Bug 1395: Teach the DNS negative-cache about TTL value from the SOA in
609 accordance with RFC 2308. Previously there was no expiry, so a longlived
610 receive process (eg. due to ACL delays) versus a short SOA value could
613 HS/05 Handle trailing backslash gracefully. (CVE-2019-15846)
615 JH/39 Promote DMARC support to mainline.
617 JH/40 Bug 2452: Add a References: header to DSNs.
619 JH/41 With GnuTLS 3.6.0 (and later) do not attempt to manage Diffie-Hellman
620 parameters. The relevant library call is documented as "Deprecated: This
621 function is unnecessary and discouraged on GnuTLS 3.6.0 or later. Since
622 3.6.0, DH parameters are negotiated following RFC7919."
624 HS/06 Change the default of dnssec_request_domains to "*"
626 JH/42 Bug 2545: Fix CHUNKING for all RCPT commands rejected. Previously we
627 carried on and emitted a BDAT command, even when PIPELINING was not
630 JH/43 Bug 2465: Fix taint-handling in dsearch lookup. Previously a nontainted
631 buffer was used for the filename, resulting in a trap when tainted
632 arguments (eg. $domain) were used.
634 JH/44 With OpenSSL 1.1.1 (onwards) disable renegotiation for TLS1.2 and below;
635 recommended to avoid a possible server-load attack. The feature can be
636 re-enabled via the openssl_options main cofiguration option.
638 JH/45 local_scan API: documented the current smtp_printf() call. This changed
639 for version 4.90 - adding a "more data" boolean to the arguments.
640 Bumped the ABI version number also, this having been missed previously;
641 release versions 4.90 to 4.92.3 inclusive were effectively broken in
642 respect of usage of smtp_printf() by either local_scan code or libraries
643 accessed via the ${dlfunc } expansion item. Both will need coding
644 adjustment for any calls to smtp_printf() to match the new function
645 signature; a FALSE value for the new argument is always safe.
647 JH/46 FreeBSD: fix use of the sendfile() syscall. The shim was not updating
648 the file-offset (which the Linux syscall does, and exim expects); this
649 resulted in an indefinite loop.
651 JH/47 ARC: fix crash in signing, triggered when a configuration error failed
652 to do ARC verification. The Authentication-Results: header line added
653 by the configuration then had no ARC item.
659 JH/01 Remove code calling the customisable local_scan function, unless a new
660 definition "HAVE_LOCAL_SCAN=yes" is present in the Local/Makefile.
662 JH/02 Bug 1007: Avoid doing logging from signal-handlers, as that can result in
663 non-signal-safe functions being used.
665 JH/03 Bug 2269: When presented with a received message having a stupidly large
666 number of DKIM-Signature headers, disable DKIM verification to avoid
667 a resource-consumption attack. The limit is set at twenty.
669 JH/04 Add variables $arc_domains, $arc_oldest_pass for ARC verify. Fix the
670 report of oldest_pass in ${authres } in consequence, and separate out
671 some descriptions of reasons for verification fail.
673 JH/05 Bug 2273: Cutthrough delivery left a window where the received messsage
674 files in the spool were present and unlocked. A queue-runner could spot
675 them, resulting in a duplicate delivery. Fix that by doing the unlock
676 after the unlink. Investigation by Tim Stewart. Take the opportunity to
677 add more error-checking on spoolfile handling while that code is being
680 PP/01 Refuse to open a spool data file (*-D) if it's a symlink.
681 No known attacks, no CVE, this is defensive hardening.
683 JH/06 Bug 2275: The MIME ACL unlocked the received message files early, and
684 a queue-runner could start a delivery while other operations were ongoing.
685 Cutthrough delivery was a common victim, resulting in duplicate delivery.
686 Found and investigated by Tim Stewart. Fix by using the open message data
687 file handle rather than opening another, and not locally closing it (which
688 releases a lock) for that case, while creating the temporary .eml format
689 file for the MIME ACL. Also applies to "regex" and "spam" ACL conditions.
691 JH/07 Bug 177: Make a random-recipient callout success visible in ACL, by setting
692 $sender_verify_failure/$recipient_verify_failure to "random".
694 JH/08 When generating a selfsigned cert, use serial number 1 since zero is not
697 JH/09 Bug 2274: Fix logging of cmdline args when starting in an unlinked cwd.
698 Previously this would segfault.
700 JH/10 Fix ARC signing for case when DKIM signing failed. Previously this would
703 JH/11 Bug 2264: Exim now only follows CNAME chains one step by default. We'd
704 like zero, since the resolver should be doing this for us, But we need one
705 as a CNAME but no MX presence gets the CNAME returned; we need to check
706 that doesn't point to an MX to declare it "no MX returned" rather than
707 "error, loop". A new main option is added so the older capability of
708 following some limited number of chain links is maintained.
710 JH/12 Add client-ip info to non-pass iprev ${authres } lines.
712 JH/13 For receent Openssl versions (1.1 onward) use modern generic protocol
713 methods. These should support TLS 1.3; they arrived with TLS 1.3 and the
714 now-deprecated earlier definitions used only specified the range up to TLS
715 1.2 (in the older-version library docs).
717 JH/14 Bug 2284: Fix DKIM signing for body lines starting with a pair of dots.
719 JH/15 Rework TLS client-side context management. Stop using a global, and
720 explicitly pass a context around. This enables future use of TLS for
721 connections to service-daemons (eg. malware scanning) while a client smtp
722 connection is using TLS; with cutthrough connections this is quite likely.
724 JH/16 Fix ARC verification to do AS checks in reverse order.
726 JH/17 Support a "tls" option on the ${readsocket } expansion item.
728 JH/18 Bug 2287: Fix the protocol name (eg utf8esmtp) for multiple messages
729 using the SMTPUTF8 option on their MAIL FROM commands, in one connection.
730 Previously the "utf8" would be re-prepended for every additional message.
732 JH/19 Reject MAIL FROM commands with SMTPUTF8 when the facility was not advertised.
733 Previously thery were accepted, resulting in issues when attempting to
734 forward messages to a non-supporting MTA.
736 PP/02 Let -n work with printing macros too, not just options.
738 JH/20 Bug 2296: Fix cutthrough for >1 address redirection. Previously only
739 one parent address was copied, and bogus data was used at delivery-logging
740 time. Either a crash (after delivery) or bogus log data could result.
741 Discovery and analysis by Tim Stewart.
743 PP/03 Make ${utf8clean:} expansion operator detect incomplete final character.
744 Previously if the string ended mid-character, we did not insert the
745 promised '?' replacement.
747 PP/04 Documentation: current string operators work on bytes, not codepoints.
749 JH/21 Change as many as possible of the global flags into one-bit bitfields; these
750 should pack well giving a smaller memory footprint so better caching and
751 therefore performance. Group the declarations where this can't be done so
752 that the byte-sized flag variables are not interspersed among pointer
753 variables, giving a better chance of good packing by the compiler.
755 JH/22 Bug 1896: Fix the envelope from for DMARC forensic reports to be possibly
756 non-null, to avoid issues with sites running BATV. Previously reports were
757 sent with an empty envelope sender so looked like bounces.
759 JH/23 Bug 2318: Fix the noerror command within filters. It wasn't working.
760 The ignore_error flag wasn't being returned from the filter subprocess so
761 was not set for later routers. Investigation and fix by Matthias Kurz.
763 JH/24 Bug 2310: Raise a msg:fail:internal event for each undelivered recipient,
764 and a msg:complete for the whole, when a message is manually removed using
765 -Mrm. Developement by Matthias Kurz, hacked on by JH.
767 JH/25 Avoid fixed-size buffers for pathnames in DB access. This required using
768 a "Gnu special" function, asprintf() in the DB utility binary builds; I
769 hope that is portable enough.
771 JH/26 Bug 2311: Fix DANE-TA verification under GnuTLS. Previously it was also
772 requiring a known-CA anchor certificate; make it now rely entirely on the
773 TLSA as an anchor. Checking the name on the leaf cert against the name
774 on the A-record for the host is still done for TA (but not for EE mode).
776 JH/27 Fix logging of proxy address. Previously, a pointless "PRX=[]:0" would be
777 included in delivery lines for non-proxied connections, when compiled with
778 SUPPORT_SOCKS and running with proxy logging enabled.
780 JH/28 Bug 2314: Fire msg:fail:delivery event even when error is being ignored.
781 Developement by Matthias Kurz, tweaked by JH. While in that bit of code,
782 move the existing event to fire before the normal logging of message
783 failure so that custom logging is bracketed by normal logging.
785 JH/29 Bug 2322: A "fail" command in a non-system filter (file) now fires the
786 msg:fail:internal event. Developement by Matthias Kurz.
788 JH/30 Bug 2329: Increase buffer size used for dns lookup from 2k, which was
789 far too small for todays use of crypto signatures stored there. Go all
790 the way to the max DNS message size of 64kB, even though this might be
791 overmuch for IOT constrained device use.
793 JH/31 Fix a bad use of a copy function, which could be used to pointlessly
794 copy a string over itself. The library routine is documented as not
795 supporting overlapping copies, and on MacOS it actually raised a SIGABRT.
797 JH/32 For main options check_spool_space and check_inode_space, where the
798 platform supports 64b integers, support more than the previous 2^31 kB
799 (i.e. more than 2 TB). Accept E, P and T multipliers in addition to
800 the previous G, M, k.
802 JH/33 Bug 2338: Fix the cyrus-sasl authenticator to fill in the
803 $authenticated_fail_id variable on authentication failure. Previously
806 JH/34 Increase RSA keysize of autogen selfsign cert from 1024 to 2048. RHEL 8.0
807 OpenSSL didn't want to use such a weak key. Do for GnuTLS also, and for
808 more-modern GnuTLS move from GNUTLS_SEC_PARAM_LOW to
809 GNUTLS_SEC_PARAM_MEDIUM.
811 JH/35 OpenSSL: fail the handshake when SNI processing hits a problem, server
812 side. Previously we would continue as if no SNI had been received.
814 JH/36 Harden the handling of string-lists. When a list consisted of a sole
815 "<" character, which should be a list-separator specification, we walked
816 off past the nul-terimation.
818 JH/37 Bug 2341: Send "message delayed" warning MDNs (restricted to external
819 causes) even when the retry time is not yet met. Previously they were
820 not, meaning that when (say) an account was over-quota and temp-rejecting,
821 and multiple senders' messages were queued, only one sender would get
822 notified on each configured delay_warning cycle.
824 JH/38 Bug 2351: Log failures to extract envelope addresses from message headers.
826 JH/39 OpenSSL: clear the error stack after an SSL_accept(). With anon-auth
827 cipher-suites, an error can be left on the stack even for a succeeding
828 accept; this results in impossible error messages when a later operation
831 AM/01 Bug 2359: GnuTLS: repeat lowlevel read and write operations while they
832 return error codes indicating retry. Under TLS1.3 this becomes required.
834 JH/40 Fix the feature-cache refresh for EXPERIMENTAL_PIPE_CONNECT. Previously
835 it only wrote the new authenticators, resulting in a lack of tracking of
836 peer changes of ESMTP extensions until the next cache flush.
838 JH/41 Fix the loop reading a message header line to check for integer overflow,
839 and more-often against header_maxsize. Previously a crafted message could
840 induce a crash of the recive process; now the message is cleanly rejected.
842 JH/42 Bug 2366: Fix the behaviour of the dkim_verify_signers option. It had
843 been totally disabled for all of 4.91. Discovery and fix by "Mad Alex".
849 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
850 When redis_servers is set to a list of > 1 element, and the Redis servers
851 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
852 case of MOVED into a DEFER case instead, thus moving the query onto the
853 next server in the list. For a cluster of N elements, all N servers must
854 be defined in redis_servers.
856 GF/02 Catch and remove uninitialized value warning in exiqsumm
857 Check for existence of @ARGV before looking at $ARGV[0]
859 JH/01 Replace the store_release() internal interface with store_newblock(),
860 which internalises the check required to safely use the old one, plus
861 the allocate and data copy operations duplicated in both (!) of the
862 extant use locations.
864 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
865 modifier. This matches the restriction on the commandline.
867 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
868 Previously only the last row was returned.
870 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
871 we assumed that tags in the header were well-formed, and parsed the
872 element content after inspecting only the first char of the tag.
873 Assumptions at that stage could crash the receive process on malformed
876 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
877 While running the DKIM ACL we operate on the Permanent memory pool so that
878 variables created with "set" persist to the DATA ACL. Also (at any time)
879 DNS lookups that fail create cache records using the Permanent pool. But
880 expansions release any allocations made on the current pool - so a dnsdb
881 lookup expansion done in the DKIM ACL releases the memory used for the
882 DNS negative-cache, and bad things result. Solution is to switch to the
883 Main pool for expansions.
884 While we're in that code, add checks on the DNS cache during store_reset,
885 active in the testsuite.
886 Problem spotted, and debugging aided, by Wolfgang Breyha.
888 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
889 When none of the hosts presented to a transport match an already-open
890 connection, close it and proceed with the list. Previously we would
891 queue the message. Spotted by Lena with Yahoo, probably involving
894 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
895 Previously a spurious "250 OK id=" response was appended to the proper
898 JH/08 The "support for" informational output now, which built with Content
899 Scanning support, has a line for the malware scanner interfaces compiled
900 in. Interface can be individually included or not at build time.
902 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
903 by the template makefile "src/EDITME". The "STREAM" support for an older
904 ClamAV interface method is removed.
906 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
907 rows affected is given instead).
909 JH/11 The runtime Berkeley DB library version is now additionally output by
910 "exim -d -bV". Previously only the compile-time version was shown.
912 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
913 SMTP connection. Previously, when one had more recipients than the
914 first, an abortive onward connection was made. Move to full support for
915 multiple onward connections in sequence, handling cutthrough connection
916 for all multi-message initiating connections.
918 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
919 routers. Previously, a multi-recipient message would fail to match the
920 onward-connection opened for the first recipient, and cause its closure.
922 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
923 a timeout on read on a GnuTLS initiating connection, resulting in the
924 initiating connection being dropped. This mattered most when the callout
925 was marked defer_ok. Fix to keep the two timeout-detection methods
928 JH/15 Relax results from ACL control request to enable cutthrough, in
929 unsupported situations, from error to silently (except under debug)
930 ignoring. This covers use with PRDR, frozen messages, queue-only and
933 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
935 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
936 metadata, resulting in a crash in free().
938 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
939 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
940 Broken also in d185889f4, with init system revamp.
942 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
943 Previously we abruptly closed the connection after reading a malware-
944 found indication; now we go on to read the "scan ok" response line,
947 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
948 ACL. Previously, a crash would result.
950 JH/19 Speed up macro lookups during configuration file read, by skipping non-
951 macro text after a replacement (previously it was only once per line) and
952 by skipping builtin macros when searching for an uppercase lead character.
954 JH/20 DANE support moved from Experimental to mainline. The Makefile control
955 for the build is renamed.
957 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
958 was allocated for every new TLS startup, meaning one per message. Fix
959 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
961 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
962 reported the original. Fix to report (as far as possible) the ACL
963 result replacing the original.
965 JH/23 Fix memory leak during multi-message connections using STARTTLS under
966 OpenSSL. Certificate information is loaded for every new TLS startup,
967 and the resources needed to be freed.
969 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
971 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
974 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
975 DATA response info to the (existing) per-recipient response info for
976 the "C=" log element. It can have useful tracking info from the
977 destination system. Patch from Simon Arlott.
979 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
980 length value. Previously this would segfault.
982 HS/02 Support Avast multiline protoocol, this allows passing flags to
983 newer versions of the scanner.
985 JH/28 Ensure that variables possibly set during message acceptance are marked
986 dead before release of memory in the daemon loop. This stops complaints
987 about them when the debug_store option is enabled. Discovered specifically
988 for sender_rate_period, but applies to a whole set of variables.
989 Do the same for the queue-runner and queue-list loops, for variables set
990 from spool message files. Do the same for the SMTP per-message loop, for
991 certain variables indirectly set in ACL operations.
993 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
994 as a multi-recipient message from a mailinglist manager). The coding had
995 an arbitrary cutoff number of characters while checking for more input;
996 enforced by writing a NUL into the buffer. This corrupted long / fast
997 input. The problem was exposed more widely when more pipelineing of SMTP
998 responses was introduced, and one Exim system was feeding another.
999 The symptom is log complaints of SMTP syntax error (NUL chars) on the
1000 receiving system, and refused recipients seen by the sending system
1001 (propating to people being dropped from mailing lists).
1002 Discovered and pinpointed by David Carter.
1004 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
1005 replaced by the ${authresults } expansion.
1007 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
1009 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
1010 allows proper process termination in container environments.
1012 JH/32 Bug 2258: Fix spool_wireformat in combination with LMTP transport.
1013 Previously the "final dot" had a newline after it; ensure it is CR,LF.
1015 JH/33 SPF: remove support for the "spf" ACL condition outcome values "err_temp"
1016 and "err_perm", deprecated since 4.83 when the RFC-defined words
1017 "temperror" and "permerror" were introduced.
1019 JH/34 Re-introduce enforcement of no cutthrough delivery on transports having
1020 transport-filters or DKIM-signing. The restriction was lost in the
1021 consolidation of verify-callout and delivery SMTP handling.
1022 Extend the restriction to also cover ARC-signing.
1024 JH/35 Cutthrough: for a final-dot response timeout (and nonunderstood responses)
1025 in defer=pass mode supply a 450 to the initiator. Previously the message
1028 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
1029 tls_require_ciphers is used as before.
1031 HS/03 Malware Avast: Better match the Avast multiline protocol. Add
1032 "pass_unscanned". Only tmpfails from the scanner are written to
1033 the paniclog, as they may require admin intervention (permission
1034 denied, license issues). Other scanner errors (like decompression
1035 bombs) do not cause a paniclog entry.
1037 JH/36 Fix reinitialisation of DKIM logging variable between messages.
1038 Previously it was possible to log spurious information in receive log
1041 JH/37 Bug 2255: Revert the disable of the OpenSSL session caching. This
1042 triggered odd behaviour from Outlook Express clients.
1044 PP/03 Add util/renew-opendmarc-tlds.sh script for safe renewal of public
1047 JH/38 DKIM: accept Ed25519 pubkeys in SubjectPublicKeyInfo-wrapped form,
1048 since the IETF WG has not yet settled on that versus the original
1049 "bare" representation.
1051 JH/39 Fix syslog logging for syslog_timestamp=no and log_selector +millisec.
1052 Previously the millisecond value corrupted the output.
1053 Fix also for syslog_pid=no and log_selector +pid, for which the pid
1054 corrupted the output.
1060 JH/01 Rework error string handling in TLS interface so that the caller in
1061 more cases is responsible for logging. This permits library-sourced
1062 string to be attached to addresses during delivery, and collapses
1063 pairs of long lines into single ones.
1065 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
1066 during configuration. Wildcards are allowed and expanded.
1068 JH/02 Rework error string handling in DKIM to pass more info back to callers.
1069 This permits better logging.
1071 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
1072 do not close it down and have the child transport start it up again on
1073 the passed-on TCP connection. Instead, proxy the child (and any
1074 subsequent ones) for TLS via a unix-domain socket channel. Logging is
1075 affected: the continued delivery log lines do not have any DNSSEC, TLS
1076 Certificate or OCSP information. TLS cipher information is still logged.
1078 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
1079 identical IP addresses on different listening ports. Will also affect
1082 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
1083 add noisy ifdef guards to special-case this sillyness.
1084 Patch from Bernd Kuhls.
1086 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
1087 than 255 are no longer allowed.
1089 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
1090 Disable the session-cache too, which might reduce our load. Since we
1091 currrectly use a new context for every connection, both as server and
1092 client, there is no benefit for these.
1093 GnuTLS appears to not support tickets server-side by default (we don't
1094 call gnutls_session_ticket_enable_server()) but client side is enabled
1095 by default on recent versions (3.1.3 +) unless the PFS priority string
1098 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
1099 <https://reproducible-builds.org/specs/source-date-epoch/>.
1101 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
1102 the check for any unsuccessful recipients did not notice the limit, and
1103 erroneously found still-pending ones.
1105 JH/08 Pipeline CHUNKING command and data together, on kernels that support
1106 MSG_MORE. Only in-clear (not on TLS connections).
1108 JH/09 Avoid using a temporary file during transport using dkim. Unless a
1109 transport-filter is involved we can buffer the headers in memory for
1110 creating the signature, and read the spool data file once for the
1111 signature and again for transmission.
1113 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
1114 4.77 as the kernel support then wasn't solid, having issues in 64bit
1115 mode. Now, it's been long enough. Add support for FreeBSD also.
1117 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
1118 case where the routing stage had gathered several addresses to send to
1119 a host before calling the transport for the first, we previously failed
1120 to close down TLS in the old transport process before passing the TCP
1121 connection to the new process. The new one sent a STARTTLS command
1122 which naturally failed, giving a failed delivery and bloating the retry
1123 database. Investigation and fix prototype from Wolfgang Breyha.
1125 JH/12 Fix check on SMTP command input synchronisation. Previously there were
1126 false-negatives in the check that the sender had not preempted a response
1127 or prompt from Exim (running as a server), due to that code's lack of
1128 awareness of the SMTP input buffering.
1130 PP/04 Add commandline_checks_require_admin option.
1131 Exim drops privileges sanely, various checks such as -be aren't a
1132 security problem, as long as you trust local users with access to their
1133 own account. When invoked by services which pass untrusted data to
1134 Exim, this might be an issue. Set this option in main configuration
1135 AND make fixes to the calling application, such as using `--` to stop
1138 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
1139 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
1140 responses to those, into a single TLS record each way (this usually means
1141 a single packet). As a side issue, smtp_enforce_sync now works on TLS
1144 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
1145 affects you only if you're dancing at the edge of the param size limits.
1146 If you are, and this message makes sense to you, then: raise the
1147 configured limit or use OpenSSL 1.1. Nothing we can do for older
1150 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
1151 cmdline element to get the documented default one. Previously it was
1154 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
1155 are unsuable for non-message-specific reasons - eg. connection timeout,
1156 banner-time rejection.
1158 JH/16 Fix logging of delivery remote port, when specified by router, under
1161 PP/06 Repair manualroute's ability to take options in any order, even if one
1162 is the name of a transport.
1165 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
1167 JH/17 Change the list-building routines interface to use the expanding-string
1168 triplet model, for better allocation and copying behaviour.
1170 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
1171 Previously it was constructed the first time a possibly-matching string
1172 was met in the configuration file input during startup; now it is done
1175 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
1176 compatible one, to avoid the (poorly documented) possibility of a config
1177 file in the working directory redirecting the DB files, possibly correpting
1178 some existing file. CVE-2017-10140 assigned for BDB.
1180 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
1181 cache-hot. Previously, although the result was properly cached, the
1182 initial verify call returned a defer.
1184 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
1185 the main verify for receipient in uncached-mode.
1187 JH/22 Retire historical build files to an "unsupported" subdir. These are
1188 defined as "ones for which we have no current evidence of testing".
1190 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
1191 if present. Previously it was ignored.
1193 JH/24 Start using specified-initialisers in C structure init coding. This is
1194 a C99 feature (it's 2017, so now considered safe).
1196 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
1197 if was a fixed-sized field and bitmask ops via macros; it is now more
1200 PP/07 GitHub PR 56: Apply MariaDB build fix.
1201 Patch provided by Jaroslav Škarvada.
1203 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
1204 during Coverity cleanups [4.87 JH/47]
1205 Diagnosis and fix provided by Michael Fischer v. Mollard.
1207 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
1208 the right size to place the terminating semicolon on its own folded
1209 line, the header hash was calculated to an incorrect value thanks to
1210 the (relaxed) space the fold became.
1212 HS/02 Fix Bug 2130: large writes from the transport subprocess were chunked
1213 and confused the parent.
1215 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
1216 which could crash as a result. This could lead to undeliverable messages.
1218 JH/28 Logging: "next input sent too soon" now shows where input was truncated
1221 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
1222 matters on fast-turnover and PID-randomising systems, which were getting
1223 out-of-order delivery.
1225 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
1226 a possibly-overlapping copy. The symptom was that "Remote host closed
1227 connection in response to HELO" was logged instead of the actual 4xx
1230 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
1231 Previously only that bufferd was discarded, resulting in SYMTP command
1234 JH/32 DKIM: when a message has multiple signatures matching an identity given
1235 in dkim_verify_signers, run the dkim acl once for each. Previously only
1236 one run was done. Bug 2189.
1238 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
1239 "panic the current process" to "deliberately defer". The panic log is
1240 still written with the problem list name; the mail and reject logs now
1241 get a temp-reject line for the message that was being handled, saying
1242 something like "domains check lookup or other defer". The SMTP 451
1243 message is still "Temporary local problem".
1245 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
1246 A crafted sequence of BDAT commands could result in in-use memory beeing
1247 freed. CVE-2017-16943.
1249 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
1250 from SMTP input. Previously it was always done; now only done for DATA
1251 and not BDAT commands. CVE-2017-16944.
1253 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
1254 to the message (such as an overlong header line). Previously this was
1255 not done and we did not exit BDAT mode. Followon from the previous item
1256 though a different problem.
1262 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
1263 than -2003 did; needs libidn2 in addition to libidn.
1265 JH/02 The path option on a pipe transport is now expanded before use.
1267 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
1268 Patch provided by "Björn", documentation fix added too.
1270 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
1271 missing a wire-to-host endian conversion.
1273 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
1274 close after a BDAT command line could be taken as a following command,
1275 giving a synch failure. Fix by only checking for synch immediately
1276 before acknowledging the chunk.
1278 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
1279 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
1280 macro. Patches provided by Josh Soref.
1282 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
1283 Previously we did not; the RFC seems ambiguous and VRFY is not listed
1284 by IANA as a service extension. However, John Klensin suggests that we
1287 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
1288 the dkim code may be unix-mode line endings rather than smtp wire-format
1289 CRLF, so prepend a CR to any bare LF.
1291 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
1292 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
1294 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
1295 the first were themselves being wrongly included in the feed into dkim
1296 processing; with most chunk sizes in use this resulted in an incorrect
1297 body hash calculated value.
1299 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
1300 DKIM signature block, for verification. Although advised against by
1301 standards it is specifically not ruled illegal.
1303 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
1305 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
1306 missing a body hash (the bh= tag).
1308 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
1309 It seems that HAProxy sends the Proxy Protocol information in clear and
1310 only then does a TLS startup, so do the same.
1312 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
1313 TCP connections (such as for Spamd) unless the daemon successfully set
1314 Fast Open mode on its listening sockets. This fixes breakage seen on
1315 too-old kernels or those not configured for Fast Open, at the cost of
1316 requiring both directions being enabled for TFO, and TFO never being used
1317 by non-daemon-related Exim processes.
1319 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
1320 endings, at least on the first header line. Try to canonify any that get
1321 past that check, despite the cost.
1323 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
1324 now limited to an arbitrary five deep, while parsing addresses with the
1325 strip_excess_angle_brackets option enabled.
1327 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
1328 instead leave the unprompted TLS handshake in socket buffer for the
1329 TLS library to consume.
1331 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
1333 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
1335 JH/16 Drop variables when they go out of scope. Memory management drops a whole
1336 region in one operation, for speed, and this leaves assigned pointers
1337 dangling. Add checks run only under the testsuite which checks all
1338 variables at a store-reset and panics on a dangling pointer; add code
1339 explicitly nulling out all the variables discovered. Fixes one known
1340 bug: a transport crash, where a dangling pointer for $sending_ip_address
1341 originally assigned in a verify callout, is re-used.
1343 PP/06 Drop '.' from @INC in various Perl scripts.
1345 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
1347 PP/08 Reduce a number of compilation warnings under clang; building with
1348 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
1349 should be warning-free.
1351 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
1353 HS/01 Fix portability problems introduced by PP/08 for platforms where
1354 realloc(NULL) is not equivalent to malloc() [SunOS et al].
1356 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
1357 chunk. This allows us to accept broken chunked messages. We need a more
1358 general solution here.
1360 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
1361 already-broken messages in the queue.
1363 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
1365 JH/19 Fix reference counting bug in routing-generated-address tracking.
1371 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
1372 supports it and a size is available (ie. the sending peer gave us one).
1374 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
1375 years of being deprecated). The replacements are the ACLs
1376 acl_smtp_mime and acl_not_smtp_mime.
1378 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
1379 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
1380 or even an in-clear connection were permitted. Now, if the host lookup
1381 was dnssec and dane was requested then the host is only used if the
1382 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
1383 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
1384 if one fails this test.
1385 This means that a poorly-configured remote DNS will make it incommunicado;
1386 but it protects against a DNS-interception attack on it.
1388 JH/04 Bug 1810: make continued-use of an open smtp transport connection
1389 non-noisy when a race steals the message being considered.
1391 JH/05 If main configuration option tls_certificate is unset, generate a
1392 self-signed certificate for inbound TLS connections.
1394 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
1395 in rewrites and routers.
1397 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
1398 and logged a warning sing 4.83; now they are a configuration file error.
1400 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
1401 (lacking @domain). Apply the same qualification processing as RCPT.
1403 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
1405 JH/10 Support ${sha256:} applied to a string (as well as the previous
1408 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
1409 a cutthrough deliver is pending, as we always want to make a connection.
1410 This also avoids re-routing the message when later placing the cutthrough
1411 connection after a verify cache hit.
1412 Do not update it with the verify result either.
1414 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
1415 when routing results in more than one destination address.
1417 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
1418 signing (which inhibits the cutthrough capability). Previously only
1419 the presence of an option was tested; now an expansion evaluating as
1420 empty is permissible (obviously it should depend only on data available
1421 when the cutthrough connection is made).
1423 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
1424 the relevant preceding SMTP command did not note the pipelining mode.
1426 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
1427 Previously they were not counted.
1429 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
1430 as one having no matching records. Previously we deferred the message
1431 that needed the lookup.
1433 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
1434 distinguished as "(=".
1436 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
1437 for missing MX records. Previously it only worked for missing A records.
1439 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
1441 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
1442 after the data-go-ahead and data-ack. Patch from Jason Betts.
1444 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
1445 even for a "none" policy. Patch from Tony Meyer.
1447 JH/22 Fix continued use of a connection for further deliveries. If a port was
1448 specified by a router, it must also match for the delivery to be
1451 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
1452 When one of the recipients of a message was unsuitable for the connection
1453 (has no matching addresses), we lost track of needing to mark it
1454 deferred. As a result mail would be lost.
1456 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
1458 JH/25 Decoding ACL controls is now done using a binary search; the source code
1459 takes up less space and should be simpler to maintain. Merge the ACL
1460 condition decode tables also, with similar effect.
1462 JH/26 Fix problem with one_time used on a redirect router which returned the
1463 parent address unchanged. A retry would see the parent address marked as
1464 delivered, so not attempt the (identical) child. As a result mail would
1467 JH/27 Fix a possible security hole, wherein a process operating with the Exim
1468 UID can gain a root shell. Credit to http://www.halfdog.net/ for
1469 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
1472 JH/28 Enable {spool,log} filesystem space and inode checks as default.
1473 Main config options check_{log,spool}_{inodes,space} are now
1474 100 inodes, 10MB unless set otherwise in the configuration.
1476 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
1477 Previously it only applied to the main-section connection policy
1480 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
1482 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
1483 by me. Added RFC7919 DH primes as an alternative.
1485 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
1486 are not in the system include path.
1488 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
1489 GnuTLS, when a session startup failed (eg because the client disconnected)
1490 Exim did stdio operations after fclose. This was exposed by a recent
1491 change which nulled out the file handle after the fclose.
1493 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
1494 signed directly by the cert-signing cert, rather than an intermediate
1495 OCSP-signing cert. This is the model used by LetsEncrypt.
1497 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
1499 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
1500 an incoming connection.
1502 HS/02 Bug 1802: Do not half-close the connection after sending a request
1505 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
1506 fallback to "prime256v1".
1508 JH/34 SECURITY: Use proper copy of DATA command in error message.
1509 Could leak key material. Remotely exploitable. CVE-2016-9963.
1515 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
1516 and 3.4.4 - once the server is enabled to respond to an OCSP request
1517 it does even when not requested, resulting in a stapling non-aware
1518 client dropping the TLS connection.
1520 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
1521 support variable-length bit vectors. No functional change.
1523 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
1524 The I= interface field on outgoing lines is now after the H= remote
1525 host field, same as incoming lines. There is a separate
1526 outgoing_interface log selector which allows you to disable the
1529 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
1530 If not running log_selector +smtp_connection the mainlog would be held
1531 open indefinitely after a "too many connections" event, including to a
1532 deleted file after a log rotate. Leave the per net connection logging
1533 leaving it open for efficiency as that will be quickly detected by the
1534 check on the next write.
1536 HS/01 Bug 1671: Fix post transport crash.
1537 Processing the wait-<transport> messages could crash the delivery
1538 process if the message IDs didn't exist for some reason. When
1539 using 'split_spool_directory=yes' the construction of the spool
1540 file name failed already, exposing the same netto behaviour.
1542 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
1543 mime_regex ACL conditions.
1545 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
1546 to DSN fail messages (bounces): remote IP, remote greeting, remote response
1547 to HELO, local diagnostic string.
1549 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
1550 log line to debug. Even when configured with a tls authenticator many
1551 client connections are expected to not authenticate in this way, so
1552 an authenticate fail is not an error.
1554 HS/02 Add the Exim version string to the process info. This way exiwhat
1555 gives some more detail about the running daemon.
1557 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
1558 matter for fast-change records such as DNSBLs.
1560 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
1561 retry record, even if constant. There may be multiple transports with
1562 different interface settings and the retry behaviour needs to be kept
1565 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
1567 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
1569 JH/10 Bug 840: fix log_defer_output option of pipe transport
1571 JH/11 Bug 830: use same host for all RCPTS of a message, even under
1572 hosts_randomize. This matters a lot when combined with mua_wrapper.
1574 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
1575 ${quote_pgsql:<string>} operator.
1577 JH/13 Bug 1708: avoid misaligned access in cached lookup.
1579 JH/14 Change header file name for freeradius-client. Relevant if compiling
1580 with Radius support; from the Gentoo tree and checked under Fedora.
1582 JH/15 Bug 1712: Introduce $prdr_requested flag variable
1584 JH/16 Bug 1714: Permit an empty string as expansion result for transport
1585 option transport_filter, meaning no filtering.
1587 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
1589 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
1590 defaults to "*" (all hosts). The variable is now available when not built
1591 with TLS, default unset, mainly to enable keeping the testsuite sane.
1592 If a server certificate is not supplied (via tls_certificate) an error is
1593 logged, and clients will find TLS connections fail on startup. Presumably
1594 they will retry in-clear.
1595 Packagers of Exim are strongly encouraged to create a server certificate
1596 at installation time.
1598 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
1599 with the $config_file variable.
1601 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
1602 in transport context, after the attempt, and per-recipient. The latter type
1603 is per host attempted. The event data is the error message, and the errno
1604 information encodes the lookup type (A vs. MX) used for the (first) host,
1605 and the trailing two digits of the smtp 4xx response.
1607 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
1608 to write to mainlog (or rejectlog, paniclog) in the window between file
1609 creation and permissions/ownership being changed. Particularly affects
1610 installations where exicyclog is run as root, rather than exim user;
1611 result is that the running daemon panics and dies.
1613 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
1615 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
1616 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
1617 "pri" and "weight". Note that the previous implicit priority given by the
1618 list order is no longer honoured.
1620 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
1621 for DKIM processing.
1623 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
1624 by defining SUPPORT_SOCKS.
1626 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
1627 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
1628 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
1629 variables are renamed to proxy_{local,external}_{address,port}.
1631 JH/27 Move Internationalisation support from Experimental to mainline, enabled
1632 for a build by defining SUPPORT_I18N
1634 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
1635 of the query string, and make ${quote_redis:} do that quoting.
1637 JH/29 Move Events support from Experimental to mainline, enabled by default
1638 and removable for a build by defining DISABLE_EVENT.
1640 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
1642 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
1643 cached by the daemon.
1645 JH/32 Move Redis support from Experimental to mainline, enabled for a build
1646 by defining LOOKUP_REDIS. The libhiredis library is required.
1648 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
1649 keys are given for lookup.
1651 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
1652 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
1653 only supported when built with TLS support. The PolarSSL SHA routines
1654 are still used when the TLS library is too old for convenient support.
1656 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
1657 openssl_options), for security. OpenSSL forces this from version 1.1.0
1658 server-side so match that on older versions.
1660 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
1661 allocation for $value could be released as the expansion processing
1662 concluded, but leaving the global pointer active for it.
1664 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
1665 and to use the domains and local_parts ACL conditions.
1667 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
1668 incorrectly not doubled on cutthrough transmission, hence seen as a
1669 body-termination at the receiving system - resulting in truncated mails.
1670 Commonly the sender saw a TCP-level error, and retransmitted the message
1671 via the normal store-and-forward channel. This could result in duplicates
1672 received - but deduplicating mailstores were liable to retain only the
1673 initial truncated version.
1675 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
1677 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
1679 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
1680 we're in there, support oversigning also; bug 1309.
1682 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
1684 HS/04 Add support for keep_environment and add_environment options.
1686 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
1687 either intentional arithmetic overflow during PRNG, or testing config-
1690 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
1691 delivery resulted in actual delivery. Cancel cutthrough before DATA
1694 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
1695 rejects a recipient: pass the reject to the originator.
1697 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
1698 Many were false-positives and ignorable, but it's worth fixing the
1701 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
1702 for the new environment-manipulation done at startup. Move the routines
1703 from being local to tls.c to being global via the os.c file.
1705 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
1706 an extract embedded as result-arg for a map, the first arg for extract
1707 is unavailable so we cannot tell if this is a numbered or keyed
1708 extraction. Accept either.
1714 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
1717 JH/02 The smtp transport option "multi_domain" is now expanded.
1719 JH/03 The smtp transport now requests PRDR by default, if the server offers
1722 JH/04 Certificate name checking on server certificates, when exim is a client,
1723 is now done by default. The transport option tls_verify_cert_hostnames
1724 can be used to disable this per-host. The build option
1725 EXPERIMENTAL_CERTNAMES is withdrawn.
1727 JH/05 The value of the tls_verify_certificates smtp transport and main options
1728 default to the word "system" to access the system default CA bundle.
1729 For GnuTLS, only version 3.0.20 or later.
1731 JH/06 Verification of the server certificate for a TLS connection is now tried
1732 (but not required) by default. The verification status is now logged by
1733 default, for both outbound TLS and client-certificate supplying inbound
1736 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
1739 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
1740 Status Notification (bounce) messages are now MIME format per RFC 3464.
1741 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
1742 under the control of the dsn_advertise_hosts option, and routers may
1743 have a dsn_lasthop option.
1745 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
1746 default, modifiable by a malware= option. The list separator for
1747 the options can now be changed in the usual way. Bug 68.
1749 JH/10 The smtp_receive_timeout main option is now expanded before use.
1751 JH/11 The incoming_interface log option now also enables logging of the
1752 local interface on delivery outgoing connections.
1754 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
1755 if the interface and destination host and port all match.
1757 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
1760 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
1761 Patch from Andrew Lewis.
1763 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
1764 now supports optional time-restrictions, weighting, and priority
1765 modifiers per server. Patch originally by <rommer@active.by>.
1767 JH/16 The spamd_address main option now supports a mixed list of local
1768 and remote servers. Remote servers can be IPv6 addresses, and
1769 specify a port-range.
1771 JH/17 Bug 68: The spamd_address main option now supports an optional
1772 timeout value per server.
1774 JH/18 Bug 1581: Router and transport options headers_add/remove can
1775 now have the list separator specified.
1777 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
1780 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
1783 JH/21 Support for the A6 type of dns record is withdrawn.
1785 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
1786 rather than the verbs used.
1788 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
1789 from 255 to 1024 chars.
1791 JH/24 Verification callouts now attempt to use TLS by default.
1793 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
1794 are generic router options now. The defaults didn't change.
1796 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
1797 Original patch from Alexander Shikoff, worked over by JH.
1799 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
1800 files if ZCAT_COMMAND is not executable.
1802 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
1804 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
1806 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
1807 Normally benign, it bites when the pair was led to by a CNAME;
1808 modern usage is to not canonicalize the domain to a CNAME target
1809 (and we were inconsistent anyway for A-only vs AAAA+A).
1811 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
1813 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
1814 when evaluating $sender_host_dnssec.
1816 JH/31 Check the HELO verification lookup for DNSSEC, adding new
1817 $sender_helo_dnssec variable.
1819 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
1821 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
1823 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
1825 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
1826 documented as working, but never had. Support all but $spam_report.
1828 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
1829 added for tls authenticator.
1831 HS/03 Add perl_taintmode main config option
1837 TL/01 When running the test suite, the README says that variables such as
1838 no_msglog_check are global and can be placed anywhere in a specific
1839 test's script, however it was observed that placement needed to be near
1840 the beginning for it to behave that way. Changed the runtest perl
1841 script to read through the entire script once to detect and set these
1842 variables, reset to the beginning of the script, and then run through
1843 the script parsing/test process like normal.
1845 TL/02 The BSD's have an arc4random API. One of the functions to induce
1846 adding randomness was arc4random_stir(), but it has been removed in
1847 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
1848 function when detected.
1850 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
1851 cause callback expansion.
1853 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
1854 syntax errors in an expansion can be treated as a string instead of
1855 logging or causing an error, due to the internal use of bool_lax
1856 instead of bool when processing it.
1858 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
1859 server certificates when making smtp deliveries.
1861 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
1863 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
1865 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1867 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1868 Merged patch from Sebastian Wiedenroth.
1870 JH/05 Fix results-pipe from transport process. Several recipients, combined
1871 with certificate use, exposed issues where response data items split
1872 over buffer boundaries were not parsed properly. This eventually
1873 resulted in duplicates being sent. This issue only became common enough
1874 to notice due to the introduction of connection certificate information,
1875 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1877 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1878 size buffer was used, resulting in syntax errors when an expansion
1881 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1882 version 3.3.6 or later.
1884 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1885 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1886 both become "event_action", the variables become $event_name, $event_data
1887 and $event_defer_errno. There is a new variable $verify_mode, usable in
1888 routers, transports and related events. The tls:cert event is now also
1889 raised for inbound connections, if the main configuration event_action
1892 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1893 early OCSP support, but no stapling (appears to be less than 1.0.0).
1895 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1896 server certificate names available under the smtp transport option
1897 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1900 JH/10 Time-related extraction expansions from certificates now use the main
1901 option "timezone" setting for output formatting, and are consistent
1902 between OpenSSL and GnuTLS compilations. Bug 1541.
1904 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1905 encoded parameter in the incoming message. Bug 1558.
1907 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1908 include certificate info, eximon was claiming there were spoolfile
1911 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1913 JH/14 Log delivery-related information more consistently, using the sequence
1914 "H=<name> [<ip>]" wherever possible.
1916 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1917 are problematic for Debian distribution, omit them from the release
1920 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1922 JH/16 Fix string representation of time values on 64bit time_t architectures.
1925 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1926 output list separator was used.
1931 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1932 checkers that were complaining about end of non-void function with no
1935 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1936 This was a regression introduced in 4.83 by another bugfix.
1938 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1940 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1941 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1947 TF/01 Correctly close the server side of TLS when forking for delivery.
1949 When a message was received over SMTP with TLS, Exim failed to clear up
1950 the incoming connection properly after forking off the child process to
1951 deliver the message. In some situations the subsequent outgoing
1952 delivery connection happened to have the same fd number as the incoming
1953 connection previously had. Exim would try to use TLS and fail, logging
1954 a "Bad file descriptor" error.
1956 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1957 utilities have not been installed.
1959 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1960 temporary space as the ACL may create new global variables.
1962 TL/01 LDAP support uses per connection or global context settings, depending
1963 upon the detected version of the libraries at build time.
1965 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1966 to extract and use the src ip:port in logging and expansions as if it
1967 were a direct connection from the outside internet. PPv2 support was
1968 updated based on HAProxy spec change in May 2014.
1970 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1972 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1973 Properly escape header and check for NULL return.
1975 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1978 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1980 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1981 characters in header names, implemented as a verify condition.
1982 Contributed by Michael Fischer v. Mollard.
1984 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1985 results permerror and temperror. Previous values are deprecated but
1986 still accepted. In a future release, err_perm and err_temp will be
1987 completely removed, which will be a backward incompatibility if the
1988 ACL tests for either of these two old results. Patch contributed by
1989 user bes-internal on the mailing list.
1991 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1993 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1994 selectors, in both main and reject logs.
1996 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1999 JH/07 Add malware type "sock" for talking to simple daemon.
2001 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
2003 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
2004 routers/transports under cutthrough routing.
2006 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
2007 numbers. Touch up "bool" conditional to keep the same definition.
2009 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
2011 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
2013 JH/12 Expand items in router/transport headers_add or headers_remove lists
2014 individually rather than the list as a whole. Bug 1452.
2016 Required for reasonable handling of multiple headers_ options when
2017 they may be empty; requires that headers_remove items with embedded
2018 colons must have them doubled (or the list-separator changed).
2020 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
2021 view the policy declared in the DMARC record. Currently, $dmarc_status
2022 is a combined value of both the record presence and the result of the
2025 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
2027 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
2028 dnslookup router and the smtp transport (applying to the forward
2031 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
2032 of ldap servers used for a specific lookup. Patch provided by Heiko
2035 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
2036 New variable $lookup_dnssec_authenticated for observability.
2038 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
2039 Patch submitted by Lars Timman.
2041 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
2043 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
2044 Requires trusted mode and valid format message id, aborts otherwise.
2045 Patch contributed by Heiko Schlichting.
2047 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
2048 certextract with support for various fields. Bug 1358.
2050 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
2051 is requested by default, modifiable by smtp transport option
2054 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
2055 operate on certificate variables to give certificate fingerprints
2056 Also new ${sha256:cert_variable}.
2058 JH/23 The PRDR feature is moved from being Experimental into the mainline.
2060 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
2061 Christian Aistleitner.
2063 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
2065 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
2066 file. Patch from Wolfgang Breyha.
2068 JH/25 Expand the coverage of the delivery $host and $host_address to
2069 client authenticators run in verify callout. Bug 1476.
2071 JH/26 Port service names are now accepted for tls_on_connect_ports, to
2072 align with daemon_smtp_ports. Bug 72.
2074 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
2075 support and error reporting did not work properly.
2077 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
2078 and is readable. Patch from Andrew Colin Kissa.
2080 TL/14 Enhance documentation of ${run expansion and how it parses the
2081 commandline after expansion, particularly in the case when an
2082 unquoted variable expansion results in an empty value.
2084 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
2086 PP/02 Fix internal collision of T_APL on systems which support RFC3123
2087 by renaming away from it. Addresses GH issue 15, reported by
2090 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
2092 TL/15 SECURITY: prevent double expansion in math comparison functions
2093 (can expand unsanitized data). Not remotely exploitable.
2100 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
2102 PP/02 Make -n do something, by making it not do something.
2103 When combined with -bP, the name of an option is not output.
2105 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
2108 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
2109 $sender_host_name and config options to manage this, and basic check
2112 PP/05 DSCP support for outbound connections and control modifier for inbound.
2114 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
2115 (Only plugin which currently uses this is kerberos4, which nobody should
2116 be using, but we should make it available and other future plugins might
2117 conceivably use it, even though it would break NAT; stuff *should* be
2118 using channel bindings instead).
2120 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
2121 name; added for Sendmail compatibility; requires admin caller.
2122 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
2123 just ignore it); requires trusted caller.
2124 Also parse but ignore: -Ac -Am -X<logfile>
2127 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
2129 TL/02 Add +smtp_confirmation as a default logging option.
2131 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
2132 Patch by Magnus Holmgren from 2007-02-20.
2134 TL/04 Bugzilla 1281 - Spec typo.
2135 Bugzilla 1283 - Spec typo.
2136 Bugzilla 1290 - Spec grammar fixes.
2138 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
2140 TL/06 Add Experimental DMARC support using libopendmarc libraries.
2142 TL/07 Fix an out of order global option causing a segfault. Reported to dev
2143 mailing list by by Dmitry Isaikin.
2145 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
2147 JH/02 Support "G" suffix to numbers in ${if comparisons.
2149 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
2151 NM/01 Bugzilla 1197 - Spec typo
2152 Bugzilla 1196 - Spec examples corrections
2154 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
2156 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
2157 gnutls_enable_pkcs11, but renamed to more accurately indicate its
2160 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
2161 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
2163 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
2164 "acl {{name}{arg}...}", and optional args on acl condition
2167 JH/05 Permit multiple router/transport headers_add/remove lines.
2169 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
2171 JH/07 Avoid using a waiting database for a single-message-only transport.
2172 Performance patch from Paul Fisher. Bugzilla 1262.
2174 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
2177 JH/09 Add $headers_added variable, with content from use of ACL modifier
2178 add_header (but not yet added to the message). Bugzilla 199.
2180 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
2181 Pulled from Bugzilla 817 by Wolfgang Breyha.
2183 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
2185 (nb: this is the same fix as in Exim 4.80.1)
2187 JH/11 Add A= logging on delivery lines, and a client_set_id option on
2190 JH/12 Add optional authenticated_sender logging to A= and a log_selector
2193 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
2195 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
2196 advertise SMTP AUTH mechanism to us, instead of a generic
2197 protocol violation error. Also, make Exim more robust to bad
2198 data from the Dovecot auth socket.
2200 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
2202 When a queue runner is handling a message, Exim first routes the
2203 recipient addresses, during which it prunes them based on the retry
2204 hints database. After that it attempts to deliver the message to
2205 any remaining recipients. It then updates the hints database using
2208 So if a recipient address works intermittently, it can get repeatedly
2209 deferred at routing time. The retry hints record remains fresh so the
2210 address never reaches the final cutoff time.
2212 This is a fairly common occurrence when a user is bumping up against
2213 their storage quota. Exim had some logic in its local delivery code
2214 to deal with this. However it did not apply to per-recipient defers
2215 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
2217 This change adds a proper retry rule check during routing so that the
2218 final cutoff time is checked against the message's age. We only do
2219 this check if there is an address retry record and there is not a
2220 domain retry record; this implies that previous attempts to handle
2221 the address had the retry_use_local_parts option turned on. We use
2222 this as an approximation for the destination being like a local
2223 delivery, as in LMTP.
2225 I suspect this new check makes the old local delivery cutoff check
2226 redundant, but I have not verified this so I left the code in place.
2228 TF/02 Correct gecos expansion when From: is a prefix of the username.
2230 Test 0254 submits a message to Exim with the header
2234 When I ran the test suite under the user fanf2, Exim expanded
2235 the header to contain my full name, whereas it should have added
2236 a Resent-Sender: header. It erroneously treats any prefix of the
2237 username as equal to the username.
2239 This change corrects that bug.
2241 GF/01 DCC debug and logging tidyup
2242 Error conditions log to paniclog rather than rejectlog.
2243 Debug lines prefixed by "DCC: " to remove any ambiguity.
2245 TF/03 Avoid unnecessary rebuilds of lookup-related code.
2247 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
2248 Bug spotted by Jeremy Harris; was flawed since initial commit.
2249 Would have resulted in OCSP responses post-SNI triggering an Exim
2250 NULL dereference and crash.
2252 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
2254 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
2255 Bug detection, analysis and fix by Samuel Thibault.
2256 Bugzilla 1331, Debian bug #698092.
2258 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
2260 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
2261 Server implementation by Todd Lyons, client by JH.
2262 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
2263 config variable "prdr_enable" controls whether the server
2264 advertises the facility. If the client requests PRDR a new
2265 acl_data_smtp_prdr ACL is called once for each recipient, after
2266 the body content is received and before the acl_smtp_data ACL.
2267 The client is controlled by both of: a hosts_try_prdr option
2268 on the smtp transport, and the server advertisement.
2269 Default client logging of deliveries and rejections involving
2270 PRDR are flagged with the string "PRDR".
2272 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
2273 fclose(). Diagnosis by Todd Lyons.
2275 PP/17 Update configure.default to handle IPv6 localhost better.
2276 Patch by Alain Williams (plus minor tweaks).
2279 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
2280 This is now consistent with GnuTLS, and is now documented: the
2281 previous undocumented portable approach to treating the option as
2282 unset was to force an expansion failure. That still works, and
2283 an empty string is now equivalent.
2285 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
2286 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
2287 not performing validation itself.
2289 PP/20 Added force_command boolean option to pipe transport.
2290 Patch from Nick Koston, of cPanel Inc.
2292 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
2295 TF/04 Added udpsend ACL modifier and hexquote expansion operator
2297 PP/21 Fix eximon continuous updating with timestamped log-files.
2298 Broken in a format-string cleanup in 4.80, missed when I repaired the
2299 other false fix of the same issue.
2300 Report and fix from Heiko Schlichting.
2303 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
2304 Report from Prashanth Katuri.
2306 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
2307 It's SecureTransport, so affects any MacOS clients which use the
2308 system-integrated TLS libraries, including email clients.
2310 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
2311 using a MIME ACL for non-SMTP local injection.
2312 Report and assistance in diagnosis by Warren Baker.
2314 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
2316 JH/16 Fix comparisons for 64b. Bugzilla 1385.
2318 TL/09 Add expansion variable $authenticated_fail_id to keep track of
2319 last id that failed so it may be referenced in subsequent ACL's.
2321 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
2324 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
2325 ldap library initialization, allowing self-signed CA's to be
2326 used. Also properly sets require_cert option later in code by
2327 using NULL (global ldap config) instead of ldap handle (per
2328 session). Bug diagnosis and testing by alxgomz.
2330 TL/12 Enhanced documentation in the ratelimit.pl script provided in
2331 the src/util/ subdirectory.
2333 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
2334 renamed to Transport Post Delivery Action by Jeremy Harris, as
2337 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
2338 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
2339 redis_servers = needs to be configured which will be used by the redis
2340 lookup. Patch from Warren Baker, of The Packet Hub.
2342 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
2344 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
2345 hostname or reverse DNS when processing a host list. Used suggestions
2346 from multiple comments on this bug.
2348 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
2350 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
2351 Missed a few lines, added it to make the runtest require no keyboard
2354 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
2355 contains upper case chars. Make router use caseful_local_part.
2357 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
2358 support when GnuTLS has been built with p11-kit.
2364 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
2366 This, or similar/improved, will also be change PP/11 of 4.82.
2372 PP/01 Handle short writes when writing local log-files.
2373 In practice, only affects FreeBSD (8 onwards).
2374 Bugzilla 1053, with thanks to Dmitry Isaikin.
2376 NM/01 Bugzilla 949 - Documentation tweak
2378 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
2381 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
2383 PP/02 Implemented gsasl authenticator.
2385 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
2387 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
2388 `pkg-config foo` for cflags/libs.
2390 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
2391 with rest of GSASL and with heimdal_gssapi.
2393 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
2394 `pkg-config foo` for cflags/libs for the TLS implementation.
2396 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
2397 properties get this fed in as external SSF. A number of robustness
2398 and debugging improvements to the cyrus_sasl authenticator.
2400 PP/08 cyrus_sasl server now expands the server_realm option.
2402 PP/09 Bugzilla 1214 - Log authentication information in reject log.
2403 Patch by Jeremy Harris.
2405 PP/10 Added dbmjz lookup type.
2407 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
2409 PP/12 MAIL args handles TAB as well as SP, for better interop with
2410 non-compliant senders.
2411 Analysis and variant patch by Todd Lyons.
2413 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
2414 Bug report from Lars Müller <lars@samba.org> (via SUSE),
2415 Patch from Dirk Mueller <dmueller@suse.com>
2417 PP/13 tls_peerdn now print-escaped for spool files.
2418 Observed some $tls_peerdn in wild which contained \n, which resulted
2419 in spool file corruption.
2421 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
2422 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
2423 or write after TLS renegotiation, which otherwise led to messages
2426 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
2427 as a tracking header (ie: a signed header comes before the signature).
2428 Patch from Wolfgang Breyha.
2430 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
2431 comma-sep list; embedded commas doubled.
2433 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
2435 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
2437 Report and patch from Dmitry Banschikov.
2439 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
2440 Removed SSL_clear() after SSL_new() which led to protocol negotiation
2441 failures. We appear to now support TLS1.1+ with Exim.
2443 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
2444 lets Exim select keys and certificates based upon TLS SNI from client.
2445 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
2446 before an outbound SMTP session. New log_selector, +tls_sni.
2448 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
2449 NULL dereference. Report and patch from Alun Jones.
2451 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
2452 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
2453 Not seeing resolver debug output on NetBSD, but suspect this is a
2454 resolver implementation change.
2456 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
2457 Left warnings. Added "eximon gdb" invocation mode.
2459 PP/21 Defaulting "accept_8bitmime" to true, not false.
2461 PP/22 Added -bw for inetd wait mode support.
2463 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
2464 locate the relevant includes and libraries. Made this the default.
2466 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
2467 Bugzilla 1246, report and most of solution from Tomasz Kusy.
2469 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
2470 This may cause build issues on older platforms.
2472 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
2473 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
2474 gnutls_require_mac & gnutls_require_protocols (no longer supported).
2475 Added SNI support via GnuTLS too.
2476 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
2478 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
2480 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
2481 Applied second patch from Janne, implementing suggestion to default
2482 multiple-strings-in-record handling to match SPF spec.
2484 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
2486 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
2487 read-only, out of scope).
2488 Patch from Wolfgang Breyha, report from Stuart Northfield.
2490 PP/29 Fix three issues highlighted by clang analyser static analysis.
2491 Only crash-plausible issue would require the Cambridge-specific
2492 iplookup router and a misconfiguration.
2493 Report from Marcin Mirosław.
2495 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
2497 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
2498 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
2499 As part of this, removing so much warning spew let me fix some minor
2500 real issues in debug logging.
2502 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
2503 assignment on my part. Fixed.
2505 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
2506 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
2507 Janne Snabb (who went above and beyond: thank you).
2509 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
2510 string otherwise requires a connection and a bunch more work and it's
2511 relatively easy to get wrong. Should also expose TLS library linkage
2514 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
2515 64-bit ${eval} (JH/03).
2517 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
2518 GNU libc to support some of the 64-bit stuff, should not lead to
2519 conflicts. Defined before os.h is pulled in, so if a given platform
2520 needs to override this, it can.
2522 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
2523 protection layer was required, which is not implemented.
2524 Bugzilla 1254, patch from Wolfgang Breyha.
2526 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
2527 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
2528 tls_dhparam take prime identifiers. Also unbreak combination of
2529 OpenSSL+DH_params+TLSSNI.
2531 PP/39 Disable SSLv2 by default in OpenSSL support.
2537 PP/01 Solaris build fix for Oracle's LDAP libraries.
2538 Bugzilla 1109, patch from Stephen Usher.
2540 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
2542 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
2545 TF/02 Fix a couple more cases where we did not log the error message
2546 when unlink() failed. See also change 4.74-TF/03.
2548 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
2549 lock up or crash if it happened to be inside a call to libc when it
2550 got a SIGUSR1 from exiwhat.
2552 The SIGUSR1 handler appends the current process status to the process
2553 log which is later printed by exiwhat. It used to use the general
2554 purpose logging code to do this, but several functions it calls are
2555 not safe for signals.
2557 The new output code in the SIGUSR1 handler is specific to the process
2558 log, and simple enough that it's easy to inspect for signal safety.
2559 Removing some special cases also simplifies the general logging code.
2560 Removing the spurious timestamps from the process log simplifies
2563 TF/04 Improved ratelimit ACL condition.
2565 The /noupdate option has been deprecated in favour of /readonly which
2566 has clearer semantics. The /leaky, /strict, and /readonly update modes
2567 are mutually exclusive. The update mode is no longer included in the
2568 database key; it just determines when the database is updated. (This
2569 means that when you upgrade Exim will forget old rate measurements.)
2571 Exim now checks that the per_* options are used with an update mode that
2572 makes sense for the current ACL. For example, when Exim is processing a
2573 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
2574 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
2575 must specify per_mail/readonly. If you omit the update mode it defaults to
2576 /leaky where that makes sense (as before) or /readonly where required.
2578 The /noupdate option is now undocumented but still supported for
2579 backwards compatibility. It is equivalent to /readonly except that in
2580 ACLs where /readonly is required you may specify /leaky/noupdate or
2581 /strict/noupdate which are treated the same as /readonly.
2583 A useful new feature is the /count= option. This is a generalization
2584 of the per_byte option, so that you can measure the throughput of other
2585 aggregate values. For example, the per_byte option is now equivalent
2586 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
2588 The per_rcpt option has been generalized using the /count= mechanism
2589 (though it's more complicated than the per_byte equivalence). When it is
2590 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
2591 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
2592 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
2593 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
2594 Note that using per_rcpt with a non-readonly update mode in more than
2595 one ACL will cause the recipients to be double-counted. (The per_mail
2596 and per_byte options don't have this problem.)
2598 The handling of very low rates has changed slightly. If the computed rate
2599 is less than the event's count (usually one) then this event is the first
2600 after a long gap. In this case the rate is set to the same as this event's
2601 count, so that the first message of a spam run is counted properly.
2603 The major new feature is a mechanism for counting the rate of unique
2604 events. The new per_addr option counts the number of different
2605 recipients that someone has sent messages to in the last time period. It
2606 behaves like per_rcpt if all the recipient addresses are different, but
2607 duplicate recipient addresses do not increase the measured rate. Like
2608 the /count= option this is a general mechanism, so the per_addr option
2609 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
2610 example, measure the rate that a client uses different sender addresses
2611 with the options per_mail/unique=$sender_address. There are further
2612 details in the main documentation.
2614 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
2616 TF/06 Removed a few PCRE remnants.
2618 TF/07 Automatically extract Exim's version number from tags in the git
2619 repository when doing development or release builds.
2621 PP/02 Raise smtp_cmd_buffer_size to 16kB.
2622 Bugzilla 879. Patch from Paul Fisher.
2624 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
2625 Heavily based on revision 40f9a89a from Simon Arlott's tree.
2628 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
2630 PP/05 Variable $av_failed, true if the AV scanner deferred.
2631 Bugzilla 1078. Patch from John Horne.
2633 PP/06 Stop make process more reliably on build failure.
2634 Bugzilla 1087. Patch from Heiko Schlittermann.
2636 PP/07 Make maildir_use_size_file an _expandable_ boolean.
2637 Bugzilla 1089. Patch from Heiko Schlittermann.
2639 PP/08 Handle ${run} returning more data than OS pipe buffer size.
2640 Bugzilla 1131. Patch from Holger Weiß.
2642 PP/09 Handle IPv6 addresses with SPF.
2643 Bugzilla 860. Patch from Wolfgang Breyha.
2645 PP/10 GnuTLS: support TLS 1.2 & 1.1.
2647 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
2650 PP/11 match_* no longer expand right-hand-side by default.
2651 New compile-time build option, EXPAND_LISTMATCH_RHS.
2652 New expansion conditions, "inlist", "inlisti".
2654 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
2656 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
2658 PP/14 fix log_write() format string regression from TF/03.
2659 Bugzilla 1152. Patch from Dmitry Isaikin.
2665 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
2667 PP/02 Harmonised TLS library version reporting; only show if debugging.
2668 Layout now matches that introduced for other libraries in 4.74 PP/03.
2670 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
2672 PP/04 New "dns_use_edns0" global option.
2674 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
2677 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
2678 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
2680 TK/01 Updated PolarSSL code to 0.14.2.
2681 Bugzilla 1097. Patch from Andreas Metzler.
2683 PP/07 Catch divide-by-zero in ${eval:...}.
2684 Fixes bugzilla 1102.
2686 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
2689 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
2690 format-string attack -- SECURITY: remote arbitrary code execution.
2692 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
2693 time unintentionally subject to list matching rules, letting the header
2694 cause arbitrary Exim lookups (of items which can occur in lists, *not*
2695 arbitrary string expansion). This allowed for information disclosure.
2697 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
2698 INT_MIN/-1 -- value coerced to INT_MAX.
2704 NM/01 Workaround for PCRE version dependency in version reporting
2707 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
2708 This fixes portability to compilers other than gcc, notably
2709 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
2711 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
2712 makefiles for portability to HP-UX and POSIX correctness.
2714 PP/01 Permit LOOKUP_foo enabling on the make command-line.
2715 Also via indented variable definition in the Makefile.
2716 (Debugging by Oliver Heesakkers).
2718 PP/02 Restore caching of spamd results with expanded spamd_address.
2719 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
2721 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
2722 Improves build reliability. Fix from: Frank Elsner
2724 NM/02 Fix wide character breakage in the rfc2047 coding
2725 Fixes bug 1064. Patch from Andrey N. Oktyabrski
2727 NM/03 Allow underscore in dnslist lookups
2728 Fixes bug 1026. Patch from Graeme Fowler
2730 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
2731 Code patches from Adam Ciarcinski of NetBSD.
2733 NM/04 Fixed exiqgrep to cope with mailq missing size issue
2736 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
2737 is logged, to avoid truncation. Patch from John Horne.
2739 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
2740 Patch from Jakob Hirsch.
2742 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
2743 SQL string expansion failure details.
2744 Patch from Andrey Oktyabrski.
2746 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
2747 Patch from Simon Arlott.
2749 PP/09 New lookups functionality failed to compile on old gcc which rejects
2750 extern declarations in function scope.
2751 Patch from Oliver Fleischmann
2753 PP/10 Use sig_atomic_t for flags set from signal handlers.
2754 Check getgroups() return and improve debugging.
2755 Fixed developed for diagnosis in bug 927 (which turned out to be
2758 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
2759 Patch from Mark Zealey.
2761 PP/12 Bugzilla 1056: Improved spamd server selection.
2762 Patch from Mark Zealey.
2764 PP/13 Bugzilla 1086: Deal with maildir quota file races.
2765 Based on patch from Heiko Schlittermann.
2767 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
2768 Patch from Uwe Doering, sign-off by Michael Haardt.
2770 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
2771 variable declaration deep within a block. Bug and patch from
2774 PP/15 lookups-Makefile IRIX compatibility coercion.
2776 PP/16 Make DISABLE_DKIM build knob functional.
2778 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
2779 Patch by Simon Arlott
2781 TF/03 Fix valgrind.h portability to C89 compilers that do not support
2782 variable argument macros. Our copy now differs from upstream.
2788 TF/01 Failure to get a lock on a hints database can have serious
2789 consequences so log it to the panic log.
2791 TF/02 Log LMTP confirmation messages in the same way as SMTP,
2792 controlled using the smtp_confirmation log selector.
2794 TF/03 Include the error message when we fail to unlink a spool file.
2796 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
2797 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
2798 for maintaining out-of-tree patches for some time.
2800 PP/01 Bugzilla 139: Documentation and portability issues.
2801 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
2802 Handle per-OS dynamic-module compilation flags.
2804 PP/02 Let /dev/null have normal permissions.
2805 The 4.73 fixes were a little too stringent and complained about the
2806 permissions on /dev/null. Exempt it from some checks.
2807 Reported by Andreas M. Kirchwitz.
2809 PP/03 Report version information for many libraries, including
2810 Exim version information for dynamically loaded libraries. Created
2811 version.h, now support a version extension string for distributors
2812 who patch heavily. Dynamic module ABI change.
2814 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
2815 privilege escalation vulnerability whereby the Exim run-time user
2816 can cause root to append content of the attacker's choosing to
2819 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
2822 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
2823 If dropping privileges for untrusted macros, we disabled normal logging
2824 on the basis that it would fail; for the Exim run-time user, this is not
2825 the case, and it resulted in successful deliveries going unlogged.
2826 Fixed. Reported by Andreas Metzler.
2832 PP/01 Date: & Message-Id: revert to normally being appended to a message,
2833 only prepend for the Resent-* case. Fixes regression introduced in
2834 Exim 4.70 by NM/22 for Bugzilla 607.
2836 PP/02 Include check_rfc2047_length in configure.default because we're seeing
2837 increasing numbers of administrators be bitten by this.
2839 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
2841 PP/03 Bugzilla 994: added openssl_options main configuration option.
2843 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
2845 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
2847 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
2849 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
2850 without a peer certificate, leading to a segfault because of an
2851 assumption that peers always have certificates. Be a little more
2852 paranoid. Problem reported by Martin Tscholak.
2854 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
2855 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
2856 NB: ClamAV planning to remove STREAM in "middle of 2010".
2857 CL also introduces -bmalware, various -d+acl logging additions and
2858 more caution in buffer sizes.
2860 PP/09 Implemented reverse_ip expansion operator.
2862 PP/10 Bugzilla 937: provide a "debug" ACL control.
2864 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2866 PP/12 Bugzilla 973: Implement --version.
2868 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2870 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2872 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2874 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2875 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2876 ignore trailing whitespace.
2878 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2880 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2883 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2884 Notification from Dr Andrew Aitchison.
2886 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2887 ExtendedDetectionInfo response format.
2888 Notification from John Horne.
2890 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2893 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2894 XSL and documented dependency on system catalogs, with examples of how
2897 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2900 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2901 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2902 configuration file which is writeable by the Exim user or group.
2904 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2905 of configuration files to cover files specified with the -C option if
2906 they are going to be used with root privileges, not just the default
2909 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2910 option (effectively making it always true).
2912 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2913 files to be used while preserving root privileges.
2915 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2916 that rogue child processes cannot use them.
2918 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2919 run-time user, instead of root.
2921 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2922 Exim run-time user without dropping privileges.
2924 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2925 result string, instead of calling string_vformat() twice with the same
2928 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2929 for other users. Others should always drop root privileges if they use
2930 -C on the command line, even for a whitelisted configure file.
2932 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2934 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2940 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2941 $data_path, and $header_path variables; fixed documentation bugs and
2944 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2945 exipick to access non-standard spools, including the "frozen" queue
2948 NM/01 Bugzilla 965: Support mysql stored procedures.
2949 Patch from Alain Williams
2951 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2953 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2954 Patch from Andreas Metzler
2956 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2957 Patch from Kirill Miazine
2959 NM/05 Bugzilla 671: Added umask to procmail example.
2961 JJ/03 installed exipick 20100323.0, fixing doc bug
2963 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2964 directory. Notification and patch from Dan Rosenberg.
2966 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2968 TK/02 Improve log output when DKIM signing operation fails.
2970 MH/01 Treat the transport option dkim_domain as a colon separated
2971 list, not as a single string, and sign the message with each element,
2972 omitting multiple occurences of the same signer.
2974 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2975 Bugzilla 985, 986. Patch by Simon Arlott
2977 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2978 Patch by Simon Arlott
2980 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2981 MBX locking. Notification from Dan Rosenberg.
2987 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2989 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2991 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2993 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2995 NM/04 Bugzilla 915: Flag broken perl installation during build.
3001 TK/01 Added patch by Johannes Berg that expands the main option
3002 "spamd_address" if it starts with a dollar sign.
3004 TK/02 Write list of recipients to X-Envelope-Sender header when building
3005 the mbox-format spool file for content scanning (suggested by Jakob
3008 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
3009 (http://www.dcc-servers.net/) support via dccifd. Activated by
3010 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
3012 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
3013 by Mark Daniel Reidel <mr@df.eu>.
3015 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
3016 When building exim an external PCRE library is now needed -
3017 PCRE is a system library on the majority of modern systems.
3018 See entry on PCRE_LIBS in EDITME file.
3020 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
3021 conversation. Added nologin parameter to request.
3022 Patch contributed by Kirill Miazine.
3024 TF/01 Do not log submission mode rewrites if they do not change the address.
3026 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
3028 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
3029 log files in place. Contributed by Roberto Lima.
3031 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
3033 TF/03 Bugzilla 615: When checking the local_parts router precondition
3034 after a local_part_suffix or local_part_prefix option, Exim now
3035 does not use the address's named list lookup cache, since this
3036 contains cached lookups for the whole local part.
3038 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
3039 Robert Millan. Documentation is in experimental-spec.txt.
3041 TF/04 Bugzilla 668: Fix parallel build (make -j).
3043 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
3045 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
3046 Patch provided by Jan Srzednicki.
3048 TF/05 Leading white space used to be stripped from $spam_report which
3049 wrecked the formatting. Now it is preserved.
3051 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
3052 that they are available at delivery time.
3054 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
3056 TF/08 TLS error reporting now respects the incoming_interface and
3057 incoming_port log selectors.
3059 TF/09 Produce a more useful error message if an SMTP transport's hosts
3060 setting expands to an empty string.
3062 NM/06 Bugzilla 744: EXPN did not work under TLS.
3063 Patch provided by Phil Pennock.
3065 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
3066 Patch provided by Richard Godbee.
3068 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
3069 acl_smtp_notquit, added index entry.
3071 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
3072 Patch provided by Eugene Bujak.
3074 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
3075 accept(). Patch provided by Maxim Dounin.
3077 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
3078 Patch provided by Phil Pennock.
3080 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
3082 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
3083 Patch provided by Brad "anomie" Jorsch.
3085 NM/14 Bugzilla 622: Added timeout setting to transport filter.
3086 Patch provided by Dean Brooks.
3088 TK/05 Add native DKIM support (does not depend on external libraries).
3090 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
3091 Patch provided by Graeme Fowler.
3093 NM/16 Bugzilla 851: Documentation example syntax fix.
3095 NM/17 Changed NOTICE file to remove references to embedded PCRE.
3097 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
3100 NM/19 Bugzilla 745: TLS version reporting.
3101 Patch provided by Phil Pennock.
3103 NM/20 Bugzilla 167: bool: condition support.
3104 Patch provided by Phil Pennock.
3106 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
3107 clients. Patch provided by Phil Pennock.
3109 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
3110 Patch provided by Brad "anomie" Jorsch.
3112 NM/23 Bugzilla 687: Fix misparses in eximstats.
3113 Patch provided by Heiko Schlittermann.
3115 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
3116 Patch provided by Heiko Schlittermann.
3118 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
3119 plus update to original patch.
3121 NM/26 Bugzilla 799: Documentation correction for ratelimit.
3123 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
3124 Patch provided by David Brownlee.
3126 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
3128 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
3130 NM/30 Bugzilla 888: TLS documentation bugfixes.
3132 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
3134 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
3135 Unlike the original bugzilla I have changed all shell scripts in src tree.
3137 NM/33 Bugzilla 898: Transport filter timeout fix.
3138 Patch by Todd Rinaldo.
3140 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
3141 Patch by Serge Demonchaux.
3143 NM/35 Bugzilla 39: Base64 decode bug fixes.
3144 Patch by Jakob Hirsch.
3146 NM/36 Bugzilla 909: Correct connect() call in dcc code.
3148 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
3150 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
3152 NM/39 Bugzilla 911: Fixed MakeLinks build script.
3158 TK/01 Add preliminary DKIM support. Currently requires a forked version of
3159 ALT-N's libdkim that I have put here:
3160 http://duncanthrax.net/exim-experimental/
3162 Note to Michael Haardt: I had to rename some vars in sieve.c. They
3163 were called 'true' and it seems that C99 defines that as a reserved
3164 keyword to be used with 'bool' variable types. That means you could
3165 not include C99-style headers which use bools without triggering
3166 build errors in sieve.c.
3168 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
3169 as mailq or other aliases. Changed the --help handling significantly
3170 to do whats expected. exim_usage() emits usage/help information.
3172 SC/01 Added the -bylocaldomain option to eximstats.
3174 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
3176 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
3178 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
3184 PH/01 Another patch from the Sieve maintainer.
3186 PH/02 When an IPv6 address is converted to a string for single-key lookup
3187 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
3188 dots are used instead of colons so that keys in lsearch files need not
3189 contain colons. This was done some time before quoting was made available
3190 in lsearch files. However, iplsearch files do require colons in IPv6 keys
3191 (notated using the quote facility) so as to distinguish them from IPv4
3192 keys. This meant that lookups for IP addresses in host lists did not work
3193 for iplsearch lookups.
3195 This has been fixed by arranging for IPv6 addresses to be expressed with
3196 colons if the lookup type is iplsearch. This is not incompatible, because
3197 previously such lookups could never work.
3199 The situation is now rather anomalous, since one *can* have colons in
3200 ordinary lsearch keys. However, making the change in all cases is
3201 incompatible and would probably break a number of configurations.
3203 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
3206 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
3207 conversion specification without a maximum field width, thereby enabling
3208 a rogue spamd server to cause a buffer overflow. While nobody in their
3209 right mind would setup Exim to query an untrusted spamd server, an
3210 attacker that gains access to a server running spamd could potentially
3211 exploit this vulnerability to run arbitrary code as the Exim user.
3213 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
3214 $primary_hostname instead of what libspf2 thinks the hosts name is.
3216 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
3217 a directory entry by the name of the lookup key. Previously, if a
3218 symlink pointed to a non-existing file or a file in a directory that
3219 Exim lacked permissions to read, a lookup for a key matching that
3220 symlink would fail. Now it is enough that a matching directory entry
3221 exists, symlink or not. (Bugzilla 503.)
3223 PH/03 The body_linecount and body_zerocount variables are now exported in the
3226 PH/04 Added the $dnslist_matched variable.
3228 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
3229 This means they are set thereafter only if the connection becomes
3232 PH/06 Added the client_condition to authenticators so that some can be skipped
3233 by clients under certain conditions.
3235 PH/07 The error message for a badly-placed control=no_multiline_responses left
3236 "_responses" off the end of the name.
3238 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
3240 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
3241 (without spaces) instead of just copying the configuration text.
3243 PH/10 Added the /noupdate option to the ratelimit ACL condition.
3245 PH/11 Added $max_received_linelength.
3247 PH/12 Added +ignore_defer and +include_defer to host lists.
3249 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
3250 way in which PCRE > 7.0 is built.
3252 PH/14 Implemented queue_only_load_latch.
3254 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
3255 MAIL command. The effect was to mangle the value on 64-bit systems.
3257 PH/16 Another patch from the Sieve maintainer.
3259 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
3261 PH/18 If a system quota error occurred while trying to create the file for
3262 a maildir delivery, the message "Mailbox is full" was not appended to the
3263 bounce if the delivery eventually timed out. Change 4.67/27 below applied
3264 only to a quota excession during the actual writing of the file.
3266 PH/19 It seems that peer DN values may contain newlines (and other non-printing
3267 characters?) which causes problems in log lines. The DN values are now
3268 passed through string_printing() before being added to log lines.
3270 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
3271 and InterBase are left for another time.)
3273 PH/21 Added message_body_newlines option.
3275 PH/22 Guard against possible overflow in moan_check_errorcopy().
3277 PH/23 POSIX allows open() to be a macro; guard against that.
3279 PH/24 If the recipient of an error message contained an @ in the local part
3280 (suitably quoted, of course), incorrect values were put in $domain and
3281 $local_part during the evaluation of errors_copy.
3287 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
3288 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
3291 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
3292 issue a MAIL command.
3294 PH/02 In an ACL statement such as
3296 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
3298 if a client was not listed at all, or was listed with a value other than
3299 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
3300 the condition was not true (as it should be), so access was not denied.
3301 The bug was that the ! inversion was incorrectly passed on to the second
3302 item. This has been fixed.
3304 PH/03 Added additional dnslists conditions == and =& which are different from
3305 = and & when the dns lookup returns more than one IP address.
3307 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
3308 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
3310 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
3311 FSYNC, which compiles an option called disable_fsync that allows for
3312 bypassing fsync(). The documentation is heavily laced with warnings.
3314 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
3316 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
3317 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
3318 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
3319 including adding "make clean"; (3) Added -fPIC when compiling the test
3320 dynamically loaded module, to get rid of a warning.
3322 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
3323 message fails, move_frozen_messages = true and ignore_bounce_errors_after
3324 = 0s. The bug is otherwise harmless.
3326 PH/07 There was a bug in the dovecot authenticator such that the value of
3327 $auth1 could be overwritten, and so not correctly preserved, after a
3328 successful authentication. This usually meant that the value preserved by
3329 the server_setid option was incorrect.
3331 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
3333 PH/09 Installed PCRE release 7.0.
3335 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
3336 run for batched SMTP input. It is now run at the start of every message
3337 in the batch. While fixing this I discovered that the process information
3338 (output by running exiwhat) was not always getting set for -bs and -bS
3339 input. This is fixed, and it now also says "batched" for BSMTP.
3341 PH/11 Added control=no_pipelining.
3343 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
3344 patch, slightly modified), and move the expansion of helo_data till after
3345 the connection is made in the smtp transport (so it can use these
3348 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
3350 PH/14 Added log_selector = +pid.
3352 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
3354 PH/16 Add ${if forany and ${if forall.
3356 PH/17 Added dsn_from option to vary the From: line in DSNs.
3358 PH/18 Flush SMTP output before performing a callout, unless control =
3359 no_callout_flush is set.
3361 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
3362 was true (the default) a successful delivery failed to delete the retry
3363 item, thus causing premature timeout of the address. The bug is now
3366 PH/20 Added hosts_avoid_pipelining to the smtp transport.
3368 PH/21 Long custom messages for fakedefer and fakereject are now split up
3369 into multiline responses in the same way that messages for "deny" and
3370 other ACL rejections are.
3372 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
3373 with slight modification.
3375 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
3376 draft, changing the syntax and factoring some duplicate code".
3378 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
3379 for deliveries of the second and subsequent messages over the same SMTP
3382 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
3383 ${reduce, with only minor "tidies".
3385 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
3387 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
3388 expansion side effects.
3390 PH/27 When a message times out after an over-quota error from an Exim-imposed
3391 quota, the bounce message says "mailbox is full". This message was not
3392 being given when it was a system quota that was exceeded. It now should
3395 MH/03 Made $recipients available in local_scan(). local_scan() already has
3396 better access to the recipient list through recipients_list[], but
3397 $recipients can be useful in postmaster-provided expansion strings.
3399 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
3400 in the case of a MAIL command with additional options following the
3401 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
3402 were accidentally chopped off.
3404 PH/29 SMTP synchronization checks are implemented when a command is read -
3405 there is a check that no more input is waiting when there shouldn't be
3406 any. However, for some commands, a delay in an ACL can mean that it is
3407 some time before the response is written. In this time, more input might
3408 arrive, invalidly. So now there are extra checks after an ACL has run for
3409 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
3410 pipelining has not been advertised.
3412 PH/30 MH's patch to allow iscntrl() characters to be list separators.
3414 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
3415 returned in the SMTP response when smtp_return_error_details was false.
3416 This has been fixed.
3418 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
3419 instead of the C I/O that was originally supplied, because problems were
3420 reported on Solaris.
3422 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
3423 Exim which did not show up earlier: it was assuming that a call to
3424 SSL_CTX_set_info_callback() might give an error value. In fact, there is
3425 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
3426 was a macro that became an assignment, so it seemed to work. This has
3427 changed to a proper function call with a void return, hence the compile
3428 error. Exim's code has been fixed.
3430 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
3433 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
3435 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
3437 PH/37 If a message is not accepted after it has had an id assigned (e.g.
3438 because it turns out to be too big or there is a timeout) there is no
3439 "Completed" line in the log. When some messages of this type were
3440 selected by exigrep, they were listed as "not completed". Others were
3441 picked up by some special patterns. I have improved the selection
3442 criteria to be more general.
3444 PH/38 The host_find_failed option in the manualroute router can now be set
3445 to "ignore", to completely ignore a host whose IP address cannot be
3446 found. If all hosts are ignored, the behaviour is controlled by the new
3447 host_all_ignored option.
3449 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
3450 homing or because of multiple MX records with /mx) generated more than
3451 one IP address, and the following item turned out to be the local host,
3452 all the secondary addresses of the first item were incorrectly removed
3453 from the list, along with the local host and any following hosts (which
3454 is what is supposed to happen).
3456 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
3457 whoever called Exim into the -H file. In the case of the daemon it was
3458 behaving confusingly. When first started, it used values for whoever
3459 started the daemon, but after a SIGHUP it used the Exim user (because it
3460 calls itself on a restart). I have changed the code so that it now always
3463 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
3464 message are rejected with the same error (e.g. no authentication or bad
3465 sender address), and a DATA command is nevertheless sent (as can happen
3466 with PIPELINING or a stupid MUA), the error message that was given to the
3467 RCPT commands is included in the rejection of the DATA command. This is
3468 intended to be helpful for MUAs that show only the final error to their
3471 PH/42 Another patch from the Sieve maintainer.
3473 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
3474 Eximstats - Fixed some broken HTML links and added missing column headers
3476 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
3477 columns for Rejects, Temp Rejects, Ham, and Spam rows.
3479 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
3481 PH/43 Yet another patch from the Sieve maintainer.
3483 PH/44 I found a way to check for a TCP/IP connection going away before sending
3484 the response to the final '.' that terminates a message, but only in the
3485 case where the client has not sent further data following the '.'
3486 (unfortunately, this is allowed). However, in many cases there won't be
3487 any further data because there won't be any more messages to send. A call
3488 to select() can be used: if it shows that the input is "ready", there is
3489 either input waiting, or the socket has been closed. An attempt to read
3490 the next input character can distinguish the two cases. Previously, Exim
3491 would have sent an OK response which the client would never have see.
3492 This could lead to message repetition. This fix should cure that, at
3493 least in a lot of common cases.
3495 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
3496 advertised in response to EHLO.
3502 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
3503 fixed by 4.65/MH/01 (is this a record?) are fixed:
3505 (i) An empty string was always treated as zero by the numeric comparison
3506 operators. This behaviour has been restored.
3508 (ii) It is documented that the numeric comparison operators always treat
3509 their arguments as decimal numbers. This was broken in that numbers
3510 starting with 0 were being interpreted as octal.
3512 While fixing these problems I realized that there was another issue that
3513 hadn't been noticed. Values of message_size_limit (both the global option
3514 and the transport option) were treated as octal if they started with 0.
3515 The documentation was vague. These values are now always treated as
3516 decimal, and I will make that clear in the documentation.
3522 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
3523 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
3526 MH/01 Don't check that the operands of numeric comparison operators are
3527 integers when their expansion is in "skipping" mode (fixes bug
3528 introduced by 4.64-PH/07).
3530 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
3531 child addresses, Exim now panics and dies. Previously, because the count
3532 is held in a short int, deliveries were likely to be lost. As such a
3533 large number of recipients for a single message is ridiculous
3534 (performance will be very, very poor), I have chosen to impose a limit
3535 rather than extend the field.
3541 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
3542 leftover -K file (the existence of which was triggered by #402).
3543 While we were at it, introduced process PID as part of the -K
3544 filename. This should rule out race conditions when creating
3547 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
3548 processing considerably. Previous code took too long for large mails,
3549 triggering a timeout which in turn triggers #401.
3551 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
3552 in the DK code in transports.c. sendfile() is not really portable,
3553 hence the _LINUX specificness.
3555 TF/01 In the add_headers option to the mail command in an Exim filter,
3556 there was a bug that Exim would claim a syntax error in any
3557 header after the first one which had an odd number of characters
3560 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
3561 callout verification, Exim cached a "reject" for the entire domain. This
3562 is correct for most verifications, but it is not correct for a recipient
3563 verification with use_sender or use_postmaster set, because in that case
3564 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
3565 case of MAIL FROM:<> rejection from other early rejections (e.g.
3566 rejection of HELO). When verifying a recipient using a non-null MAIL
3567 address, the cache is ignored if it shows MAIL FROM:<> rejection.
3568 Whatever the result of the callout, the value of the domain cache is
3569 left unchanged (for any other kind of callout, getting as far as trying
3570 RCPT means that the domain itself is ok).
3572 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
3575 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
3576 manifest itself as EPIPE rather than ECONNECT. When tidying away a
3577 session, the daemon ignores ECONNECT errors and logs others; it now
3578 ignores EPIPE as well.
3580 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
3581 (quoted-printable decoding).
3583 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
3584 later the small subsequent patch to fix an introduced bug.
3586 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
3588 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
3590 PH/08 An error is now given if message_size_limit is specified negative.
3592 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
3593 to be given (somewhat) arbitrary names.
3595 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
3598 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
3599 miscellaneous code fixes
3601 PH/10 Added the log_reject_target ACL modifier to specify where to log
3604 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
3605 hostname. This is wrong, because it relates to the incoming message (and
3606 probably the interface on which it is arriving) and not to the outgoing
3607 callout (which could be using a different interface). This has been
3608 changed to use the value of the helo_data option from the smtp transport
3609 instead - this is what is used when a message is actually being sent. If
3610 there is no remote transport (possible with a router that sets up host
3611 addresses), $smtp_active_hostname is used.
3613 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
3614 tweaks were necessary in order to get it to work (see also 21 below):
3615 (a) The code assumed that strncpy() returns a negative number on buffer
3616 overflow, which isn't the case. Replaced with Exim's string_format()
3618 (b) There were several signed/unsigned issues. I just did the minimum
3619 hacking in of casts. There is scope for a larger refactoring.
3620 (c) The code used strcasecmp() which is not a standard C function.
3621 Replaced with Exim's strcmpic() function.
3622 (d) The code set only $1; it now sets $auth1 as well.
3623 (e) A simple test gave the error "authentication client didn't specify
3624 service in request". It would seem that Dovecot has changed its
3625 interface. Fortunately there's a specification; I followed it and
3626 changed what the client sends and it appears to be working now.
3628 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
3631 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
3632 address A is aliased to B and C, where B exists and C does not. Without
3633 -v the output is "A verified" because verification stops after a
3634 successful redirection if more than one address is generated. However,
3635 with -v the child addresses are also verified. Exim was outputting "A
3636 failed to verify" and then showing the successful verification for C,
3637 with its parentage. It now outputs "B failed to verify", showing B's
3638 parentage before showing the successful verification of C.
3640 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
3641 look up a TXT record in a specific list after matching in a combined
3644 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
3645 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
3646 they consult the DNS. I had assumed they would set it the way they
3647 wanted; and indeed my experiments on Linux seem to show that in some
3648 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
3649 To be on the safe side, however, I have now made the interface to
3650 host_find_byname() similar to host_find_bydns(), with an argument
3651 containing the DNS resolver options. The host_find_byname() function now
3652 sets these options at its start, just as host_find_bydns() does. The smtp
3653 transport options dns_qualify_single and dns_search_parents are passed to
3654 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
3655 of host_find_byname() use the default settings of RES_DEFNAMES
3656 (qualify_single) but not RES_DNSRCH (search_parents).
3658 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
3659 spool_read_header() do less string testing, by means of a preliminary
3660 switch on the second character of optional "-foo" lines. (This is
3661 overdue, caused by the large number of possibilities that now exist.
3662 Originally there were few.) While I was there, I also converted the
3663 str(n)cmp tests so they don't re-test the leading "-" and the first
3664 character, in the hope this might squeeze out yet more improvement.
3666 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
3667 flag allowing group syntax was set by the header_syntax check but not
3668 turned off, possible causing trouble later; (2) The flag was not being
3669 set at all for the header_verify test, causing "group"-style headers to
3670 be rejected. I have now set it in this case, and also caused header_
3671 verify to ignore an empty address taken from a group. While doing this, I
3672 came across some other cases where the code for allowing group syntax
3673 while scanning a header line wasn't quite right (mostly, not resetting
3674 the flag correctly in the right place). These bugs could have caused
3675 trouble for malformed header lines. I hope it is now all correct.
3677 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
3678 with the "reply" argument non-NULL. The code, however (which originally
3679 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
3680 but it didn't always do it. This confused somebody who was copying the
3681 code for some other use. I have removed all the tests.
3683 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
3684 feature that was used to support insecure browsers during the U.S. crypto
3685 embargo. It requires special client support, and Exim is probably the
3686 only MTA that supported it -- and would never use it because real RSA is
3687 always available. This code has been removed, because it had the bad
3688 effect of slowing Exim down by computing (never used) parameters for the
3689 RSA_EXPORT functionality.
3691 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
3692 authenticator to fail if there's a tab character in the incoming data
3693 (there should never be unless someone is messing about, as it's supposed
3694 to be base64-encoded). Also added, on Timo's advice, the "secured" option
3695 if the connection is using TLS or if the remote IP is the same as the
3696 local IP, and the "valid-client-cert option" if a client certificate has
3699 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
3700 authenticators. This can be used for authorization after authentication
3701 succeeds. (In the case of plaintext, it servers for both authentication
3704 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
3705 if any retry times were supplied.
3707 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
3708 connection, where there is no client IP address to check. In this
3709 situation, the verify now always succeeds.
3711 PH/25 Applied John Jetmore's -Mset patch.
3713 PH/26 Added -bem to be like -Mset, but loading a message from a file.
3715 PH/27 In a string expansion for a processed (not raw) header when multiple
3716 headers of the same name were present, leading whitespace was being
3717 removed from all of them, but trailing whitespace was being removed only
3718 from the last one. Now trailing whitespace is removed from each header
3719 before concatenation. Completely empty headers in a concatenation (as
3720 before) are ignored.
3722 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
3723 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
3725 PH/29 [Removed. This was a change that I later backed out, and forgot to
3726 correct the ChangeLog entry (that I had efficiently created) before
3727 committing the later change.]
3729 PH/30 Exim was sometimes attempting to deliver messages that had suffered
3730 address errors (4xx response to RCPT) over the same connection as other
3731 messages routed to the same hosts. Such deliveries are always "forced",
3732 so retry times are not inspected. This resulted in far too many retries
3733 for the affected addresses. The effect occurred only when there were more
3734 hosts than the hosts_max_try setting in the smtp transport when it had
3735 the 4xx errors. Those hosts that it had tried were not added to the list
3736 of hosts for which the message was waiting, so if all were tried, there
3737 was no problem. Two fixes have been applied:
3739 (i) If there are any address or message errors in an SMTP delivery, none
3740 of the hosts (tried or untried) are now added to the list of hosts
3741 for which the message is waiting, so the message should not be a
3742 candidate for sending over the same connection that was used for a
3743 successful delivery of some other message. This seems entirely
3744 reasonable: after all the message is NOT "waiting for some host".
3745 This is so "obvious" that I'm not sure why it wasn't done
3746 previously. Hope I haven't missed anything, but it can't do any
3747 harm, as the worst effect is to miss an optimization.
3749 (ii) If, despite (i), such a delivery is accidentally attempted, the
3750 routing retry time is respected, so at least it doesn't keep
3751 hammering the server.
3753 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
3754 in ${readsocket because some servers need this prod.
3756 PH/32 Added some extra debug output when updating a wait-xxx database.
3758 PH/33 The hint "could be header name not terminated by colon", which has been
3759 given for certain expansion errors for a long time, was not being given
3760 for the ${if def:h_colon_omitted{... case.
3762 PH/34 The spec says: "With one important exception, whenever a domain list is
3763 being scanned, $domain contains the subject domain." There was at least
3764 one case where this was not true.
3766 PH/35 The error "getsockname() failed: connection reset by peer" was being
3767 written to the panic log as well as the main log, but it isn't really
3768 panic-worthy as it just means the connection died rather early on. I have
3769 removed the panic log writing for the ECONNRESET error when getsockname()
3772 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
3773 runs only) independently of the message's sender address. This meant
3774 that, if the 4xx error was in fact related to the sender, a different
3775 message to the same recipient with a different sender could confuse
3776 things. In particular, this can happen when sending to a greylisting
3777 server, but other circumstances could also provoke similar problems.
3778 I have changed the default so that the retry time for these errors is now
3779 based a combination of the sender and recipient addresses. This change
3780 can be overridden by setting address_retry_include_sender=false in the
3783 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
3784 remote server are returned as part of bounce messages. This was not
3785 happening for LMTP over a pipe (the lmtp transport), but now it is the
3786 same for both kinds of LMTP.
3788 PH/38 Despite being documented as not happening, Exim was rewriting addresses
3789 in header lines that were in fact CNAMEs. This is no longer the case.
3791 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
3792 and queue runs started by the daemon processed all messages. This has
3793 been fixed so that -R and -S can now usefully be given with -q<time>.
3795 PH/40 Import PCRE release 6.7 (fixes some bugs).
3797 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
3799 PH/42 Give an error if -q is specified more than once.
3801 PH/43 Renamed the variables $interface_address and $interface_port as
3802 $received_ip_address and $received_port, to make it clear that these
3803 values apply to message reception, and not to the outgoing interface when
3804 a message is delivered. (The old names remain recognized, of course.)
3806 PH/44 There was no timeout on the connect() call when using a Unix domain
3807 socket in the ${readsocket expansion. There now is.
3809 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
3810 be meaningful with "accept".
3812 SC/01 Eximstats V1.43
3813 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
3815 SC/02 Eximstats V1.44
3816 Use a glob alias rather than an array ref in the generated
3817 parser. This improves both readability and performance.
3819 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
3820 Collect SpamAssassin and rejection statistics.
3821 Don't display local sender or destination tables unless
3822 there is data to show.
3823 Added average volumes into the top table text output.
3825 SC/04 Eximstats V1.46
3826 Collect data on the number of addresses (recipients)
3827 as well as the number of messages.
3829 SC/05 Eximstats V1.47
3830 Added 'Message too big' to the list of mail rejection
3831 reasons (thanks to Marco Gaiarin).
3833 SC/06 Eximstats V1.48
3834 Mainlog lines which have GMT offsets and are too short to
3835 have a flag are now skipped.
3837 SC/07 Eximstats V1.49 (Alain Williams)
3838 Added the -emptyok flag.
3840 SC/08 Eximstats V1.50
3841 Fixes for obtaining the IP address from reject messages.
3843 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
3844 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
3845 whitespace changes from 4.64-PH/27
3847 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
3850 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
3851 are found, allow negative numbers in numeric criteria)
3853 JJ/06 exipick.20061117.2, added new $message_body_missing variable
3855 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
3856 to match changes made in 4.64-PH/43
3858 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
3860 PH/47 Put in an explicit test for a DNS lookup of an address record where the
3861 "domain" is actually an IP address, and force a failure. This locks out
3862 those revolvers/nameservers that support "A-for-A" lookups, in
3863 contravention of the specifications.
3865 PH/48 When a host name was looked up from an IP address, and the subsequent
3866 forward lookup of the name timed out, the host name was left in
3867 $sender_host_name, contrary to the specification.
3869 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3870 restricted to single-key lookups, Exim was not diagnosing an error if
3871 * or *@ was used with a query-style lookup.
3873 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3875 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3876 long ago, but noone interested enough thought of it. Let's just say that
3877 the "1.1" means that there are some new functions that weren't there at
3878 some point in the past.
3880 PH/51 Error processing for expansion failure of helo_data from an smtp
3881 transport during callout processing was broken.
3883 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3884 tested/used via the -bh/-bhc/-bs options.
3886 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3887 bug, fixed in subsequent PCRE releases).
3889 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3890 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3892 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3898 SC/01 Use a glob alias rather than an array ref in eximstats generated
3899 parser. This improves both readability and performance.
3901 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3902 Don't display local sender or destination tables in eximstats unless
3903 there is data to show.
3904 Added average volumes into the eximstats top table text output.
3906 SC/03 Collect data on the number of addresses (recipients) as well
3907 as the number of messages in eximstats.
3909 TF/01 Correct an error in the documentation for the redirect router. Exim
3910 does (usually) call initgroups() when daemonizing.
3912 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3913 with consistent privilege compared to when running as a daemon.
3915 TF/03 Note in the spec that $authenticated_id is not set for local
3916 submissions from trusted users.
3918 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3919 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3921 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3922 by adding some example configuration directives to the default
3923 configuration file. A little bit of work is required to uncomment the
3924 directives and define how usernames and passwords are checked, but
3925 there is now a framework to start from.
3927 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3928 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3929 without this. I don't know how relevant this is to other LDAP libraries.
3931 PH/02 Add the verb name to the "unknown ACL verb" error.
3933 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3935 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3937 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3938 directory not expanded when it should be if an expanded home directory
3939 was set for the address (which is overridden by the transport).
3941 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3944 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3945 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3946 because it is too late at that time, and has no effect.
3948 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3949 security issue with \' (bugzilla #107). I could not use the
3950 PQescapeStringConn() function, because it needs a PGconn value as one of
3953 PH/08 When testing addresses using -bt, indicate those final addresses that
3954 are duplicates that would not cause an additional delivery. At least one
3955 person was confused, thinking that -bt output corresponded to deliveries.
3956 (Suppressing duplicates isn't a good idea as you lose the information
3957 about possibly different redirections that led to the duplicates.)
3959 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3960 systems where poll() doesn't work, in particular OS X.
3962 PH/10 Added more information to debugging output for retry time not reached.
3964 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3965 operations in malware.c.
3967 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3970 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3971 syslog_duplication set false, log lines that would normally be written
3972 both the the main log and to the reject log were not written to syslog at
3975 PH/14 In the default configuration, change the use of "message" in ACL warn
3976 statements to "add_header".
3978 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3979 not followed by a command (e.g. "seen endif").
3981 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3982 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3985 PH/17 Added extra conditions to the default value of delay_warning_condition
3989 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3990 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3991 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3994 The Auto-Submitted: and various List- headers are standardised, whereas I
3995 don't think Precedence: ever was.
3997 PH/18 Refactored debugging code in route_finduser() to show more information,
3998 in particular, the error code if getpwnam() issues one.
4000 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
4001 This is apparently needed in addition to the PH/07 change above to avoid
4002 any possible encoding problems.
4004 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
4005 but not after initializing Perl.
4007 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
4008 output them only if debugging. By default they are written stderr,
4009 apparently, which is not desirable.
4011 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
4014 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
4017 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
4019 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
4020 authenticated or an ident call has been made. Suppress the default
4021 values for $authenticated_id and $authenticated_sender (but permit -oMai
4022 and -oMas) when testing with -bh.
4024 PH/24 Re-jigged the order of the tests in the default configuration so that the
4025 tests for valid domains and recipients precede the DNS black list and CSA
4026 tests, on the grounds that those ones are more expensive.
4028 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
4029 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
4030 This bug exists in every version of Exim that I still have, right back to
4033 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
4034 However, an attempt to turn on case-sensitivity in a regex key by
4035 including (?-i) didn't work because the subject string was already
4036 lowercased, and the effects were non-intuitive. It turns out that a
4037 one-line patch can be used to allow (?-i) to work as expected.
4043 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
4044 other effects) broke the use of negated acl sub-conditions.
4046 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
4049 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
4050 "Deny" causes Exim to reject the incoming connection with a 554 error.
4051 Unfortunately, if there is a major crisis, such as a disk failure,
4052 tcp-wrappers gives "deny", whereas what one would like would be some
4053 kind of temporary error. A kludge has been added to help with this.
4054 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
4055 554 error is used if errno is still zero or contains ENOENT (which occurs
4056 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
4059 PH/03 Add -lutil to the default FreeBSD LIBS setting.
4061 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
4062 errors. Otherwise a message that provokes a temporary error (when other
4063 messages do not) can cause a whole host to time out.
4065 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
4066 addresses were routed directly to files or pipes from a redirect router.
4067 File deliveries just didn't batch; pipe deliveries might have suffered
4070 PH/06 A failure to get a lock for a hints database would erroneously always say
4071 "Failed to get write lock", even when it was really a read lock.
4073 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
4074 of 0600. This has been changed to use the value of the lockfile_mode
4075 option (which defaults to 0600).
4077 PH/08 Applied small patch from the Sieve maintainer.
4079 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
4080 folder from quota calculations, a direct delivery into this folder messed
4081 up the contents of the maildirsize file. This was because the regex was
4082 used only to exclude .Trash (or whatever) when the size of the mailbox
4083 was calculated. There was no check that a delivery was happening into an
4084 excluded directory. This bug has been fixed by ignoring all quota
4085 processing for deliveries into excluded directories.
4087 PH/10 Added the maildirfolder_create_regex option to appendfile.
4093 PH/01 The code for finding all the local interface addresses on a FreeBSD
4094 system running IPv6 was broken. This may well have applied to all BSD
4095 systems, as well as to others that have similar system calls. The broken
4096 code found IPv4 interfaces correctly, but gave incorrect values for the
4097 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
4098 that it would not match correctly against @[] and not recognize the IPv6
4101 PH/02 The ipliteral router was not recognizing addresses of the form user@
4102 [ipv6:....] because it didn't know about the "ipv6:" prefix.
4104 PH/03 Added disable_ipv6.
4106 PH/04 Changed $reply_address to use the raw form of the headers instead of the
4107 decoded form, because it is most often used to construct To: headers
4108 lines in autoreplies, and the decoded form may well be syntactically
4109 invalid. However, $reply_address has leading white space removed, and all
4110 newlines turned into spaces so that the autoreply transport does not
4113 PH/05 If group was specified without a user on a router, and no group or user
4114 was specified on a transport, the group from the router was ignored.
4116 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
4117 for visible compile-time settings that can be used to change these
4118 numbers, for those that want even more. Backwards compatibility with old
4119 spool files has been maintained. However, going back to a previous Exim
4120 release will lost any variables that are in spool files.
4122 PH/07 Two small changes when running in the test harness: increase delay when
4123 passing a TCP/IP connection to a new process, in case the original
4124 process has to generate a bounce, and remove special handling of
4125 127.0.0.2 (sic), which is no longer necessary.
4127 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
4128 be the same on different OS.
4130 PH/09 Moved a debug statement in filter processing to avoid a race problem when
4133 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
4134 whether --show-vars was specified or not
4136 JJ/02 exipick: Added support for new ACL variable spool format introduced
4139 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
4140 syntactically invalid From: or Reply-to: line, and a filter used this to
4141 generate an autoreply, and therefore failed to obtain an address for the
4142 autoreply, Exim could try to deliver to a non-existent relative file
4143 name, causing unrelated and misleading errors. What now happens is that
4144 it logs this as a hard delivery error, but does not attempt to create a
4147 PH/11 The exinext utility has a -C option for testing purposes, but although
4148 the given file was scanned by exinext itself; it wasn't being passed on
4149 when Exim was called.
4151 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
4152 an end-of-file indication when reading a command response.
4154 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
4155 compiled. In many other places in Exim, IPv6 addresses are always
4156 recognized, so I have changed this. It also means that IPv4 domain
4157 literals of the form [IPV4:n.n.n.n] are now always recognized.
4159 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
4160 used if the router is not running as root, for example, when verifying at
4161 ACL time, or when using -bh. The debugging output from this situation was
4162 non-existent - all you got was a failure to exec. I have made two
4165 (a) Failures to set uid/gid, the current directory, or a process leader
4166 in a subprocess such as that created by queryprogram now generate
4167 suitable debugging output when -d is set.
4169 (b) The queryprogram router detects when it is not running as root,
4170 outputs suitable debugging information if -d is set, and then runs
4171 the subprocess without attempting to change uid/gid.
4173 PH/15 Minor change to Makefile for building test_host (undocumented testing
4176 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
4177 additional section of a DNS packet that returns MX or SRV records.
4178 Instead, it always explicitly searches for A/AAAA records. This avoids
4179 major problems that occur when a DNS server includes only records of one
4180 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
4181 fixed another bug: if SRV records were looked up and the corresponding
4182 address records were *not* found in the additional section, the port
4183 values from the SRV records were lost.
4185 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
4186 using the correct key (the original address) when searching the retry
4187 rules in order to find which one to use for generating the retry hint.
4189 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
4190 adding the default one. Similarly, if it contains a Reply-To: header, the
4191 errors_reply_to option, if set, is not used.
4193 PH/19 When calculating a retry time, Exim used to measure the "time since
4194 failure" by looking at the "first failed" field in the retry record. Now
4195 it does not use this if it is later than than the arrival time of the
4196 message. Instead it uses the arrival time. This makes for better
4197 behaviour in cases where some deliveries succeed, thus re-setting the
4198 "first failed" field. An example is a quota failure for a huge message
4199 when small messages continue to be delivered. Without this change, the
4200 "time since failure" will always be short, possible causing more frequent
4201 delivery attempts for the huge message than are intended.
4202 [Note: This change was subsequently modified - see PH/04 for 4.62.]
4204 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
4205 $1, $2, $3) because the numerical variables can be reset during some
4206 expansion items (e.g. "match"), thereby losing the authentication data.
4208 PH/21 Make -bV show the size of off_t variables so that the test suite can
4209 decide whether to run tests for quotas > 2G.
4211 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
4212 mailbox_size, and mailbox_filecount in the appendfile transport. If a
4213 filecount value is greater than 2G or if a quota value is greater than 2G
4214 on a system where the size of off_t is not greater than 4, a panic error
4217 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
4218 never match. The debug and -bh output now contains an explicit error
4219 message indicating a malformed IPv4 address or mask.
4221 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
4222 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
4223 PH/23 above applies.
4225 PH/25 Do not write to syslog when running in the test harness. The only
4226 occasion when this arises is a failure to open the main or panic logs
4227 (for which there is an explicit test).
4229 PH/26 Added the /no_tell option to "control=freeze".
4231 PH/27 If a host name lookup failed very early in a connection, for example, if
4232 the IP address matched host_lookup and the reverse lookup yielded a name
4233 that did not have a forward lookup, an error message of the form "no IP
4234 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
4235 could be logged. Now it outputs the IP address instead of "NULL".
4237 PH/28 An enabling patch from MH: add new function child_open_exim2() which
4238 allows the sender and the authenticated sender to be set when
4239 submitting a message from within Exim. Since child_open_exim() is
4240 documented for local_scan(), the new function should be too.
4242 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
4243 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
4244 results in an empty string is now treated as unset.
4246 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
4248 PH/31 Added sender_verify_fail logging option.
4250 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
4251 needed by RFC 822 but not by RFC 2822 was commented out. I have now
4252 tidied the source and removed it altogether.
4254 PH/33 When a queue run was abandoned because the load average was too high, a
4255 log line was always written; now it is written only if the queue_run log
4256 selector is set. In addition, the log line for abandonment now contains
4257 information about the queue run such as the pid. This is always present
4258 in "start" and "stop" lines but was omitted from the "abandon" line.
4260 PH/34 Omit spaces between a header name and the colon in the error message that
4261 is given when verify = headers_syntax fails (if there are lots of them,
4262 the message gets confusing).
4264 PH/35 Change the default for dns_check_names_pattern to allow slashes within
4265 names, as there are now some PTR records that contain slashes. This check
4266 is only to protect against broken name servers that fall over on strange
4267 characters, so the fact that it applies to all lookups doesn't matter.
4269 PH/36 Now that the new test suite is complete, we can remove some of the
4270 special code in Exim that was needed for the old test suite. For example,
4271 sorting DNS records because real resolvers return them in an arbitrary
4272 order. The new test suite's fake resolver always returns records in the
4275 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
4276 bounces) except when queue_only is set, to avoid logging races between
4277 the different processes.
4279 PH/38 Panic-die if .include specifies a non-absolute path.
4281 PH/39 A tweak to the "H" retry rule from its user.
4283 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
4284 a label. They prevented compilation on older perls.
4286 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
4287 a warning to be raised on newish perls.
4289 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
4290 on queue. Changes to match documented behaviour of showing count of
4291 messages matching specified criteria.
4293 PH/40 Changed the default ident timeout from 30s to 5s.
4295 PH/41 Added support for the use of login_cap features, on those BSD systems
4296 that have them, for controlling the resources used by pipe deliveries.
4298 PH/42 The content-scanning code uses fopen() to create files in which to put
4299 message data. Previously it was not paying any attention to the mode of
4300 the files. Exim runs with umask(0) because the rest of the code creates
4301 files with open(), and sets the required mode explicitly. Thus, these
4302 files were ending up world-writeable. This was not a big issue, because,
4303 being within the spool directory, they were not world-accessible. I have
4304 created a function called modefopen, which takes an additional mode
4305 argument. It sets umask(777), creates the file, chmods it to the required
4306 mode, then resets the umask. All the relevant calls to fopen() in the
4307 content scanning code have been changed to use this function.
4309 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
4310 to 24 hours. This avoids potential overflow problems when processing G
4311 and H retry rules. I suspect nobody ever tinkers with this value.
4313 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
4315 PH/45 When the plaintext authenticator is running as a client, the server's
4316 challenges are checked to ensure they are valid base64 strings. By
4317 default, the authentication attempt is cancelled if an invalid string is
4318 received. Setting client_ignore_invalid_base64 true ignores these errors.
4319 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
4320 they are received. Thus, the responses can be made to depend on the
4321 challenges. If an invalid string is ignored, an empty string is placed in
4324 PH/46 Messages that are created by the autoreply transport now contains a
4325 References: header, in accordance with RFCs 2822 and 3834.
4327 PH/47 Added authenticated_sender_force to the smtp transport.
4329 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
4331 PH/49 Installed latest patch from the Sieve maintainer.
4333 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
4334 was also set, the appendfile transport was unnecessarily scanning a
4335 directory of message files (e.g. for maildir delivery) to find the count
4336 of files (along with the size), even though it did not need this
4337 information. It now does the scan only if it needs to find either the
4338 size of the count of files.
4340 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
4342 PH/52 Two bugs concerned with error handling when the smtp transport is
4345 (i) Exim was not creating retry information for temporary errors given
4346 for individual recipients after the DATA command when the smtp transport
4347 was used in LMTP mode. This meant that they could be retried too
4348 frequently, and not timed out correctly.
4350 (ii) Exim was setting the flag that allows error details to be returned
4351 for LMTP errors on RCPT commands, but not for LMTP errors for individual
4352 recipients that were returned after the DATA command.
4354 PH/53 This is related to PH/52, but is more general: for any failing address,
4355 when detailed error information was permitted to be returned to the
4356 sender, but the error was temporary, then after the final timeout, only
4357 "retry timeout exceeded" was returned. Now it returns the full error as
4358 well as "retry timeout exceeded".
4360 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
4361 do this, and (what is worse) MTAs that accept it.
4363 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
4364 will now be deprecated.
4366 PH/56 New os.c-cygwin from the Cygwin maintainer.
4368 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
4369 formats (previously only available in exim formats via -bpr, -bpru,
4370 and -bpra. Now also available in native and exiqgrep formats)
4372 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
4373 with very large, slow to parse queues
4375 JJ/08 exipick: added ! as generic prefix to negate any criteria format
4377 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
4379 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
4380 responses to authentication challenges, though it was showing the
4381 challenges; (ii) I've removed the CR characters from the debug output for
4384 PH/58 Allow for the insertion of a newline as well as a space when a string
4385 is turned into more than one encoded-word during RFC 2047 encoding. The
4386 Sieve code now uses this.
4388 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
4389 data_4xx, lost_connection, tls_required.
4391 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
4392 message was being sent as an SMTP response.
4394 PH/61 Add -l and -k options to exicyclog.
4396 PH/62 When verifying, if an address was redirected to one new address, so that
4397 verification continued, and the new address failed or deferred after
4398 having set something in $address_data, the value of $address_data was not
4399 passed back to the ACL. This was different to the case when no
4400 redirection occurred. The value is now passed back in both cases.
4402 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
4403 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
4404 use login_cap.h, so on its own it isn't the distinguishing feature. The
4405 new name refers directly to the setclassresources() function.
4407 PH/65 Added configuration files for NetBSD3.
4409 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
4411 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
4412 is preferred over IPv4.
4414 PH/68 The bounce_return_message and bounce_return_body options were not being
4415 honoured for bounces generated during the reception of non-SMTP messages.
4416 In particular, this applied to messages rejected by the ACL. This bug has
4417 been fixed. However, if bounce_return_message is true and bounce_return_
4418 body is false, the headers that are returned for a non-SMTP message
4419 include only those that have been read before the error was detected.
4420 (In the case of an ACL rejection, they have all been read.)
4422 PH/69 The HTML version of the specification is now built in a directory called
4423 spec_html instead of spec.html, because the latter looks like a path with
4424 a MIME-type, and this confuses some software.
4426 PH/70 Catch two compiler warnings in sieve.c.
4428 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
4429 function verify_get_ident() calls ip_connect() to connect a socket, but
4430 if the "connect()" function timed out, ip_connect() used to close the
4431 socket. However, verify_get_ident() also closes the socket later, and in
4432 between Exim writes to the log, which may get opened at this point. When
4433 the socket was closed in ip_connect(), the log could get the same file
4434 descriptor number as the socket. This naturally causes chaos. The fix is
4435 not to close the socket in ip_connect(); the socket should be closed by
4436 the function that creates it. There was only one place in the code where
4437 this was missing, in the iplookup router, which I don't think anybody now
4438 uses, but I've fixed it anyway.
4440 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
4441 well as to direct DNS lookups. Otherwise the handling of names in host
4442 lists is inconsistent and therefore confusing.
4448 PH/01 Two changes to the default runtime configuration:
4450 (1) Move the checks for relay_from_hosts and authenticated clients from
4451 after to before the (commented out) DNS black list checks.
4453 (2) Add control=submission to the relay_from_hosts and authenticated
4454 clients checks, on the grounds that messages accepted by these
4455 statements are most likely to be submissions.
4457 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
4459 (1) Generate an error if the third argument for the ${prvs expansion is
4462 (2) Treat a missing third argument of ${prvscheck as if it were an empty
4465 (3) Reset the variables that are obtained from the first argument of
4466 ${prvscheck and used in the second argument before leaving the code,
4467 because their memory is reclaimed, so using them afterwards may do
4470 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
4471 one (it's much easier than Tom thought :-).
4473 (5) Because of (4), we can now allow for the use of $prvscheck_result
4474 inside the third argument.
4476 PH/03 For some reason, the default setting of PATH when running a command from
4477 a pipe transport was just "/usr/bin". I have changed it to
4480 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
4481 anything to be listed in the output from -bV.
4483 PH/05 When a filter generated an autoreply, the entire To: header line was
4484 quoted in the delivery log line, like this:
4486 => >A.N.Other <ano@some.domain> <original@ddress> ...
4488 This has been changed so that it extracts the operative address. There
4489 may be more than one such address. If so, they are comma-separated, like
4492 => >ano@some.domain,ona@other.domain <original@ddress> ...
4494 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
4495 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
4496 not being looked up in the rDNS to get a host name, Exim was showing the
4497 IP address twice in Received: lines, even though the IP addresses were
4498 identical. For example:
4500 Received: from [1.2.3.4] (helo=[1.2.3.4])
4502 However, if the real host name was known, it was omitting the HELO data
4503 if it matched the actual IP address. This has been tidied up so that it
4504 doesn't show the same IP address twice.
4506 PH/07 When both +timestamp and +memory debugging was on, the value given by
4507 $tod_xxx expansions could be wrong, because the tod_stamp() function was
4508 called by the debug printing, thereby overwriting the timestamp buffer.
4509 Debugging no longer uses the tod_stamp() function when +timestamp is set.
4511 PH/08 When the original message was included in an autoreply transport, it
4512 always said "this is a copy of the message, including all the headers",
4513 even if body_only or headers_only was set. It now gives an appropriate
4516 PH/09 Applied a patch from the Sieve maintainer which:
4518 o fixes some comments
4519 o adds the (disabled) notify extension core
4520 o adds some debug output for the result of if/elsif tests
4521 o points to the current vacation draft in the documentation
4522 and documents the missing references header update
4526 o fixes a bug in processing the envelope test (when testing
4527 multiple envelope elements, the last element determined the
4530 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
4531 Electronic Mail") by including:
4533 Auto-submitted: auto-generated
4535 in the messages that it generates (bounce messages and others, such as
4536 warnings). In the case of bounce messages for non-SMTP messages, there was
4537 also a typo: it was using "Auto_submitted" (underscore instead of
4538 hyphen). Since every message generated by Exim is necessarily in response
4539 to another message, thes have all been changed to:
4541 Auto-Submitted: auto-replied
4543 in accordance with these statements in the RFC:
4545 The auto-replied keyword:
4547 - SHOULD be used on messages sent in direct response to another
4548 message by an automatic process,
4550 - MUST NOT be used on manually-generated messages,
4552 - MAY be used on Delivery Status Notifications (DSNs) and Message
4553 Disposition Notifications (MDNs),
4555 - MUST NOT be used on messages generated by automatic or periodic
4556 processes, except for messages which are automatic responses to
4559 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
4560 to the default Received: header definition.
4562 PH/12 Added log selector acl_warn_skipped (default on).
4564 PH/13 After a successful wildlsearch lookup, discard the values of numeric
4565 variables because (a) they are in the wrong storage pool and (b) even if
4566 they were copied, it wouldn't work properly because of the caching.
4568 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
4569 checking when decoding. Apparently there are clients that generate
4570 overlong encoded strings. Why am I not surprised?
4572 PH/15 If the first argument of "${if match_address" was not empty, but did not
4573 contain an "@" character, Exim crashed. Now it writes a panic log message
4574 and treats the condition as false.
4576 PH/16 In autoreply, treat an empty string for "once" the same as unset.
4578 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
4579 extension "envelope-auth". The code is finished and in agreement with
4580 other implementations, but there is no documentation so far and in fact,
4581 nobody wrote the draft yet. This extension is currently #undef'ed, thus
4582 not changing the active code.
4584 Print executed "if" and "elsif" statements when debugging is used. This
4585 helps a great deal to understand what a filter does.
4587 Document more things not specified clearly in RFC3028. I had all this
4588 sorted out, when out of a sudden new issues came to my mind. Oops."
4590 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
4593 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
4594 canonical form (as documented). However, after a host name lookup from
4595 the IP address, check_host() was doing a simple string comparison with
4596 addresses acquired from the DNS when checking that the found name did
4597 have the original IP as one of its addresses. Since any found IPv6
4598 addresses are likely to be in abbreviated form, the comparison could
4599 fail. Luckily, there already exists a function for doing the comparison
4600 by converting both addresses to binary, so now that is used instead of
4601 the text comparison.
4603 PH/20 There was another similar case to PH/19, when a complete host name was
4604 given in a host list; looking up its IP address could give an abbreviated
4605 form, whereas the current host's name might or might not be abbreviated.
4606 The same fix has been applied.
4612 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
4613 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
4616 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
4617 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
4619 PH/03 Typo: missing ".o" in src/pcre/Makefile.
4621 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
4622 header line, restrict the check to what is listed in RFCs 2369 and 2929.
4623 Also, for "Auto-Submitted", treat anything other than "no" as
4624 non-personal, in accordance with RFC 3834. (Previously it treated
4625 anything starting "auto-" as non-personal.)
4627 TF/01 The control=submission/name=... option had a problem with syntax
4628 errors if the name included a slash character. The /name= option
4629 now slurps the rest of the string, so it can include any characters
4630 but it must come last in the list of options (after /sender_retain
4633 PH/05 Some modifications to the interface to the fake nameserver for the new
4641 TK/01 Added the "success_on_redirect" address verification option. See
4642 NewStuff for rationale and an example.
4644 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
4646 PH/02 Patch to exigrep to allow it to work on syslog lines.
4648 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
4649 fread() to skip over the body file's header line, because in Cygwin the
4650 header line is locked and is inaccessible.
4652 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
4653 co-exist for some time) to make it clear that it is the Exim ID that is
4654 referenced, not the Message-ID: header line.
4656 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
4657 string_format() function, because snprintf() does not exist on all
4660 PH/06 The use of forbid_filter_existstest now also locks out the use of the
4661 ${stat: expansion item.
4663 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
4664 protocol synchronization error", to keep the pedants happy.
4666 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
4667 well as for IRIX systems, when gcc is being used. See the host.c source
4670 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
4672 PH/10 Named domain lists were not working if used in a queue_smtp_domains
4675 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
4676 transport and to the smtp transport in LMTP mode.
4678 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
4680 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
4681 run a filter in a subprocess. This could lead to confusion in subsequent
4682 lookups in the parent process. There should also be a search_tidyup() at
4683 the end of the subprocess.
4685 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
4686 only if the host matched helo_try_verify_hosts, which caused the
4687 verification to occur when the EHLO/HELO command was issued. The ACL just
4688 tested the remembered result. Now, if a previous verification attempt has
4689 not happened, "verify = helo" does it there and then.
4691 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
4693 TK/03 Fix log output including CR from clamd.
4695 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
4696 exist provoked a memory error which could cause a segfault.
4698 PH/15 Installed PCRE 6.2
4700 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
4702 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
4703 of the problem. Specifically, suggested +O2 rather than +O1 for the
4706 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
4708 PH/20 If a delivery was routed to a non-standard port by means of an SRV
4709 record, the port was not correctly logged when the outgoing_port log
4710 selector was set (it logged the transort's default port).
4712 PH/21 Added support for host-specific ports to manualroute, queryprogram,
4713 fallback_hosts, and "hosts" in the smtp transport.
4715 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
4716 host errors such as "Connection refused".
4718 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
4719 authentication with radiusclient 0.4.9:
4721 - Error returned from rc_read_config was caught wrongly
4722 - Username/password not passed on to radius server due to wrong length.
4724 The presumption is that some radiusclient API changes for 4.51/PH/17
4725 were not taken care of correctly. The code is still untested by me (my
4726 Linux distribution still has 0.3.2 of radiusclient), but it was
4727 contributed by a Radius user.
4729 PH/24 When doing a callout, the value of $domain wasn't set correctly when
4730 expanding the "port" option of the smtp transport.
4732 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
4733 while reading a MIME header. Thanks to Tom Hughes for a patch.
4735 PH/24 Include config.h inside local_scan.h so that configuration settings are
4738 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
4739 that in an ACL for RCPT (for example), you can examine exactly what was
4742 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
4743 commands, but it was not correctly comparing the address with the actual
4744 client host address. Thus, it would show the EHLO address in Received:
4745 header lines when this was not necessary.
4747 PH/27 Added the % operator to ${eval:}.
4749 PH/28 Exim tries to create and chdir to its spool directory when it starts;
4750 it should be ignoring failures (because with -C, for example, it has lost
4751 privilege). It wasn't ignoring creation failures other than "already
4754 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
4757 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
4758 because an input line was too long, either on its own, or by virtue of
4759 too many continuations, the temporary file was not being removed, and the
4760 return code was incorrect.
4762 PH/31 Missing "BOOL" in function definition in filtertest.c.
4764 PH/32 Applied Sieve patches from the maintainer.
4766 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
4768 PH/33 Added "verify = not_blind".
4770 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
4771 Local/Makefile (with some defaults set). These are used in built scripts
4772 such as exicyclog, but they have never been used in the exim_install
4773 script (though there are many overriding facilities there). I have
4774 arranged that the exim_install script now takes note of these two
4777 PH/35 Installed configuration files for Dragonfly.
4779 PH/36 When a locally submitted message by a trusted user did not contain a
4780 From: header, and the sender address was obtained from -f or from an SMTP
4781 MAIL command, and the trusted user did not use -F to supply a sender
4782 name, $originator_name was incorrectly used when constructing a From:
4783 header. Furthermore, $originator_name was used for submission mode
4784 messages from external hosts without From: headers in a similar way,
4785 which is clearly wrong.
4787 PH/37 Added control=suppress_local_fixups.
4789 PH/38 When log_selector = +received_sender was set, and the addition of the
4790 sender made the log line's construction buffer exactly full, or one byte
4791 less than full, an overflow happened when the terminating "\n" was
4794 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
4795 when the result of a list match is failure because a DNS lookup failed.
4797 PH/40 RM_COMMAND is now used in the building process.
4799 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
4800 the "build-* directories that it finds.
4802 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
4803 address in a domain literal was a prefix of an interface address.
4805 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
4806 when verifying a sender address, unless rewrite_headers is false.
4808 PH/44 Wrote a long comment about why errors_to addresses are verified as
4809 recipients, not senders.
4811 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
4812 the ratelimit ACL was added.
4814 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
4816 PH/46 Added extra information about PostgreSQL errors to the error string.
4818 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
4819 avoiding the need to install special zones in a real server. This is
4820 backwards compatible; if it can't find the fake resolver, it drops back.
4821 Thus, both old and new test suites can be run.
4823 TF/02 Added util/ratelimit.pl
4825 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
4826 clock is set back in time.
4828 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
4829 Candler <B.Candler@pobox.com>.
4831 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
4832 OK for addresses that are the result of redirections.
4834 PH/48 A number of further additions for the benefit of the new test suite,
4835 including a fake gethostbyname() that interfaces to the fake DNS resolver
4838 TF/06 The fix for widen_domains has also been applied to qualify_single and
4839 search_parents which are the other dnslookup options that can cause
4842 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
4845 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
4847 TF/07 Exim produced the error message "an SRV record indicated no SMTP
4848 service" if it encountered an MX record with an empty target hostname.
4849 The message is now "an MX or SRV record indicated no SMTP service".
4851 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
4852 if the DNS of the sending site is misconfigured. This is quite a
4853 common situation. This change restores the behaviour of treating a
4854 helo verification defer as a failure.
4856 PH/51 If self=fail was set on a router, the bounce message did not include the
4857 actual error message.
4863 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
4865 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4866 command itself ended in error, the underlying message about the transport
4867 filter timeout was being overwritten with the pipe command error. Now the
4868 underlying error message should be appended to the second error message.
4870 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4872 PH/02 Reduce the amount of output that "make" produces by default. Full output
4873 can still be requested.
4875 PH/03 The warning log line about a condition test deferring for a "warn" verb
4876 was being output only once per connection, rather than after each
4877 occurrence (because it was using the same function as for successful
4878 "warn" verbs). This seems wrong, so I have changed it.
4880 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4881 it should not have, which might have caused a crash in the right
4882 circumstances, but probably never did.
4884 PH/04 Installed a modified version of Tony Finch's patch to make submission
4885 mode fix the return path as well as the Sender: header line, and to
4886 add a /name= option so that you can make the user's friendly name appear
4889 TF/03 Added the control = fakedefer ACL modifier.
4891 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4892 Mark Lowes for thorough testing.
4894 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4896 TK/03 Merged latest SRS patch from Miles Wilton.
4898 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4899 with the definition in sysexits.h (which is #included earlier).
4900 Fortunately, Exim does not actually use EX_OK. The code used to try to
4901 preserve the sysexits.h value, by assuming that macro definitions were
4902 scanned for macro replacements. I have been disabused of this notion,
4903 so now the code just undefines EX_OK before #including unistd.h.
4905 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4906 in the smtp transport. When a block could not be written in a single
4907 write() function, the timeout was being re-applied to each part-write.
4908 This seems wrong - if the receiver was accepting one byte at a time it
4909 would take for ever. The timeout is now adjusted when this happens. It
4910 doesn't have to be particularly precise.
4912 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4913 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4915 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4916 without a domain if the check to <postmaster@domain> fails.
4918 SC/01 Eximstats: added -xls and the ability to specify output files
4919 (patch written by Frank Heydlauf).
4921 SC/02 Eximstats: use FileHandles for outputting results.
4923 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4925 SC/04 Eximstats: fixed display of large numbers with -nvr option
4927 SC/05 Eximstats: fixed merging of reports with empty tables.
4929 SC/06 Eximstats: added the -include_original_destination flag
4931 SC/07 Eximstats: removed tabs and trailing whitespace.
4933 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4935 TK/06 MBOX spool code: Add real "From " MBOX separator line
4936 so the .eml file is really in mbox format (even though
4937 most programs do not really care). Patch from Alex Miller.
4939 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4940 The latter is generated from $received_to and is only set if the
4941 message has one envelope recipient. SA can use these headers,
4942 obviously out-of-the-box. Patch from Alex Miller.
4944 PH/08 The ${def test on a variable was returning false if the variable's
4945 value was "0", contrary to what the specification has always said!
4946 The result should be true unless the variable is empty.
4948 PH/09 The syntax error of a character other than { following "${if
4949 def:variable_name" (after optional whitespace) was not being diagnosed.
4950 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4951 accidental colon was present, for example, could give incorrect results.
4953 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4954 result is used (not including appendfile, where other changes are about
4957 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4958 This involved changing a lot of size variables from int to off_t. It
4959 should work with maildirs and everything.
4961 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4962 spamd dying while we are connected to it.
4964 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4967 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4968 function for BDB 4.3.
4970 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4972 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4973 the sources for PCRE have been re-arranged and more files are now
4976 PH/15 The code I had for printing potentially long long variables in PH/11
4977 above was not the best (it lost precision). The length of off_t variables
4978 is now inspected at build time, and an appropriate printing format (%ld
4979 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4980 to be "long long int" or "long int". This is needed for the internal
4981 formatting function string_vformat().
4983 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4984 the configuration file to be ":syslog", then the script "guesses" where
4985 the logs files are, rather than using the compiled in default. In our
4986 case the guess is not the same as the compiled default, so the script
4987 suddenly stopped working when I started to use syslog. The patch checks
4988 to see if log_file_path is "". If so, it attempts to read it from exim
4989 with no configuration file to get the compiled in version, before it
4990 falls back to the previous guessing code."
4992 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4993 implementing BATV in an Exim configuration. See NewStuff for the gory
4996 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4997 Makefile that are specific to HP-UX.
4999 PH/18 If the "use_postmaster" option was set for a recipient callout together
5000 with the "random" option, the postmaster address was used as the MAIL
5001 FROM address for the random test, but not for the subsequent recipient
5002 test. It is now used for both.
5004 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
5005 patch removes a few documentation additions to RFC 3028, because the
5006 latest draft now contains them. It adds the new en;ascii-case comparator
5007 and a new error check for 8bit text in MIME parts. Comparator and
5008 require names are now matched exactly. I enabled the subaddress
5009 extension, but it is not well tested yet (read: it works for me)."
5011 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
5012 rework some of the code of TK/09 above to avoid the hardwired use of
5013 "%lld" and "long long". Replaced the call to snprintf() with a call to
5016 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
5017 records point to non-existent hosts", "retry timeout exceeded", and
5018 "retry time not reached for any host after a long failure period".
5020 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
5021 experimental DomainKeys support:
5023 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
5024 (2) On an error such as an illegally used "control", the wrong name for
5025 the control was given.
5027 These problems did NOT occur unless DomainKeys support was compiled.
5029 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
5031 PH/24 Added ${if match_ip condition.
5033 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
5034 databases so that it will be absolutely obvious if a crash occurs in the
5035 DB library. This is a regular occurrence (often caused by mis-matched
5038 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
5039 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
5040 picked up on a user's system that detects such things. There doesn't seem
5041 to be a gcc warning option for this - only an attribute that has to be
5042 put on the function's prototype. It seems that in Fedora Core 4 they have
5043 set this on a number of new functions. No doubt there will be more in due
5046 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
5047 specify a transport. However, if an address that was verified by such a
5048 router was the subject of a callout, Exim crashed because it tried to
5049 read the rcpt_include_affixes from the non-existent transport. Now it
5050 just assumes that the setting of that option is false. This bug was
5051 introduced by 4.51/PH/31.
5053 PH/28 Changed -d+all to exclude +memory, because that information is very
5054 rarely of interest, but it makes the output a lot bigger. People tend to
5055 do -d+all out of habit.
5057 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
5058 code in os-type was giving problems when libc.so lives in lib64, like on
5061 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
5062 aren't the modern standard, and it seems that some systems' include files
5063 don't always have them. Exim was already checking for some of the newer
5064 ones like T_AAAA, and defining it itself. I've added checks for all the
5065 record types that Exim uses.
5067 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
5068 not automatically generating a new one, as it is supposed to. This
5069 prevented TLS from working. If the file did exist, but contained invalid
5070 data, a new version was generated, as expected. It was only the case of a
5071 non-existent file that was broken.
5073 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
5074 with a change in libdomainkeys > 0.64.
5076 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
5077 from DNS. If the selector record carries the flag, it now has
5078 precedence over the domain-wide flag.
5080 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
5082 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
5083 the use of a port name that isn't defined in /etc/services) occurred, the
5084 message was deferred as in a normal delivery, and thus remained on the
5085 spool, instead of being failed because of the mua_wrapper setting. This
5086 is now fixed, and I tidied up some of the mua_wrapper messages at the
5089 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
5090 the messages in a hash of arrays rather than using individual hashes.
5091 This is a bit cleaner and results in dramatic memory savings, albeit
5092 at a slight CPU cost.
5094 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
5095 as requested by Marc Sherman.
5097 SC/10 Eximstats: added histograms for user specified patterns as requested
5100 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
5102 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
5103 fopen() in the content-scanning modules that did not already have it.
5109 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
5110 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
5112 TK/02 Fix ACL "control" statement not being available in MIME ACL.
5114 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
5116 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
5117 to test Sieve filters that use "vacation".
5119 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
5120 that changes the way the GnuTLS parameters are stored in the cache file.
5121 The new format can be generated externally. For backward compatibility,
5122 if the data in the cache doesn't make sense, Exim assumes it has read an
5123 old-format file, and it generates new data and writes a new file. This
5124 means that you can't go back to an older release without removing the
5127 PH/03 A redirect router that has both "unseen" and "one_time" set does not
5128 work if there are any delivery delays because "one_time" forces the
5129 parent to be marked "delivered", so its unseen clone is never tried
5130 again. For this reason, Exim now forbids the simultaneous setting of
5133 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
5134 redirected to themselves ("homonym" addresses). Read the long ChangeLog
5135 entry if you want to know the details. The fix, however, neglected to
5136 consider the case when local delivery batching is involved. The test for
5137 "previously delivered" was not happening when checking to see if an
5138 address could be batched with a previous (undelivered) one; under
5139 certain circumstances this could lead to multiple deliveries to the same
5142 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
5143 in its include files, and this causes problems building Exim.
5145 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
5146 header_syntax) but Exim was just ignoring anything given after a slash.
5147 In particular, this caused confusion with an attempt to use "verify =
5148 reverse_host_lookup/defer_ok". An error is now given when options are
5149 supplied for verify items that do not have them. (Maybe reverse_host_
5150 lookup should have a defer_ok option, but that's a different point.)
5152 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
5153 defined by RFC 821) to 2048, because there were problems with some AUTH
5154 commands, and RFC 1869 says the size should be increased for extended
5155 SMTP commands that take arguments.
5157 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
5160 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
5161 "unknown" error; now it says that the functionality isn't in the binary.
5163 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
5164 an address' error message when a string expansion fails (syntax or
5165 whatever). Otherwise the password may appear in the log. Following change
5166 PH/42 below, there is no longer a chance of it appearing in a bounce
5169 PH/11 Installed exipick version 20050225.0 from John Jetmore.
5171 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
5172 of its addresses was ever tried. (Bugzilla bug #2.)
5174 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
5175 the result incorrectly in the debug output. (It correctly added a newline
5176 to what was transported.)
5178 TF/01 Added $received_time.
5180 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
5181 commented out examples of how to interface to a virus scanner and to
5182 SpamAssassin. Also added commented examples of av_scanner and
5183 spamd_address settings.
5185 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
5186 and controls are allowed in which ACLs. There were a couple of minor
5187 errors. Some of the entries in the conditions table (which is a table of
5188 where they are NOT allowed) were getting very unwieldy; rewrote them as a
5189 negation of where the condition IS allowed.
5191 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
5193 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
5194 header file does not have a version number, so I've had to invent a new
5195 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
5196 API. The code is untested by me (my Linux distribution still has 0.3.2 of
5197 radiusclient), but it was contributed by a Radius user.
5199 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
5200 files or directories, for OpenSSL.
5202 PH/19 When an Exim process that is running as root has to create an Exim log
5203 file, it does so in a subprocess that runs as exim:exim so as to get the
5204 ownership right at creation (otherwise, other Exim processes might see
5205 the file with the wrong ownership). There was no test for failure of this
5206 fork() call, which would lead to the process getting stuck as it waited
5207 for a non-existent subprocess. Forks do occasionally fail when resources
5208 run out. I reviewed all the other calls to fork(); they all seem to check
5211 PH/20 When checking for unexpected SMTP input at connect time (before writing
5212 the banner), Exim was not dealing correctly with a non-positive return
5213 from the read() function. If the client had disconnected by this time,
5214 the result was a log entry for a synchronization error with an empty
5215 string after "input=" when read() returned zero. If read() returned -1
5216 (an event I could not check), uninitialized data bytes were printed.
5217 There were reports of junk text (parts of files, etc) appearing after
5220 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
5222 PH/22 Added support for macro redefinition, and (re)definition in between
5223 driver and ACL definitions.
5225 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
5226 forgetting to use the resulting value; it was using the unexpanded value.
5228 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
5229 hadn't been configured. The fix is from Juergen Kreileder, who
5230 understands it better than I do:
5232 "Here's what I see happening with three configured cyrus_sasl
5233 authenticators configured (plain, login, cram-md5):
5235 On startup auth_cyrus_sasl_init() gets called for each of these.
5236 This means three calls to sasl_listmech() without a specified mech_list.
5237 => SASL tests which mechs of all available mechs actually work
5238 => three warnings about OTP not working
5239 => the returned list contains: plain, login, cram-md5, digest-md5, ...
5241 With the patch, sasl_listmech() also gets called three times. But now
5242 SASL's mech_list option is set to the server_mech specified in the the
5243 authenticator. Or in other words, the answer from sasl_listmech()
5244 gets limited to just the mech you're testing for (which is different
5246 => the return list contains just 'plain' or 'login', 'cram-md5' or
5247 nothing depending on the value of ob->server_mech.
5249 I've just tested the patch: Authentication still works fine,
5250 unavailable mechs specified in the exim configuration are still
5251 caught, and the auth.log warnings about OTP are gone."
5253 PH/25 When debugging is enabled, the contents of the command line are added
5254 to the debugging output, even when log_selector=+arguments is not
5257 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
5258 answer is "GNU", and only if the return is "GNU/something" is the answer
5261 PH/27 $acl_verify_message is now set immediately after the failure of a
5262 verification in an ACL, and so is available in subsequent modifiers. In
5263 particular, the message can be preserved by coding like this:
5265 warn !verify = sender
5266 set acl_m0 = $acl_verify_message
5268 Previously, $acl_verify_message was set only while expanding "message"
5269 and "log_message" when a very denied access.
5271 PH/28 Modified OS/os.c-Linux with
5273 -#ifndef OS_LOAD_AVERAGE
5274 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
5276 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
5277 nomenclature these days.)
5279 PH/29 Installed patch from the Sieve maintainer that adds the options
5280 sieve_useraddress and sieve_subaddress to the redirect router.
5282 PH/30 In these circumstances:
5283 . Two addresses routed to the same list of hosts;
5284 . First host does not offer TLS;
5285 . First host accepts first address;
5286 . First host gives temporary error to second address;
5287 . Second host offers TLS and a TLS session is established;
5288 . Second host accepts second address.
5289 Exim incorrectly logged both deliveries with the TLS parameters (cipher
5290 and peerdn, if requested) that were in fact used only for the second
5293 PH/31 When doing a callout as part of verifying an address, Exim was not paying
5294 attention to any local part prefix or suffix that was matched by the
5295 router that accepted the address. It now behaves in the same way as it
5296 does for delivery: the affixes are removed from the local part unless
5297 rcpt_include_affixes is set on the transport.
5299 PH/32 Add the sender address, as F=<...>, to the log line when logging a
5300 timeout during the DATA phase of an incoming message.
5302 PH/33 Sieve envelope tests were broken for match types other than :is. I have
5303 applied a patch sanctioned by the Sieve maintainer.
5305 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
5306 the uid or gid is negative. A case of a negative gid caused this to be
5307 noticed. The fix allows for either to be negative.
5309 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
5310 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
5313 PH/36 The change PH/12 above was broken. Fixed it.
5315 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
5316 the grounds that routing the same address twice would always produce the
5317 same answer. This might have been true once, but it is certainly no
5318 longer true now. Routing a child address may depend on the previous
5319 routing that produced that child. Some complicated redirection strategies
5320 went wrong when messages had multiple recipients, and made Exim's
5321 behaviour dependent on the order in which the addresses were given.
5323 I have moved the duplicate checking until after the routing is complete.
5324 Exim scans the addresses that are assigned to local and remote
5325 transports, and removes any duplicates. This means that more work will be
5326 done, as duplicates will always all be routed, but duplicates are
5327 presumably rare, so I don't expect this is of any significance.
5329 For deliveries to pipes, files, and autoreplies, the duplicate checking
5330 still happens during the routing process, since they are not going to be
5333 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
5334 It corrects a timeout issue with spamd. This is Ian's comment: "The
5335 background is that sometimes spamd either never reads data from a
5336 connection it has accepted, or it never writes response data. The exiscan
5337 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
5338 blindly assumes that writes won't block so it may never time out."
5340 PH/39 Allow G after quota size as well as K and M.
5342 PH/40 The value set for $authenticated_id in an authenticator may not contain
5343 binary zeroes or newlines because the value is written to log lines and
5344 to spool files. There was no check on this. Now the value is run through
5345 the string_printing() function so that such characters are converted to
5346 printable escape sequences.
5348 PH/41 $message_linecount is a new variable that contains the total number of
5349 lines in the message. Compare $body_linecount, which is the count for the
5352 PH/42 Exim no longer gives details of delivery errors for specific addresses in
5353 bounce and delay warning messages, except in certain special cases, which
5356 (a) An SMTP error message from a remote host;
5357 (b) A message specified in a :fail: redirection;
5358 (c) A message specified in a "fail" command in a system filter;
5359 (d) A message specified in a FAIL return from the queryprogram router;
5360 (e) A message specified by the cannot_route_message router option.
5362 In these cases only, Exim does include the error details in bounce and
5363 warning messages. There are also a few cases where bland messages such
5364 as "unrouteable address" or "local delivery error" are given.
5366 PH/43 $value is now also set for the "else" part of a ${run expansion.
5368 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
5369 being worked on, but at least Exim now implements the latest version to
5372 PH/45 In a pipe transport, although a timeout while waiting for the pipe
5373 process to complete was treated as a delivery failure, a timeout while
5374 writing the message to the pipe was logged, but erroneously treated as a
5375 successful delivery. Such timeouts include transport filter timeouts. For
5376 consistency with the overall process timeout, these timeouts are now
5377 treated as errors, giving rise to delivery failures by default. However,
5378 there is now a new Boolean option for the pipe transport called
5379 timeout_defer, which, if set TRUE, converts the failures into defers for
5380 both kinds of timeout. A transport filter timeout is now identified in
5383 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
5384 systems where "make" and "gmake" are different, calling "gmake" at top
5385 level broke things. I've arranged for the value of $(MAKE) to be passed
5386 from the Makefile to this script so that it can call the same version of
5390 A note about Exim versions 4.44 and 4.50
5391 ----------------------------------------
5393 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
5394 changes of various kinds. As a consequence, a big documentation update was
5395 needed. This delayed the release for rather longer than seemed good, especially
5396 in the light of a couple of (minor) security issues. Therefore, the changes
5397 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
5398 release. So 4.44 and 4.50 are in effect two different branches that both start
5401 I have left the 4.50 change log unchanged; it contains all the changes since
5402 4.43. The change log for 4.44 is below; many of its items are identical to
5403 those for 4.50. This seems to be the most sensible way to preserve the
5404 historical information.
5410 1. Minor wording change to the doc/README.SIEVE file.
5412 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
5413 computation of the current number of files was incorrect.
5415 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5416 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5419 4. Give more explanation in the error message when the command for a transport
5420 filter fails to execute.
5422 5. There are several places where Exim runs a non-Exim command in a
5423 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5424 was being done only for the command run by the queryprogram router. It is
5425 now done for all such subprocesses. The other cases are: ${run, transport
5426 filters, and the commands run by the lmtp and pipe transports.
5428 6. Added CONFIGURE_GROUP build-time option.
5430 7. Some older OS have a limit of 256 on the maximum number of file
5431 descriptors. Exim was using setrlimit() to set 1000 as a large value
5432 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5433 systems. I've change it so that if it can't get 1000, it tries for 256.
5435 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5436 was an oversight, and furthermore, ever since the addition of extra
5437 controls (e.g. 4.43/32), the checks on when to allow different forms of
5438 "control" were broken. There should now be diagnostics for all cases when a
5439 control that does not make sense is encountered.
5441 9. Added the /retain_sender option to "control=submission".
5443 10. $recipients is now available in the predata ACL (oversight).
5445 11. Tidy the search cache before the fork to do a delivery from a message
5446 received from the command line. Otherwise the child will trigger a lookup
5447 failure and thereby defer the delivery if it tries to use (for example) a
5448 cached ldap connection that the parent has called unbind on.
5450 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5451 of $address_data from the recipient verification was clobbered by the
5452 sender verification.
5454 13. The value of address_data from a sender verification is now available in
5455 $sender_address_data in subsequent conditions in the ACL statement.
5457 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
5459 15. Added a new option "connect=<time>" to callout options, to set a different
5462 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5463 was its contents. (It was OK if the option was not defined at all.)
5465 17. A "Completed" log line is now written for messages that are removed from
5466 the spool by the -Mrm option.
5468 18. New variables $sender_verify_failure and $recipient_verify_failure contain
5469 information about exactly what failed.
5471 19. Added -dd to debug only the daemon process.
5473 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
5474 handles timeouts, both on the server side and network timeouts. Renamed the
5475 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
5477 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
5478 It is now set to "smtps".
5480 22. $host_address is now set to the target address during the checking of
5481 ignore_target_hosts.
5483 23. When checking ignore_target_hosts for an ipliteral router, no host name was
5484 being passed; this would have caused $sender_host_name to have been used if
5485 matching the list had actually called for a host name (not very likely,
5486 since this list is usually IP addresses). A host name is now passed as
5489 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5490 code that specifies a non-restarting handler (typically sigaction() in
5491 modern systems) in an attempt to fix a rare and obscure crash bug.
5493 25. Narrowed the window for a race in the daemon that could cause it to ignore
5494 SIGCHLD signals. This is not a major problem, because they are used only to
5495 wake it up if nothing else does.
5497 26. A malformed maildirsize file could cause Exim to calculate negative values
5498 for the mailbox size or file count. Odd effects could occur as a result.
5499 The maildirsize information is now recalculated if the size or filecount
5502 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5503 support for a long time. Removed HAVE_SYS_VFS_H.
5505 28. Installed the latest version of exipick from John Jetmore.
5507 29. In an address list, if the pattern was not a regular expression, an empty
5508 subject address (from a bounce message) matched only if the pattern was an
5509 empty string. Non-empty patterns were not even tested. This was the wrong
5510 because it is perfectly reasonable to use an empty address as part of a
5511 database query. An empty address is now tested by patterns that are
5512 lookups. However, all the other forms of pattern expect the subject to
5513 contain a local part and a domain, and therefore, for them, an empty
5514 address still always fails if the pattern is not itself empty.
5516 30. Exim went into a mad DNS loop when attempting to do a callout where the
5517 host was specified on an smtp transport, and looking it up yielded more
5518 than one IP address.
5520 31. Re-factored the code for checking spool and log partition space into a
5521 function that finds that data and another that does the check. The former
5522 is then used to implement four new variables: $spool_space, $log_space,
5523 $spool_inodes, and $log_inodes.
5525 32. The RFC2047 encoding function was originally intended for short strings
5526 such as real names; it was not keeping to the 75-character limit for
5527 encoded words that the RFC imposes. It now respects the limit, and
5528 generates multiple encoded words if necessary. To be on the safe side, I
5529 have increased the buffer size for the ${rfc2047: expansion operator from
5532 33. It is now permitted to omit both strings after an "if" condition; if the
5533 condition is true, the result is "true". As before, when the second string
5534 is omitted, a false condition yields an empty string. This makes it less
5535 cumbersome to write custom ACL and router conditions.
5537 34. Failure to deliver a bounce message always caused it to be frozen, even if
5538 there was an errors_to setting on the router. The errors_to setting is now
5541 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5542 canonical form (fully expanded) before being placed in
5543 $sender_host_address.
5545 36. The table in the code that translates DNS record types into text (T_A to
5546 "A" for instance) was missing entries for NS and CNAME. It is just possible
5547 that this could have caused confusion if both these types were looked up
5548 for the same domain, because the text type is used as part of Exim's
5549 per-process caching. But the chance of anyone hitting this buglet seems
5552 37. The dnsdb lookup has been extended in a number of ways.
5554 (1) There is a new type, "zns", which walks up the domain tree until it
5555 finds some nameserver records. It should be used with care.
5557 (2) There is a new type, "mxh", which is like "mx" except that it returns
5558 just the host names, not the priorities.
5560 (3) It is now possible to give a list of domains (or IP addresses) to be
5561 looked up. The behaviour when one of the lookups defers can be
5562 controlled by a keyword.
5564 (4) It is now possible to specify the separator character for use when
5565 multiple records are returned.
5567 38. The dnslists ACL condition has been extended: it is now possible to supply
5568 a list of IP addresses and/or domains to be looked up in a particular DNS
5571 39. Added log_selector=+queue_time_overall.
5573 40. When running the queue in the test harness, wait just a tad after forking a
5574 delivery process, to get repeatability of debugging output.
5576 41. Include certificate and key file names in error message when GnuTLS fails
5577 to set them up, because the GnuTLS error message doesn't include the name
5578 of the failing file when there is a problem reading it.
5580 42. Allow both -bf and -bF in the same test run.
5582 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
5584 44. The "Exiscan patch" is now merged into the mainline Exim source.
5586 45. Sometimes the final signoff response after QUIT could fail to get
5587 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5588 < 0 before doing a fflush(). This bug looks as though it goes back to the
5589 introduction of TLS in release 3.20, but "sometimes" must have been rare
5590 because the tests only now provoked it.
5592 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
5593 (this can affect the format of dates).
5595 47. exim_tidydb, when checking for the continued existence of a message for
5596 which it has found a message-specific retry record, was not finding
5597 messages that were in split spool directories. Consequently, it was
5598 deleting retry records that should have stayed in existence.
5600 48. Steve fixed some bugs in eximstats.
5602 49. The SPA authentication driver was not abandoning authentication and moving
5603 on to the next authenticator when an expansion was forced to fail,
5604 contradicting the general specification for all authenticators. Instead it
5605 was generating a temporary error. It now behaves as specified.
5607 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
5608 (the order specifies the preference for clients). The order is now AES256,
5609 AES128, 3DES, ARCFOUR128.
5611 51. Small patch to Sieve code - explicitly set From: when generating an
5614 52. Exim crashed if a remote delivery caused a very long error message to be
5615 recorded - for instance if somebody sent an entire SpamAssassin report back
5616 as a large number of 550 error lines. This bug was coincidentally fixed by
5617 increasing the size of one of Exim's internal buffers (big_buffer) that
5618 happened as part of the Exiscan merge. However, to be on the safe side, I
5619 have made the code more robust (and fixed the comments that describe what
5622 53. Now that there can be additional text after "Completed" in log lines (if
5623 the queue_time_overall log selector is set), a one-byte patch to exigrep
5624 was needed to allow it to recognize "Completed" as not the last thing in
5627 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
5628 patch that reportedly fixes this has been added. I am not expert enough to
5629 create a test for it. This is what the patch creator wrote:
5631 "I found a little strange behaviour of ldap code when working with
5632 Windows 2003 AD Domain, where users was placed in more than one
5633 Organization Units. When I tried to give exim partial DN, the exit code
5634 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
5635 But simultaneously result of request was absolutely normal ldap result,
5636 so I produce this patch..."
5638 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
5639 REFERENCE, so I have modified the code to exclude the patch when that macro
5642 55. Some experimental protocols are using DNS PTR records for new purposes. The
5643 keys for these records are domain names, not reversed IP addresses. The
5644 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5645 leaves it alone. Component reversal etc. now happens only for IP addresses.
5648 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
5650 57. Double the size of the debug message buffer (to 2048) so that more of very
5651 long debug lines gets shown.
5653 58. The exicyclog utility now does better if the number of log files to keep
5654 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
5656 59. Two changes related to the smtp_active_hostname option:
5658 (1) $smtp_active_hostname is now available as a variable.
5659 (2) The default for smtp_banner uses $smtp_active_hostname instead
5660 of $primary_hostname.
5662 60. The host_aton() function is supposed to be passed a string that is known
5663 to be a valid IP address. However, in the case of IPv6 addresses, it was
5664 not checking this. This is a hostage to fortune. Exim now panics and dies
5665 if the condition is not met. A case was found where this could be provoked
5666 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5667 components; fortuitously, this particular loophole had already been fixed
5668 by change 4.50/55 above.
5670 If there are any other similar loopholes, the new check in host_aton()
5671 itself should stop them being exploited. The report I received stated that
5672 data on the command line could provoke the exploit when Exim was running as
5673 exim, but did not say which command line option was involved. All I could
5674 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5675 running as the user.
5678 61. There was a buffer overflow vulnerability in the SPA authentication code
5679 (which came originally from the Samba project). I have added a test to the
5680 spa_base64_to_bits() function which I hope fixes it.
5683 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
5684 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
5686 63. The daemon start-up calls getloadavg() while still root for those OS that
5687 need the first call to be done as root, but it missed one case: when
5688 deliver_queue_load_max is set with deliver_drop_privilege. This is
5689 necessary for the benefit of the queue runner, because there is no re-exec
5690 when deliver_drop_privilege is set.
5692 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
5693 This has been fixed.
5695 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5696 was in use, was not putting the data itself into the right store pool;
5697 consequently, it could be overwritten for a subsequent message in the same
5698 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5701 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
5703 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
5704 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
5705 the function were treating the return as a boolean value, which happened to
5706 work because 0=false and not-0=true, but is not correct code.
5708 68. The host_aton() function was not handling scoped IPv6 addresses (those
5709 with, for example, "%eth0" on the end) correctly.
5711 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
5712 negated items (that is, ~something) in unsigned ints. Some compilers
5713 apparently mutter when there is no cast.
5715 70. If an address verification called from an ACL failed, and did not produce a
5716 user-specific message (i.e. there was only a "system" message), nothing was
5717 put in $acl_verify_message. In this situation, it now puts the system
5720 71. Change 4.23/11 added synchronization checking at the start of an SMTP
5721 session; change 4.31/43 added the unwanted input to the log line - except
5722 that it did not do this in the start of session case. It now does.
5724 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
5725 This is wrong and can cause the other end to generate a synchronization
5726 error if it is another Exim or anything else that does the synchronization
5727 check. A QUIT command is no longer sent after a timeout.
5729 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
5730 during host lookups.
5732 74. The defer_ok option of callout verification was not working if it was used
5733 when verifying addresses in header lines, that is, for this case:
5735 verify = header_sender/callout=defer_ok
5737 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
5738 those file descriptors could be used for SMTP connections. If anything
5739 wrote to stderr (the example that came up was "warn" in embedded Perl), it
5740 could be sent to the SMTP client, causing chaos. The daemon now opens
5741 stdin, stdout, and stderr to /dev/null when it puts itself into the
5744 76. Arrange for output from Perl's "warn" command to be written to Exim's main
5745 log by default. The user can override this with suitable Perl magic.
5747 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
5748 the log message when discard triggers, was not working for the DATA ACL or
5749 for the non-SMTP ACL.
5751 78. Error message wording change in sieve.c.
5753 79. If smtp_accept_max_per_host was set, the number of connections could be
5754 restricted to fewer than expected, because the daemon was trying to set up
5755 a new connection before checking whether the processes handling previous
5756 connections had finished. The check for completed processes is now done
5757 earlier. On busy systems, this bug wouldn't be noticed because something
5758 else would have woken the daemon, and it would have reaped the completed
5761 80. If a message was submitted locally by a user whose login name contained one
5762 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
5763 It caused a spool format error. I have fixed the spool reading code. A
5764 related problem was that the "from" clause in the Received: line became
5765 illegal because of the space(s). It is now covered by ${quote_local_part.
5767 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
5770 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
5772 83. Patch from Sieve maintainer to fix unterminated string problem in
5773 "vacation" handling.
5775 84. Some minor changes to the Linux configuration files to help with other
5776 OS variants using glibc.
5778 85. One more patch for Sieve to update vacation handling to latest spec.
5781 ----------------------------------------------------
5782 See the note above about the 4.44 and 4.50 releases.
5783 ----------------------------------------------------
5789 1. Change 4.43/35 introduced a bug that caused file counts to be
5790 incorrectly computed when quota_filecount was set in an appendfile
5793 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5794 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5797 3. Give more explanation in the error message when the command for a transport
5798 filter fails to execute.
5800 4. There are several places where Exim runs a non-Exim command in a
5801 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5802 was being done only for the command run by the queryprogram router. It is
5803 now done for all such subprocesses. The other cases are: ${run, transport
5804 filters, and the commands run by the lmtp and pipe transports.
5806 5. Some older OS have a limit of 256 on the maximum number of file
5807 descriptors. Exim was using setrlimit() to set 1000 as a large value
5808 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5809 systems. I've change it so that if it can't get 1000, it tries for 256.
5811 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5812 was an oversight, and furthermore, ever since the addition of extra
5813 controls (e.g. 4.43/32), the checks on when to allow different forms of
5814 "control" were broken. There should now be diagnostics for all cases when a
5815 control that does not make sense is encountered.
5817 7. $recipients is now available in the predata ACL (oversight).
5819 8. Tidy the search cache before the fork to do a delivery from a message
5820 received from the command line. Otherwise the child will trigger a lookup
5821 failure and thereby defer the delivery if it tries to use (for example) a
5822 cached ldap connection that the parent has called unbind on.
5824 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5825 of $address_data from the recipient verification was clobbered by the
5826 sender verification.
5828 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5829 was its contents. (It was OK if the option was not defined at all.)
5831 11. A "Completed" log line is now written for messages that are removed from
5832 the spool by the -Mrm option.
5834 12. $host_address is now set to the target address during the checking of
5835 ignore_target_hosts.
5837 13. When checking ignore_target_hosts for an ipliteral router, no host name was
5838 being passed; this would have caused $sender_host_name to have been used if
5839 matching the list had actually called for a host name (not very likely,
5840 since this list is usually IP addresses). A host name is now passed as
5843 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5844 code that specifies a non-restarting handler (typically sigaction() in
5845 modern systems) in an attempt to fix a rare and obscure crash bug.
5847 15. Narrowed the window for a race in the daemon that could cause it to ignore
5848 SIGCHLD signals. This is not a major problem, because they are used only to
5849 wake it up if nothing else does.
5851 16. A malformed maildirsize file could cause Exim to calculate negative values
5852 for the mailbox size or file count. Odd effects could occur as a result.
5853 The maildirsize information is now recalculated if the size or filecount
5856 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5857 support for a long time. Removed HAVE_SYS_VFS_H.
5859 18. Updated exipick to current release from John Jetmore.
5861 19. Allow an empty sender to be matched against a lookup in an address list.
5862 Previously the only cases considered were a regular expression, or an
5865 20. Exim went into a mad DNS lookup loop when doing a callout where the
5866 host was specified on the transport, if the DNS lookup yielded more than
5869 21. The RFC2047 encoding function was originally intended for short strings
5870 such as real names; it was not keeping to the 75-character limit for
5871 encoded words that the RFC imposes. It now respects the limit, and
5872 generates multiple encoded words if necessary. To be on the safe side, I
5873 have increased the buffer size for the ${rfc2047: expansion operator from
5876 22. Failure to deliver a bounce message always caused it to be frozen, even if
5877 there was an errors_to setting on the router. The errors_to setting is now
5880 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5881 canonical form (fully expanded) before being placed in
5882 $sender_host_address.
5884 24. Updated eximstats to version 1.33
5886 25. Include certificate and key file names in error message when GnuTLS fails
5887 to set them up, because the GnuTLS error message doesn't include the name
5888 of the failing file when there is a problem reading it.
5890 26. Expand error message when OpenSSL has problems setting up cert/key files.
5893 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5894 (this can affect the format of dates).
5896 28. exim_tidydb, when checking for the continued existence of a message for
5897 which it has found a message-specific retry record, was not finding
5898 messages that were in split spool directories. Consequently, it was
5899 deleting retry records that should have stayed in existence.
5901 29. eximstats updated to version 1.35
5902 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5903 1.35 - bugfix such that pie charts by volume are generated correctly
5905 30. The SPA authentication driver was not abandoning authentication and moving
5906 on to the next authenticator when an expansion was forced to fail,
5907 contradicting the general specification for all authenticators. Instead it
5908 was generating a temporary error. It now behaves as specified.
5910 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5911 (the order specifies the preference for clients). The order is now AES256,
5912 AES128, 3DES, ARCFOUR128.
5914 31. Small patch to Sieve code - explicitly set From: when generating an
5917 32. Exim crashed if a remote delivery caused a very long error message to be
5918 recorded - for instance if somebody sent an entire SpamAssassin report back
5919 as a large number of 550 error lines. This bug was coincidentally fixed by
5920 increasing the size of one of Exim's internal buffers (big_buffer) that
5921 happened as part of the Exiscan merge. However, to be on the safe side, I
5922 have made the code more robust (and fixed the comments that describe what
5925 33. Some experimental protocols are using DNS PTR records for new purposes. The
5926 keys for these records are domain names, not reversed IP addresses. The
5927 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5928 leaves it alone. Component reversal etc. now happens only for IP addresses.
5931 34. The host_aton() function is supposed to be passed a string that is known
5932 to be a valid IP address. However, in the case of IPv6 addresses, it was
5933 not checking this. This is a hostage to fortune. Exim now panics and dies
5934 if the condition is not met. A case was found where this could be provoked
5935 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5936 components; fortuitously, this particular loophole had already been fixed
5937 by change 4.50/55 or 4.44/33 above.
5939 If there are any other similar loopholes, the new check in host_aton()
5940 itself should stop them being exploited. The report I received stated that
5941 data on the command line could provoke the exploit when Exim was running as
5942 exim, but did not say which command line option was involved. All I could
5943 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5944 running as the user.
5947 35. There was a buffer overflow vulnerability in the SPA authentication code
5948 (which came originally from the Samba project). I have added a test to the
5949 spa_base64_to_bits() function which I hope fixes it.
5952 36. The daemon start-up calls getloadavg() while still root for those OS that
5953 need the first call to be done as root, but it missed one case: when
5954 deliver_queue_load_max is set with deliver_drop_privilege. This is
5955 necessary for the benefit of the queue runner, because there is no re-exec
5956 when deliver_drop_privilege is set.
5958 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5959 was in use, was not putting the data itself into the right store pool;
5960 consequently, it could be overwritten for a subsequent message in the same
5961 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5964 38. Sometimes the final signoff response after QUIT could fail to get
5965 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5966 < 0 before doing a fflush(). This bug looks as though it goes back to the
5967 introduction of TLS in release 3.20, but "sometimes" must have been rare
5968 because the tests only now provoked it.
5974 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5975 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5976 Now it returns an integer. A number of calls were still expecting a T/F
5977 return. Fortuitously, in all cases, the tests worked in OK situations,
5978 which is the norm. However, things would have gone wrong on any write
5979 failures on the smtp file descriptor. This function is used when sending
5980 messages over SMTP and also when doing verify callouts.
5982 2. When Exim is called to do synchronous delivery of a locally submitted
5983 message (the -odf or -odi options), it no longer closes stderr before doing
5986 3. Implemented the mua_wrapper option.
5988 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5990 5. Implemented the functions header_remove(), header_testname(),
5991 header_add_at_position(), and receive_remove_recipient(), and exported them
5994 6. If an ACL "warn" statement specified the addition of headers, Exim already
5995 inserted X-ACL-Warn: at the start if there was no header name. However, it
5996 was not making this test for the second and subsequent header lines if
5997 there were newlines in the string. This meant that an invalid header could
5998 be inserted if Exim was badly configured.
6000 7. Allow an ACL "warn" statement to add header lines at the start or after all
6001 the Received: headers, as well as at the end.
6003 8. Added the rcpt_4xx retry error code.
6005 9. Added postmaster_mailfrom=xxx to callout verification option.
6007 10. Added mailfrom=xxxx to the callout verification option, for verify=
6010 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
6011 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
6013 12. Inserted some casts to stop certain compilers warning when using pointer
6014 differences as field lengths or precisions in printf-type calls (mostly
6015 affecting debugging statements).
6017 13. Added optional readline() support for -be (dynamically loaded).
6019 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
6020 same clock tick as a message's arrival, so that its received time was the
6021 same as the "first fail" time on the retry record, and that message
6022 remained on the queue past the ultimate address timeout, every queue runner
6023 would try a delivery (because it was past the ultimate address timeout) but
6024 after another failure, the ultimate address timeout, which should have then
6025 bounced the address, did not kick in. This was a "< instead of <=" error;
6026 in most cases the first failure would have been in the next clock tick
6027 after the received time, and all would be well.
6029 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
6030 being recognized when the domain list was tested by the match_domain
6031 condition in an expansion string.
6033 16. Added the ${str2b64: operator.
6035 17. Exim was always calling setrlimit() to set a large limit for the number of
6036 processes, without checking whether the existing limit was already
6037 adequate. (It did check for the limit on file descriptors.) Furthermore,
6038 errors from getrlimit() and setrlimit() were being ignored. Now they are
6039 logged to the main and panic logs, but Exim does carry on, to try to do its
6040 job under whatever limits there are.
6042 18. Imported PCRE 5.0.
6044 19. Trivial typo in log message " temporarily refused connection" (the leading
6047 20. If the log selector return_path_on_delivery was set and an address was
6048 redirected to /dev/null, the delivery process crashed because it assumed
6049 that a return path would always be set for a "successful" delivery. In this
6050 case, the whole delivery is bypassed as an optimization, and therefore no
6053 21. Internal re-arrangement: the function for sending a challenge and reading
6054 a response while authentication was assuming a zero-terminated challenge
6055 string. It's now changed to take a pointer and a length, to allow for
6056 binary data in such strings.
6058 22. Added the cyrus_sasl authenticator (code supplied by MBM).
6060 23. Exim was not respecting finduser_retries when seeking the login of the
6061 uid under which it was called; it was always trying 10 times. (The default
6062 setting of finduser_retries is zero.) Also, it was sleeping after the final
6063 failure, which is pointless.
6065 24. Implemented tls_on_connect_ports.
6067 25. Implemented acl_smtp_predata.
6069 26. If the domain in control=submission is set empty, Exim assumes that the
6070 authenticated id is a complete email address when it generates From: or
6071 Sender: header lines.
6073 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
6074 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
6075 chown and chgrp in /bin and hostname in /usr/bin.
6077 28. Exim was keeping the "process log" file open after each use, just as it
6078 does for the main log. This opens the possibility of it remaining open for
6079 long periods when the USR1 signal hits a daemon. Occasional processlog
6080 errors were reported, that could have been caused by this. Anyway, it seems
6081 much more sensible not to leave this file open at all, so that is what now
6084 29. The long-running daemon process does not normally write to the log once it
6085 has entered its main loop, and it closes the log before doing so. This is
6086 so that log files can straightforwardly be renamed and moved. However,
6087 there are a couple of unusual error situations where the daemon does write
6088 log entries, and I had neglected to close the log afterwards.
6090 30. The text of an SMTP error response that was received during a remote
6091 delivery was being truncated at 512 bytes. This is too short for some of
6092 the long messages that one sometimes sees. I've increased the limit to
6095 31. It is now possible to make retry rules that apply only when a message has a
6096 specific sender, in particular, an empty sender.
6098 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
6099 it possible to be selective about when SMTP synchronization is enforced.
6101 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
6103 32. Implemented hosts_connection_nolog.
6105 33. Added an ACL for QUIT.
6107 34. Setting "delay_warning=" to disable warnings was not working; it gave a
6110 35. Added mailbox_size and mailbox_filecount to appendfile.
6112 36. Added control = no_multiline_responses to ACLs.
6114 37. There was a bug in the logic of the code that waits for the clock to tick
6115 in the case where the clock went backwards by a substantial amount such
6116 that the microsecond fraction of "now" was more than the microsecond
6117 fraction of "then" (but the whole seconds number was less).
6119 38. Added support for the libradius Radius client library this is found on
6120 FreeBSD (previously only the radiusclient library was supported).
6126 1. When certain lookups returned multiple values in the form name=value, the
6127 quoting of the values was not always being done properly. Specifically:
6128 (a) If the value started with a double quote, but contained no whitespace,
6130 (b) If the value contained whitespace other than a space character (i.e.
6131 tabs or newlines or carriage returns) it was not quoted.
6132 This fix has been applied to the mysql and pgsql lookups by writing a
6133 separate quoting function and calling it from the lookup code. The fix
6134 should probably also be applied to nisplus, ibase and oracle lookups, but
6135 since I cannot test any of those, I have not disturbed their existing code.
6137 2. A hit in the callout cache for a specific address caused a log line with no
6138 reason for rejecting RCPT. Now it says "Previous (cached) callout
6139 verification failure".
6141 3. There was an off-by-one bug in the queryprogram router. An over-long
6142 return line was truncated at 256 instead of 255 characters, thereby
6143 overflowing its buffer with the terminating zero. As well as fixing this, I
6144 have increased the buffer size to 1024 (and made a note to document this).
6146 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
6147 when Exim is waiting for an SMTP response from a remote server, Exim
6148 restarts its select() call on the socket, thereby resetting its timeout.
6149 This is not a problem when such interrupts are rare. Somebody set up a cron
6150 job to run exiwhat every 2 minutes, which is less than the normal select()
6151 timeout (5 or 10 minutes). This meant that the select() timeout never
6152 kicked in because it was always reset. I have fixed this by comparing the
6153 time when an interrupt arrives with the time at the start of the first call
6154 to select(). If more time than the timeout has elapsed, the interrupt is
6155 treated as a timeout.
6157 5. Some internal re-factoring in preparation for the addition of Sieve
6158 extensions (by MH). In particular, the "personal" test is moved to a
6159 separate function, and given an option for scanning Cc: and Bcc: (which is
6160 not set for Exim filters).
6162 6. When Exim created an email address using the login of the caller as the
6163 local part (e.g. when creating a From: or Sender: header line), it was not
6164 quoting the local part when it contained special characters such as @.
6166 7. Installed new OpenBSD configuration files.
6168 8. Reworded some messages for syntax errors in "and" and "or" conditions to
6169 try to make them clearer.
6171 9. Callout options, other than the timeout value, were being ignored when
6172 verifying sender addresses in header lines. For example, when using
6174 verify = header_sender/callout=no_cache
6176 the cache was (incorrectly) being used.
6178 10. Added a missing instance of ${EXE} to the exim_install script; this affects
6179 only the Cygwin environment.
6181 11. When return_path_on_delivery was set as a log selector, if different remote
6182 addresses in the same message used different return paths and parallel
6183 remote delivery occurred, the wrong values would sometimes be logged.
6184 (Whenever a remote delivery process finished, the return path value from
6185 the most recently started remote delivery process was logged.)
6187 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
6188 lines when AUTH and/or TLS are in use. This is the "received protocol"
6189 field. Exim used to use "asmtp" for authenticated SMTP, without any
6190 indication (in the protocol name) for TLS use. Now it follows the RFC and
6191 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
6192 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
6193 names appear in log lines as well as in Received: header lines.
6195 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
6196 extensions, and comparison tests, and to fix some bugs.
6198 14. Changes to the "personal" filter test:
6200 (1) The test was buggy in that it was just doing the equivalent of
6201 "contains" tests on header lines. For example, if a user's address was
6202 anne@some.where, the "personal" test would incorrectly be true for
6204 To: susanne@some.where
6206 This test is now done by extracting each address from the header in turn,
6207 and checking the entire address. Other tests that are part of "personal"
6208 are now done using regular expressions (for example, to check local parts
6209 of addresses in From: header lines).
6211 (2) The list of non-personal local parts in From: addresses has been
6212 extended to include "listserv", "majordomo", "*-request", and "owner-*",
6213 taken from the Sieve specification recommendations.
6215 (3) If the message contains any header line starting with "List-" it is
6216 treated as non-personal.
6218 (4) The test for "circular" in the Subject: header line has been removed
6219 because it now seems ill-conceived.
6221 15. Minor typos in src/EDITME comments corrected.
6223 16. Installed latest exipick from John Jetmore.
6225 17. If headers_add on a router specified a text string that was too long for
6226 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
6227 of string_sprintf() is now avoided.
6229 18. $message_body_size was not set (it was always zero) when running the DATA
6230 ACL and the local_scan() function.
6232 19. For the "mail" command in an Exim filter, no default was being set for
6233 the once_repeat time, causing a random time value to be used if "once" was
6234 specified. (If the value happened to be <= 0, no repeat happened.) The
6235 default is now 0s, meaning "never repeat". The "vacation" command was OK
6236 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
6237 (I found it when inspecting the code).
6239 20. There is now an overall timeout for performing a callout verification. It
6240 defaults to 4 times the callout timeout, which applies to individual SMTP
6241 commands during the callout. The overall timeout applies when there is more
6242 than one host that can be tried. The timeout is checked before trying the
6243 next host. This prevents very long delays if there are a large number of
6244 hosts and all are timing out (e.g. when the network connections are timing
6245 out). The value of the overall timeout can be changed by specifying an
6246 additional sub-option for "callout", called "maxwait". For example:
6248 verify = sender/callout=5s,maxwait=20s
6250 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
6251 to the end before writing, but this should make it even safer).
6253 22. Exim was forgetting that it had advertised PIPELINING for the second and
6254 subsequent messages on an SMTP connection. It was also not resetting its
6255 memory on STARTTLS and an internal HELO.
6257 23. When Exim logs an SMTP synchronization error within a session, it now
6258 records whether PIPELINING has been advertised or not.
6260 24. Added 3 instances of "(long int)" casts to time_t variables that were being
6261 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
6262 rather than long int.
6264 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
6266 26. Added the never_mail option to autoreply.
6272 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
6273 crash if the getsockname() call failed; this can happen if a connection is
6274 closed very soon after it is established. The problem was simply in the
6275 order in which certain operations were done, causing Exim to try to write
6276 to the SMTP stream before it had set up the file descriptor. The bug has
6277 been fixed by making things happen in the correct order.
6283 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
6284 before the connection was closed, thus losing the rejection response.
6286 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
6287 some early Solaris releases, but causes trouble in current releases where
6288 socklen_t is defined.
6290 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
6293 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
6296 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
6297 that releasing the top of it at the end releases what was used for sub-
6298 expansions (unless the block got too big). However, discard this block if
6299 the first thing is a variable or header, so that we can use its block when
6300 it is dynamic (useful for very large $message_headers, for example).
6302 6. Lookups now cache *every* query, not just the most recent. A new, separate
6303 store pool is used for this. It can be recovered when all lookup caches are
6304 flushed. Lookups now release memory at the end of their result strings.
6305 This has involved some general refactoring of the lookup sources.
6307 7. Some code has been added to the store_xxx() functions to reduce the amount
6308 of flapping under certain conditions.
6310 8. log_incoming_interface used to affect only the <= reception log lines. Now
6311 it causes the local interface and port to be added to several more SMTP log
6312 lines, for example "SMTP connection from", and rejection lines.
6314 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
6316 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
6318 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
6319 could be overwritten at the end of the current message (or the start of a
6320 new message if it was set in a HELO ACL). The value is now preserved for
6321 the duration of the SMTP connection.
6323 12. If a transport had a headers_rewrite setting, and a matching header line
6324 contained an unqualified address, that address was qualified, even if it
6325 did not match any rewriting rules. The underlying bug was that the values
6326 of the flags that permit the existence of unqualified sender and recipient
6327 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
6328 non-local messages, and by -bnq for local messages) were not being
6329 preserved with the message after it was received.
6331 13. When Exim was logging an SMTP synchronization error, it could sometimes log
6332 "next input=" as part of the text comprising the host identity instead of
6333 the correct text. The code was using the same buffer for two different
6334 strings. However, depending on which order the printing function evaluated
6335 its arguments, the bug did not always show up. Under Linux, for example, my
6336 test suite worked just fine.
6338 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
6339 doesn't work with some older versions of Perl. It has been changed to "my",
6340 which in any case is probably the better facility to use.
6342 15. A really picky compiler found some instances of statements for creating
6343 error messages that either had too many or two few arguments for the format
6346 16. The size of the buffer for calls to the DNS resolver has been increased
6347 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
6348 for addresses that have a lot of PTR records. This alleviates a problem; it
6349 does not fully solve it.
6351 17. A dnsdb lookup for PTR records that receives more data than will fit in the
6352 buffer now truncates the list and logs the incident, which is the same
6353 action as happens when Exim is looking up a host name and its aliases.
6354 Previously in this situation something unpredictable would happen;
6355 sometimes it was "internal error: store_reset failed".
6357 18. If a server dropped the connection unexpectedly when an Exim client was
6358 using GnuTLS and trying to read a response, the client delivery process
6359 crashed while trying to generate an error log message.
6361 19. If a "warn" verb in an ACL added multiple headers to a message in a single
6362 string, for example:
6364 warn message = H1: something\nH2: something
6366 the text was added as a single header line from Exim's point of view
6367 though it ended up OK in the delivered message. However, searching for the
6368 second and subsequent header lines using $h_h2: did not work. This has been
6369 fixed. Similarly, if a system filter added multiple headers in this way,
6370 the routers could not see them.
6372 20. Expanded the error message when iplsearch is called with an invalid key to
6373 suggest using net-iplsearch in a host list.
6375 21. When running tests using -bh, any delays imposed by "delay" modifiers in
6376 ACLs are no longer actually imposed (and a message to that effect is
6379 22. If a "gecos" field in a passwd entry contained escaped characters, in
6380 particular, if it contained a \" sequence, Exim got it wrong when building
6381 a From: or a Sender: header from that name. A second bug also caused
6382 incorrect handling when an unquoted " was present following a character
6383 that needed quoting.
6385 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
6386 was not being matched caselessly.
6388 24. Arranged for all hyphens in the exim.8 source to be escaped with
6391 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
6392 properly. Recipient callout cache records were still being keyed to include
6393 the sender, even when use_sender was set false. This led to far more
6394 callouts that were necessary. The sender is no longer included in the key
6395 when use_sender is false.
6397 26. Added "control = submission" modifier to ACLs.
6399 27. Added the ${base62d: operator to decode base 62 numbers.
6401 28. dnsdb lookups can now access SRV records.
6403 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
6404 the configuration file.
6406 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
6407 (-v) mode. This makes the output for a verbose queue run more intelligible.
6409 31. Added a use_postmaster feature to recipient callouts.
6411 32. Added the $body_zerocount variable, containing the number of binary zero
6412 bytes in the message body.
6414 33. The time of last modification of the "new" subdirectory is now used as the
6415 "mailbox time last read" when there is a quota error for a maildir
6418 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
6420 35. Added +ignore_unknown as a special item in host lists.
6422 36. Code for decoding IPv6 addresses in host lists is now included, even if
6423 IPv6 support is not being compiled. This fixes a bug in which an IPv6
6424 address was recognized as an IP address, but was then not correctly decoded
6425 into binary, causing unexpected and incorrect effects when compared with
6432 1. Very minor rewording of debugging text in manualroute to say "list of
6433 hosts" instead of "hostlist".
6435 2. If verify=header_syntax was set, and a header line with an unqualified
6436 address (no domain) and a large number of spaces between the end of the
6437 name and the colon was received, the reception process suffered a buffer
6438 overflow, and (when I tested it) crashed. This was caused by some obsolete
6439 code that should have been removed. The fix is to remove it!
6441 3. When running in the test harness, delay a bit after writing a bounce
6442 message to get a bit more predictability in the log output.
6444 4. Added a call to search_tidyup() just before forking a reception process. In
6445 theory, someone could use a lookup in the expansion of smtp_accept_max_
6446 per_host which, without the tidyup, could leave open a database connection.
6448 5. Added the variables $recipient_data and $sender_data which get set from a
6449 lookup success in an ACL "recipients" or "senders" condition, or a router
6450 "senders" option, similar to $domain_data and $local_part_data.
6452 6. Moved the writing of debug_print from before to after the "senders" test
6455 7. Change 4.31/66 (moving the time when the Received: is generated) caused
6456 problems for message scanning, either using a data ACL, or using
6457 local_scan() because the Received: header was not generated till after they
6458 were called (in order to set the time as the time of reception completion).
6459 I have revised the way this works. The header is now generated after the
6460 body is received, but before the ACL or local_scan() are called. After they
6461 are run, the timestamp in the header is updated.
6467 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
6468 before starting a queue runner without re-exec. This happened only when
6469 deliver_drop_privilege was set or when the Exim user was set to root. The
6470 effect of the bug was that timeouts during subsequent deliveries caused
6471 crashes instead of being properly handled. The handler is now left at its
6472 default (and expected) setting.
6474 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
6475 message, again when deliver_drop_privilege is set or Exim is run as root.
6476 The bug described in (1) was not present in this case, but the tidying up
6477 of the other signals was missing. I have made the two cases consistent.
6479 3. The ignore_target_hosts setting on a manualroute router was being ignored
6480 for hosts that were looked up using the /MX notation.
6482 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
6485 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
6486 operated on the sender address. After changing the $sender_address to <>
6487 for the sender address verify, Exim was re-instated it as the original
6488 (before rewriting) address, but remembering that it had rewritten it, so it
6489 wasn't rewriting it again. This bug also had the effect of breaking the
6490 sender address verification caching when the sender address was rewritten.
6492 6. The ignore_target_hosts option was being ignored by the ipliteral router.
6493 This has been changed so that if the ip literal address matches
6494 ignore_target_hosts, the router declines.
6496 7. Added expansion conditions match_domain, match_address, and match_local_
6497 part (NOT match_host).
6499 8. The placeholder for the Received: header didn't have a length field set.
6501 9. Added code to Exim itself and to exim_lock to test for a specific race
6502 condition that could lead to file corruption when using MBX delivery. The
6503 issue is with the lockfile that is created in /tmp. If this file is removed
6504 after a process has opened it but before that process has acquired a lock,
6505 there is the potential for a second process to recreate the file and also
6506 acquire a lock. This could lead to two Exim processes writing to the file
6507 at the same time. The added code performs the same test as UW imapd; it
6508 checks after acquiring the lock that its file descriptor still refers to
6509 the same named file.
6511 10. The buffer for building added header lines was of fixed size, 8192 bytes.
6512 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
6515 11. Added the smtp_active_hostname option. If used, this will typically be made
6516 to depend on the incoming interface address. Because $interface_address is
6517 not set up until the daemon has forked a reception process, error responses
6518 that can happen earlier (such as "too many connections") no longer contain
6521 12. If an expansion in a condition on a "warn" statement fails because a lookup
6522 defers, the "warn" statement is abandoned, and the next ACL statement is
6523 processed. Previously this caused the whole ACL to be aborted.
6525 13. Added the iplsearch lookup type.
6527 14. Added ident_timeout as a log selector.
6529 15. Added tls_certificate_verified as a log selector.
6531 16. Added a global option tls_require_ciphers (compare the smtp transport
6532 option of the same name). This controls incoming TLS connections.
6534 17. I finally figured out how to make tls_require_ciphers do a similar thing
6535 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
6536 before starting the TLS session.
6538 18. Tabs are now shown as \t in -bP output.
6540 19. If the log selector return_path_on_delivery was set, Exim crashed when
6541 bouncing a message because it had too many Received: header lines.
6543 20. If two routers both had headers_remove settings, and the first one included
6544 a superfluous trailing colon, the final name in the first list and the
6545 first name in the second list were incorrectly joined into one item (with a
6546 colon in the middle).
6552 1. Added -C and -D options to the exinext utility, mainly to make it easier
6553 to include in the automated testing, but these could be helpful when
6554 multiple configurations are in use.
6556 2. The exinext utility was not formatting the output nicely when there was
6557 an alternate port involved in the retry record key, nor when there was a
6558 message id as well (for retries that were specific to a specific message
6559 and a specific host). It was also confused by IPv6 addresses, because of
6560 the additional colons they contain. I have fixed the IPv4 problem, and
6561 patched it up to do a reasonable job for IPv6.
6563 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
6564 delivery, the log line now contains "pipelined" if PIPELINING was used.
6566 4. An SMTP transport process used to panic and die if the bind() call to set
6567 an explicit outgoing interface failed. This has been changed; it is now
6568 treated in the same way as a connect() failure.
6570 5. A reference to $sender_host_name in the part of a conditional expansion
6571 that was being skipped was still causing a DNS lookup. This no longer
6574 6. The def: expansion condition was not recognizing references to header lines
6575 that used bh_ and bheader_.
6577 7. Added the _cache feature to named lists.
6579 8. The code for checking quota_filecount in the appendfile transport was
6580 allowing one more file than it should have been.
6582 9. For compatibility with Sendmail, the command line option
6590 and sets the incoming protocol and host name (for trusted callers). The
6591 host name and its colon can be omitted when only the protocol is to be set.
6592 Note the Exim already has two private options, -pd and -ps, that refer to
6593 embedded Perl. It is therefore impossible to set a protocol value of "d" or
6594 "s", but I don't think that's a major issue.
6596 10. A number of refactoring changes to the code, none of which should affect
6599 (a) The number of logging options was getting close to filling up the
6600 32-bit word that was used as a bit map. I have split them into two classes:
6601 those that are passed in the argument to log_write(), and those that are
6602 only ever tested independently outside of that function. These are now in
6603 separate 32-bit words, so there is plenty of room for expansion again.
6604 There is no change in the user interface or the logging behaviour.
6606 (b) When building, for example, log lines, the code previously used a
6607 macro that called string_cat() twice, in order to add two strings. This is
6608 not really sufficiently general. Furthermore, there was one instance where
6609 it was actually wrong because one of the argument was used twice, and in
6610 one call a function was used. (As it happened, calling the function twice
6611 did not affect the overall behaviour.) The macro has been replaced by a
6612 function that can join an arbitrary number of extra strings onto a growing
6615 (c) The code for expansion conditions now uses a table and a binary chop
6616 instead of a serial search (which was left over from when there were very
6617 few conditions). Also, it now recognizes conditions like "pam" even when
6618 the relevant support is not compiled in: a suitably worded error message is
6619 given if an attempt is made to use such a condition.
6621 11. Added ${time_interval:xxxxx}.
6623 12. A bug was causing one of the ddress fields not to be passed back correctly
6624 from remote delivery subprocesses. The field in question was not being
6625 subsequently used, so this caused to problems in practice.
6627 13. Added new log selectors queue_time and deliver_time.
6629 14. Might have fixed a bug in maildirsizefile handling that threw up
6630 "unexpected character" debug warnings, and recalculated the data
6631 unnecessarily. In any case, I expanded the warning message to give more
6634 15. Added the message "Restricted characters in address" to the statements in
6635 the default ACL that block characters like @ and % in local parts.
6637 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
6638 Three changes have been made:
6640 (a) There was a serious bug; a negative response to MAIL caused the whole
6641 recipient domain to be cached as invalid, thereby blocking all messages
6642 to all local parts at the same domain, from all senders. This bug has
6643 been fixed. The domain is no longer cached after a negative response to
6644 MAIL if the sender used is not empty.
6646 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
6649 (c) A new callout option, "use_sender" has been added for people who want
6650 the modified behaviour.
6656 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
6659 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
6660 indeed breaks things for older releases.
6662 3. Added additional logging to the case where there is a problem reading data
6663 from a filter that is running in a subprocess using a pipe, in order to
6664 try to track down a specific problem.
6666 4. Testing facility fudge: when running in the test harness and attempting
6667 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
6668 getting "No route to host". Convert this to a timeout.
6670 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
6673 6. Some OS don't have socklen_t but use size_t instead. This affects the
6674 fifth argument of getsockopt() amongst other things. This is now
6675 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
6676 can be set for individual OS. I have set it for SunOS5, OSF1, and
6677 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
6678 some earlier ones do not.
6680 7. Change 4.30/15 was not doing the test caselessly.
6682 8. The standard form for an IPv6 address literal was being rejected by address
6683 parsing in, for example, MAIL and RCPT commands. An example of this kind of
6684 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6685 this, as well as the form without the "IPv6" on the front (but only when
6686 address literals are enabled, of course).
6688 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
6690 10. Exim crashed if a message with an empty sender address specified by -f
6691 encountered a router with an errors_to setting. This could be provoked only
6692 by a command such as
6696 where an empty string was supplied; "<>" did not hit this bug.
6698 11. Installed PCRE release 4.5.
6700 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
6701 remained set. It is now erased.
6703 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
6704 times from message ids (which are base 36 rather than the normal 62).
6706 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
6707 were being counted as actual protocol errors, and logged if the log
6708 selector +smtp_protocol_error was set. One cannot be perfect in this test,
6709 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
6710 and DATA following a set of rejected RCPTs do not count as protocol errors.
6711 In other words, Exim assumes they were pipelined, though this may not
6712 actually be the case. Of course, in all cases the client gets an
6713 appropriate error code.
6715 15. If a lookup fails in an ACL condition, a message about the failure may
6716 be available; it is used if testing the ACL cannot continue, because most
6717 such messages specify what the cause of the deferral is. However, some
6718 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
6719 that caused an old message to be retained and used if a later statement
6720 caused a defer, replacing the real cause of the deferral.
6722 16. If an IP address had so many PTR records that the DNS lookup buffer
6723 was not large enough to hold them, Exim could crash while trying to process
6724 the truncated data. It now detects and logs this case.
6726 17. Further to 4.21/58, another change has been made: if (and only if) the
6727 first line of a message (the first header line) ends with CRLF, a bare LF
6728 in a subsequent header line has a space inserted after it, so as not to
6729 terminate the header.
6731 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
6732 unnecessarily, used atoi() instead of strtol(), and didn't check the
6733 termination when getting file sizes from file names by regex.
6735 19. Completely re-implemented the support for maildirsize files, in the light
6736 of a number of problems with the previous contributed implementation
6737 (4.30/29). In particular:
6739 . If the quota is zero, the maildirsize file is maintained, but no quota is
6742 . If the maildir directory does not exist, it is created before any attempt
6743 to write a maildirsize file.
6745 . The quota value in the file is just a cache; if the quota is changed in
6746 the transport, the new value overrides.
6748 . A regular expression is available for excluding directories from the
6751 20. The autoreply transport checks the characters in options that define the
6752 message's headers; it allows continued headers, but it was checking with
6753 isspace() after an embedded newline instead of explicitly looking for a
6756 21. If all the "regular" hosts to which an address was routed had passed their
6757 expiry times, and had not reached their retry times, the address was
6758 bounced, even if fallback hosts were defined. Now Exim should go on to try
6761 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
6762 equivalent code in the SMTP transport. Some hosts send humungous responses
6763 to HELO/EHLO, more than 1024 it seems.
6765 23. Refactoring: code in filter.c used (void *) for "any old type" but this
6766 gives compiler warnings in some environments. I've now done it "properly",
6769 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
6770 (because of problems with the built-in one) was declared to return uschar *
6771 instead of char *, causing compiler failure.
6773 25. Fixed a file descriptor leak when processing alias/forward files.
6775 26. Fixed a minor format string issue in dbfn.c.
6777 27. Typo in exim.c: ("dmbnz" for "dbmnz").
6779 28. If a filter file refered to $h_xxx or $message_headers, and the headers
6780 contained RFC 2047 "words", Exim's memory could, under certain conditions,
6783 29. When a sender address is verified, it is cached, to save repeating the test
6784 when there is more than one recipient in a message. However, when the
6785 verification involves a callout, it is possible for different callout
6786 options to be set for different recipients. It is too complicated to keep
6787 track of this in the cache, so now Exim always runs a verification when a
6788 callout is required, relying on the callout cache for the optimization.
6789 The overhead is duplication of the address routing, but this should not be
6792 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
6793 to be verified with callout=postmaster, and the main callout worked but the
6794 postmaster check failed, the verification correctly failed. However, if a
6795 subsequent RCPT command asked for sender verification *without* the
6796 postmaster check, incorrect caching caused this verification also to fail,
6799 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
6800 it was not caching the DNS options (qualify_single, search_parents) that
6801 were used when the lookup failed. A subsequent lookup with different
6802 options therefore always gave the same answer, though there were cases
6803 where it should not have. (Example: a "domains = !$mx_any" option on a
6804 dnslookup router: the "domains" option is always processed without any
6805 widening, but the router might have qualify_single set.) Now Exim uses the
6806 cached value only when the same options are set.
6808 32. Added John Jetmore's "exipick" utility to the distribution.
6810 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
6811 than a timeout (e.g. a certificate is required, and is not provided), an
6812 Exim server now closes the connection immediately. Previously it waited for
6813 the client to close - but if the client is SSL, it seems that they each
6814 wait for each other, leading to a delay before one of them times out.
6816 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
6817 maintained 0.8.x compatibility because I don't think many are using it, and
6818 it is clearly obsolete.
6820 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
6823 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
6824 client certificate was expired. A simple patch fixes this, though I don't
6825 understand the full logic of why the verify callback is called multiple
6828 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
6829 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
6830 which causes problems with some clients (such as the Certicom SSL Plus
6831 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
6832 disables the coutermeasure allowing Eudora to connect."
6834 38. Exim was not checking that a write() to a log file succeeded. This could
6835 lead to Bad Things if a log got too big, in particular if it hit a file
6836 size limit. Exim now panics and dies if it cannot write to a log file, just
6837 as it does if it cannot open a log file.
6839 39. Modified OS/Makefile-Linux so that it now contains
6841 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
6843 The two -D definitions ensure that Exim is compiled with large file
6844 support, which makes it possible to handle log files that are bigger than
6847 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
6848 instance) a domain was checked against a named list that involved a lookup,
6849 causing $domain_data to be set, then another domain was checked against the
6850 same list, then the first domain was re-checked, the value of $domain_data
6851 after the final check could be wrong. In particular, if the second check
6852 failed, it could be set empty. This bug probably also applied to
6855 41. The strip_trailing_dot option was not being applied to the address given
6856 with the -f command-line option.
6858 42. The code for reading a message's header from the spool was incrementing
6859 $received_count, but never initializing it. This meant that the value was
6860 incorrect (doubled) while delivering a message in the same process in which
6861 it was received. In the most common configuration of Exim, this never
6862 happens - a fresh exec is done - but it can happen when
6863 deliver_drop_privilege is set.
6865 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6866 it now includes up to 150 characters of the unexpected data in the log
6869 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6870 and building data strings. The size of both of these buffers was 10 000
6871 bytes - far larger than anybody would *ever* want, thought I. Needless to
6872 say, somebody hit the limit. I have increased the maximum line length to
6873 20 000 and the maximum data length of concatenated lines to 100 000. I have
6874 also fixed two bugs, because there was no checking on these buffers. Tsk,
6875 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6876 buffer is too small.
6878 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6879 lsearch lookups. Now it does.
6881 46. When parsing a route_list item in a manualroute router, a fixed-length
6882 buffer was used for the list of hosts. I made this 1024 bytes long,
6883 thinking that nobody would ever have a list of hosts that long. Wrong.
6884 Somebody had a whole pile of complicated expansion conditions, and the
6885 string was silently truncated, leading to an expansion error. It turns out
6886 that it is easier to change to an unlimited length (owing to other changes
6887 that have happened since this code was originally written) than to build
6888 structure for giving a limitation error. The length of the item that
6889 expands into the list of hosts is now unlimited.
6891 47. The lsearch lookup could not handle data where the length of text line was
6892 more than 4095 characters. Such lines were truncated, leading to shortened
6893 data being returned. It should now handle lines of any length.
6895 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6896 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6899 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6900 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6901 no longer happens in comment lines. A list of replacements is now placed
6902 at the head of all of the source files, except those whose only change is
6903 to replace PERL_COMMAND in the very first #! line.
6905 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6906 messages on the queue, with a bottom-up merge sort, using code contributed
6907 by Michael Haardt. This should make operations like -bp somewhat faster on
6908 large queues. It won't affect queue runners, except when queue_run_in_order
6911 51. Installed eximstats 1.31 in the distribution.
6913 52. Added support for SRV lookups to the dnslookup router.
6915 53. If an ACL referred to $message_body or $message_body_end, the value was not
6916 reset for any messages that followed in the same SMTP session.
6918 54. The store-handling optimization for building very long strings was not
6919 differentiating between the different store pools. I don't think this
6920 actually made any difference in practice, but I've tidied it.
6922 55. While running the routers to verify a sender address, $sender_address
6923 was still set to the sender address. This is wrong, because when routing to
6924 send a bounce to the sender, it would be empty. Therefore, I have changed
6925 it so that, while verifying a sender address, $sender_address is set to <>.
6926 (There is no change to what happens when verifying a recipient address.)
6928 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6929 target A or AAAA records (if not already returned) without resetting the
6930 qualify_single or search_parents options of the DNS resolver. These are
6931 inappropriate in this case because the targets of MX and SRV records must
6932 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6933 target that, when qualified, matched something in the local domain. These
6934 two options are now turned off when doing these lookups.
6936 57. It seems that at least some releases of Reiserfs (which does not have the
6937 concept of a fixed number of inodes) returns zero and not -1 for the
6938 number of available inodes. This interacted badly with check_spool_inodes,
6939 which assumed that -1 was the "no such thing" setting. What I have done is
6940 to check that the total number of inodes is greater than zero before doing
6941 the test of how many are available.
6943 58. When a "warn" ACL statement has a log_message modifier, the message is
6944 remembered, and not repeated. This is to avoid a lot of repetition when a
6945 message has many recipients that cause the same warning to be written.
6946 However, Exim was preserving the list of already written lines for an
6947 entire SMTP session, which doesn't seem right. The memory is now reset if a
6948 new message is started.
6950 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6951 debugging output unless log_rewrite was also set.
6953 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6954 of (int)(handle) when we know that handle contains (void *)(-1).
6956 61. The Exim daemon panic-logs an error return when it closes the incoming
6957 connection. However "connection reset by peer" seems to be common, and
6958 isn't really an error worthy of noting specially, so that particular error
6961 62. When Exim is trying to find all the local interfaces, it used to panic and
6962 die if the ioctl to get the interface flags failed. However, it seems that
6963 on at least one OS (Solaris 9) it is possible to have an interface that is
6964 included in the list of interfaces, but for which you get a failure error
6965 for this call. This happens when the interface is not "plumbed" into a
6966 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6967 failure of the "get flags" call assumes that the interface is down.
6969 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6970 makes life easier for people who are doing arithmetic on fields extracted
6971 from dates, where you often get leading zeros that should not be
6972 interpreted as octal.
6974 64. Added qualify_domain to the redirect router, to override the global
6977 65. If a pathologically long header line contained very many addresses (the
6978 report of this problem mentioned 10 000) and each of them was rewritten,
6979 Exim could use up a very large amount of memory. (It kept on making new
6980 copies of the header line as it rewrote, and never released the old ones.)
6981 At the expense of a bit more processing, the header rewriting function has
6982 been changed so that it no longer eats memory in this way.
6984 66. The generation of the Received: header has been moved from the time that a
6985 message starts to be received, to the time that it finishes. The timestamp
6986 in the Received: header should now be very close to that of the <= log
6987 line. There are two side-effects of this change:
6989 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6990 logged header lines no longer include the local Received: line, because
6991 it has not yet been created. The same applies to a copy of the message
6992 that is returned to a non-SMTP sender when a message is rejected.
6994 (b) When a filter file is tested using -bf, no additional Received: header
6995 is added to the test message. After some thought, I decided that this
6998 This change does not affect the value of $received_for. It is still set
6999 after address rewriting, but before local_scan() is called.
7001 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
7003 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
7004 gave an unhelpful panic error message, and a defer error. I have managed to
7005 change this behaviour so that it now rejects any supplied certificate,
7006 which seems right, as the list of acceptable certificates is empty.
7008 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
7009 gave an unhelpful defer error. I have not managed to make this reject any
7010 supplied certificates, but the error message it gives is "no certificate
7011 supplied", which is not helpful.
7013 70. exigrep's output now also includes lines that are not associated with any
7014 message, but which match the given pattern. Implemented by a patch from
7015 Martin Sluka, which also tidied up the Perl a bit.
7017 71. Recipient callout verification, like sender verification, was using <> in
7018 the MAIL FROM command. This isn't really the right thing, since the actual
7019 sender may affect whether the remote host accepts the recipient or not. I
7020 have changed it to use the actual sender in the callout; this means that
7021 the cache record is now keyed on a recipient/sender pair, not just the
7022 recipient address. There doesn't seem to be a real danger of callout loops,
7023 since a callout by the remote host to check the sender would use <>.
7024 [SEE ABOVE: changed after hitting problems.]
7026 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
7027 temporary errors. However, in the case of such a code being given after
7028 the end of a data transmission (i.e. after ".") Exim was failing to write
7029 a retry record for the message. (Yes, there was some broken host that was
7030 actually sending 8xx at this point.)
7032 73. An unknown lookup type in a host list could cause Exim to panic-die when
7033 the list was checked. (An example that provoked this was putting <; in the
7034 middle of a list instead of at the start.) If this happened during a DATA
7035 ACL check, a -D file could be left lying around. This kind of configuration
7036 error no longer causes Exim to die; instead it causes a defer error. The
7037 incident is still logged to the main and panic logs.
7039 74. Buglet left over from Exim 3 conversion. The message "too many messages
7040 in one connection" was written to the rejectlog but not the mainlog, except
7041 when address rewriting (yes!) was being logged.
7043 75. Added write_rejectlog option.
7045 76. When a system filter was run not as root (that is, when system_filter_user
7046 was set), the values of the $n variables were not being returned to the
7047 main process; thus, they were not subsequently available in the $sn
7050 77. Added +return_path_on_delivery log selector.
7052 78. A connection timeout was being treated differently from recipients deferred
7053 when testing hosts_max_try with a message that was older than the host's
7054 retry timeout. (The host should not be counted, thus allowing all hosts to
7055 be tried at least once before bouncing.) This may have been the cause of an
7056 occasionally reported bug whereby a message would remain on the queue
7057 longer than the retry timeout, but would be bounced if a delivery was
7058 forced. I say "may" because I never totally pinned down the problem;
7059 setting up timeout/retry tests is difficult. See also the next item.
7061 79. The ultimate address timeout was not being applied to errors that involved
7062 a combination of host plus message (for example, a timeout on a MAIL
7063 command). When an address resolved to a number of possible hosts, and they
7064 were not all tried for each delivery (e.g. because of hosts_max_try), a
7065 message could remain on the queue longer than the retry timeout.
7067 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
7070 81. Fixed an obscure SMTP outgoing bug which required at least the following
7071 conditions: (a) there was another message waiting for the same server;
7072 (b) the server returned 5xx to all RCPT commands in the first message so
7073 that the message was not completed; (c) the server dropped the connection
7074 or gave a negative response to the RSET that Exim sends to abort the
7075 transaction. The observed case was a dropped connection after DATA that had
7076 been sent in pipelining mode. That is, the server had advertised PIPELINING
7077 but was not implementing it correctly. The effect of the bug was incorrect
7078 behaviour, such as trying another host, and this could lead to a crash.
7084 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
7085 and daemon.c were passed as pointers to ints; they should have been
7086 pointers to socklen_t variables (which are typically unsigned ints).
7088 2. Some signed/unsigned type warnings in the os.c file for Linux have been
7091 3. Fixed a really odd bug that affected only the testing scheme; patching a
7092 certain fixed string in the binary changed the value of another string that
7093 happened to be identical to the end of the original first string.
7095 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
7096 name", it returns that address as the IP address. On some operating
7097 systems (e.g. Solaris), it also passes back the IP address string as the
7098 "host name". However, on others (e.g. Linux), it passes back an empty
7099 string. Exim wasn't checking for this, and was changing the host name to an
7100 empty string, assuming it had been canonicalized.
7102 5. Although rare, it is permitted to have more than one PTR record for a given
7103 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
7104 all the names associated with an address, because they do in Solaris.
7105 However, it seems that they do not in Linux for data that comes from the
7106 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
7107 I found this out when I moved to a new Linux workstation and tried to run
7108 the Exim test suite.
7110 To get round this problem I have changed the code so that it now does its
7111 own call to the DNS to look up PTR records when searching for a host name.
7112 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
7113 addresses that are only in /etc/hosts are still found.
7115 This behaviour is, however, controlled by an option called host_lookup_
7116 order, which defaults to "bydns:byaddr". If people want to use the other
7117 order, or indeed, just use one or the other means of lookup, they can
7118 specify it in this variable.
7120 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
7121 some operating systems, this comes back from gethostbyaddr() as an empty
7122 string, and this is what Exim used to test for. However, it seems that in
7123 other systems, "." is yielded. Exim now tests for this case too.
7125 7. The values of check_spool_space and check_log_space are now held internally
7126 as a number of kilobytes instead of an absolute number of bytes. If a
7127 numbers is specified without 'K' or 'M', it is rounded up to the nearest
7128 kilobyte. This means that much larger values can be stored.
7130 8. Exim monitor: an attempt to get the action menu when not actually pointing
7131 at a message produces an empty menu entitled "No message selected". This
7132 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
7133 no entries in it ("Shell widget menu has zero width and/or height"). So I
7134 have added a single, blank menu entry in this case.
7136 9. Added ${quote_local_part.
7138 10. MIME decoding is now applied to the contents of Subject: header lines when
7141 11. Now that a reference to $sender_host_address automatically causes a reverse
7142 lookup to occur if necessary (4.13/18), there is no need to arrange for a
7143 host lookup before query-style lookups in lists that might use this
7144 variable. This has therefore been abolished, and the "net-" prefix is no
7145 longer necessary for query-style lookups.
7147 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
7148 have been a typo for LFLAGS, so it has been changed.
7150 13. The install script calls Exim with "-C /dev/null" in order to find the
7151 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
7152 to be output. However, since Exim outputs its version number before the
7153 error, it didn't break the script. It just looked ugly. I fixed this by
7154 always allowing "-C /dev/null" if the caller is root.
7156 14. Ignore overlarge ACL variable number when reading spool file - insurance
7157 against a later release with more variables having written the file.
7159 15. The standard form for an IPv6 address literal was being rejected by EHLO.
7160 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
7161 this, as well as the form without the "IPv6" on the front.
7163 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
7164 OS/Makefile-Darwin file.
7166 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
7167 only to LDAP libraries that do not have LDAP_OPT_DEREF.
7169 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
7170 variable. However, some OS use ints for inodes. I've added cast to long int
7171 to get rid of the compiler warning.
7173 19. I had forgotten to lock out "/../" in configuration file names when
7174 ALT_CONFIG_PREFIX was set.
7176 20. Routers used for verification do not need to specify transports. However,
7177 if such a router generated a host list, and callout was configured, Exim
7178 crashed, because it could not find a port number from the (non-existent)
7179 transport. It now assumes port 25 in this circumstance.
7181 21. Added the -t option to exigrep.
7183 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
7184 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
7185 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
7186 them is set, LOOKUP_LSEARCH is forced.
7188 23. "exim -bV" now outputs a list of lookups that are included in the binary.
7190 24. Added sender and host information to the "rejected by local_scan()" log
7191 line; previously there was no indication of these.
7193 25. Added .include_if_exists.
7195 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
7196 Linux. It turns out that not all file systems support this. Apparently some
7197 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
7198 do it.) To cope with this, the error EINVAL, which means that sync-ing is
7199 not supported on the file descriptor, is now ignored when Exim is trying to
7200 sync a directory. This applies only to Linux.
7202 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
7204 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
7205 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
7208 29. Added support for maildirsize files from supplied patch (modified a bit).
7210 30. The use of :fail: followed by an empty string could lead Exim to respond to
7211 sender verification failures with (e.g.):
7213 550 Verification failed for <xxx>
7214 550 Sender verify failed
7216 where the first response line was missing the '-' that indicates it is not
7217 the final line of the response.
7219 31. The loop for finding the name of the user that called Exim had a hardwired
7220 limit of 10; it now uses the value of finduser_retries, which is used for
7221 all other user lookups.
7223 32. Added $received_count variable, available in data and not_smtp ACLs, and at
7226 33. Exim was neglecting to zero errno before one call of strtol() when
7227 expanding a string and expecting an integer value. On some systems this
7228 resulted in spurious "integer overflow" errors. Also, it was casting the
7229 result into an int without checking.
7231 34. Testing for a connection timeout using "timeout_connect" in the retry rules
7232 did not work. The code looks as if it has *never* worked, though it appears
7233 to have been documented since at least release 1.62. I have made it work.
7235 35. The "timeout_DNS" error in retry rules, also documented since at least
7236 1.62, also never worked. As it isn't clear exactly what this means, and
7237 clearly it isn't a major issue, I have abolished the feature by treating it
7238 as "timeout", and writing a warning to the main and panic logs.
7240 36. The display of retry rules for -brt wasn't always showing the error code
7243 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
7244 timeout_connect_A, timeout_connect_MX.
7246 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
7247 to the empty sender.
7249 39. The daemon was not analysing the content of -oX till after it had closed
7250 stderr and disconnected from the controlling terminal. This meant that any
7251 syntax errors were only noted on the panic log, and the return code from
7252 the command was 0. By re-arranging the code a little, I've made the
7253 decoding happen first, so such errors now appear on stderr, and the return
7254 code is 1. However, the actual setting up of the sockets still happens in
7255 the disconnected process, so errors there are still only recorded on the
7258 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
7259 connections (as happens on some IP stacks) was logged at start up time as
7260 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
7261 it from "IPv6 and IPv4", which means that two separate sockets are being
7264 41. The debug output for gethostbyname2() or getipnodebyname() failures now
7265 says whether AF_INET or AF_INET6 was passed as an argument.
7267 42. Exiwhat output was messed up when time zones were included in log
7270 43. Exiwhat now gives more information about the daemon's listening ports,
7271 and whether -tls-on-connect was used.
7273 44. The "port" option of the smtp transport is now expanded.
7275 45. A "message" modifier in a "warn" statement in a non-message ACL was being
7276 silently ignored. Now an error message is written to the main and panic
7279 46. There's a new ACL modifier called "logwrite" which writes to a log file
7280 as soon as it is encountered.
7282 47. Added $local_user_uid and $local_user_gid at routing time.
7284 48. Exim crashed when trying to verify a sender address that was being
7287 49. Exim was recognizing only a space character after ".include". It now also
7288 recognizes a tab character.
7290 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
7291 extracting the relevant information from the specification. The man page no
7292 longer contains scrambled data for the -d option, and I've added a section
7293 at the front about calling Exim under different names.
7295 51. Added "extra_headers" argument to the "mail" command in filter files.
7297 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
7300 53. Installed eximstats 1.29.
7302 54. Added transport_filter_timeout as a generic transport option.
7304 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
7305 Cc: header lines. This was required by RFC 822, but it not required by RFC
7308 56. Exim used to add From:, Date:, and Message-Id: header lines to any
7309 incoming messages that did not have them. Now it does so only if the
7310 message originates locally, that is, if there is no associated remote host
7311 address. When Resent- header lines are present, this applies to the Resent-
7312 lines rather than the non-Resent- lines.
7314 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
7315 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
7317 58. Messages for configuration errors now include the name of the main
7318 configuration file - useful now that there may be more than one file in a
7319 list (.included file names were always shown).
7321 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
7322 for those rare installations that do not start the daemon as root or run it
7323 setuid root. I've cut out the call to initgroups() if the daemon is not
7326 60. The Exim user and group can now be bound into the binary as text strings
7327 that are looked up at the start of Exim's processing.
7329 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
7331 62. Added $mailstore_basename variable.
7333 63. Installed patch to sieve.c from Michael Haardt.
7335 64. When Exim failed to open the panic log after failing to open the main log,
7336 the original message it was trying to log was written to stderr and debug
7337 output, but if they were not available (the usual case in production), it
7338 was lost. Now it is written to syslog before the two lines that record the
7339 failures to open the logs.
7341 65. Users' Exim filters run in subprocesses under the user's uid. It is
7342 possible for a "deliver" command or an alias in a "personal" command to
7343 provoke an address rewrite. If logging of address rewriting is configured,
7344 this fails because the process is not running as root or exim. There may be
7345 a better way of dealing with this, but for the moment (because 4.30 needs
7346 to be released), I have disabled address rewrite logging when running a
7347 filter in a non-root, non-exim process.
7353 1. The buildconfig auxiliary program wasn't quoting the value set for
7354 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
7355 not defined. This bug was masked in 4.22 by the effect that was fixed in
7358 2. Some messages that were rejected after a message id was allocated were
7359 shown as "incomplete" by exigrep. It no longer does this for messages that
7360 are rejected by local_scan() or the DATA or non-SMTP ACLs.
7362 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
7363 have allow_domain_literals set, the ID did not get logged in the <= line.
7364 Domain literals are now always recognized in Message-ID: header lines.
7366 4. The first argument for a ${extract expansion item is the key name or field
7367 number. Leading and trailing spaces in this item were not being ignored,
7368 causing some misleading effects.
7370 5. When deliver_drop_privilege was set, single queue runner processes started
7371 manually (i.e. by the command "exim -q") or by the daemon (which uses the
7372 same command in the process it spins off) were not dropping privilege.
7374 6. When the daemon running as "exim" started a queue runner, it always
7375 re-executed Exim in the spun-off process. This is a waste of effort when
7376 deliver_drop_privilege is set. The new process now just calls the
7377 queue-runner function directly.
7383 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
7386 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
7387 ignored. Though the use of -oP was forcing the writing of a pid file, it
7388 was always written to the default place.
7390 3. If the message "no IP address found for host xxxx" is generated during
7391 incoming verification, it is now followed by identification of the incoming
7392 connection (so you can more easily find what provoked it).
7394 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
7396 5. Added some features to "harden" Exim a bit more against certain attacks:
7398 (a) There is now a build-time option called FIXED_NEVER_USERS that can
7399 be put in Local/Makefile. This is like the never_users runtime option,
7400 but it cannot be overridden. The default setting is "root".
7402 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
7403 prefix string with which any file named in a -C command line option
7406 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
7407 is retained for -C and -D only if the caller of Exim is root. Without
7408 it, the exim user may also use -C and -D and retain privilege.
7410 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
7411 command line option is disabled.
7413 6. Macro names set by the -D option must start with an upper case letter, just
7414 like macro names defined in the configuration file.
7416 7. Added "dereference=" facility to LDAP.
7418 8. Two instances of the typo "uknown" in the source files are fixed.
7420 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
7421 the Configure-Makefile script screwed up while processing it.
7423 10. Incorporated PCRE 4.4.
7425 11. The SMTP synchronization check was not operating right at the start of an
7426 SMTP session. For example, it could not catch a HELO sent before the client
7427 waited for the greeting. There is now a check for outstanding input at the
7428 point when the greeting is written. Because of the duplex, asynchronous
7429 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
7430 way, but not yet received, when the check is performed.
7432 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
7433 on TCP/IP sockets, because this apparently causes some broken clients to
7436 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
7437 unchanged) from the Cygwin maintainer.
7439 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
7440 was supported instead of testing for mbx. Effectively a typo.
7442 15. The spa authenticator server code was not checking that the input it
7443 received was valid base64.
7445 16. The debug output line for the "set" modifier in ACLs was not showing the
7446 name of the variable that was being set.
7448 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
7450 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
7451 lookup on its own. Something else was needed to trigger the lookup. For
7452 example, a match in host_lookup or the need for a host name in a host list.
7453 Now, if $sender_host_name is referenced and the host name has not yet been
7454 looked up, a lookup is performed. If the lookup fails, the variable remains
7455 empty, and $host_lookup_failed is set to "1".
7457 19. Added "eqi" as a case-independent comparison operator.
7459 20. The saslauthd authentication condition could segfault if neither service
7460 nor realm was specified.
7462 21. If an overflowing value such as "2048M" was set for message_size_limit, the
7463 error message that was logged was misleading, and incoming SMTP
7464 connections were dropped. The message is now more accurate, and temporary
7465 errors are given to SMTP connections.
7467 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
7468 (except RSET) with a 421 error, until QUIT is received. However, it was
7469 failing to send a response to QUIT.
7471 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
7472 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
7473 ACL is now run after the helo_try_verify_hosts code.
7475 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
7476 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
7477 case-independent, so other case variants are also recognized). Apparently
7478 some systems use these upper case variants.
7480 25. If more than two messages were waiting for the same host, and a transport
7481 filter was specified for the transport, Exim sent two messages over the
7482 same TCP/IP connection, and then failed with "socket operation on non-
7483 socket" when it tried to send the third.
7485 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
7487 27. The extern definition of crypt16() in expand.c was not being excluded when
7488 the OS had its own crypt16() function.
7490 28. Added bounce_return_body as a new option, and bounce_return_size_limit
7491 as a preferred synonym for return_size_limit, both as an option and as an
7494 29. Added LIBS=-liconv to OS/Makefile-OSF1.
7496 30. Changed the default configuration ACL to relax the local part checking rule
7497 for addresses that are not in any local domains. For these addresses,
7498 slashes and pipe symbols are allowed within local parts, but the sequence
7499 /../ is explicitly forbidden.
7501 31. SPA server authentication was not clearing the challenge buffer before
7504 32. log_message in a "warn" ACL statement was writing to the reject log as
7505 well as to the main log, which contradicts the documentation and doesn't
7506 seem right (because no rejection is happening). So I have stopped it.
7508 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
7509 However, I am unable to do any testing of this.
7511 34. Fixed an infelicity in the appendfile transport. When checking directories
7512 for a mailbox, to see if any needed to be created, it was accidentally
7513 using path names with one or more superfluous leading slashes; tracing
7514 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
7516 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
7517 contents are added to the log line that is written for every discarded
7518 recipient. (Previously a log_message setting was ignored.)
7520 36. The ${quote: operator now quotes the string if it is empty.
7522 37. The install script runs exim in order to find its version number. If for
7523 some reason other than non-existence or emptiness, which it checks, it
7524 could not run './exim', it was installing it with an empty version number,
7525 i.e. as "exim-". This error state is now caught, and the installation is
7528 38. An argument was missing from the function that creates an error message
7529 when Exim fails to connect to the socket for saslauthd authentication.
7530 This could cause Exim to crash, or give a corrupted message.
7532 39. Added isip, isip4, and isip6 to ${if conditions.
7534 40. The ACL variables $acl_xx are now saved with the message, and can be
7535 accessed later in routers, transports, and filters.
7537 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
7538 strings in the file are not string-expanded.
7540 42. If a MAIL command specified a SIZE value that was too large to fit into an
7541 int variable, the check against message_size_limit failed. Such values are
7542 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
7543 day this will have to be increased, but I don't think I want to be around
7544 when emails are that large.
7551 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
7552 iconv() is not standard in FreeBSD.
7554 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
7555 IPv6 enabled. The observed symptom was a segmentation fault on return from
7556 the function os_common_find_running_interfaces() in src/os.c.
7558 3. In the check_special_case() function in daemon.c I had used "errno" as an
7559 argument name, which causes warnings on some systems. This was basically a
7560 typo, since it was named "eno" in the comments!
7562 4. The code that waits for the clock to tick (at a resolution of some fraction
7563 of a second) so as to ensure message-id uniqueness was always waiting for
7564 at least one whole tick, when it could have waited for less. [This is
7565 almost certainly not relevant at current processor speeds, where it is
7566 unlikely to ever wait at all. But we try to future-proof.]
7568 5. The function that sleeps for a time interval that includes fractions of a
7569 second contained a race. It did not block SIGALRM between setting the
7570 timer, and suspending (a couple of lines later). If the interval was short
7571 and the sigsuspend() was delayed until after it had expired, the suspension
7572 never ended. On busy systems this could lead to processes getting stuck for
7575 6. Some uncommon configurations may cause a lookup to happen in a queue runner
7576 process, before it forks any delivery processes. The open lookup caching
7577 mechanism meant that the open file or database connection was passed into
7578 the delivery process. The problem was that delivery processes always tidy
7579 up cached lookup data. This could cause a problem for the next delivery
7580 process started by the queue runner, because the external queue runner
7581 process does not know about the closure. So the next delivery process
7582 still has data in the lookup cache. In the case of a file lookup, there was
7583 no problem because closing a file descriptor in a subprocess doesn't affect
7584 the parent. However, if the lookup was caching a connection to a database,
7585 the connection was closed, and the second delivery process was likely to
7586 see errors such as "PGSQL: query failed: server closed the connection
7587 unexpectedly". The problem has been fixed by closing all cached lookups
7588 in a queue runner before running a delivery process.
7590 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
7591 seem to have the "const" qualifier which it has on other OS. I've
7594 8. Change 4.21/2 was too strict. It is only if there are two authenticators
7595 *of the same type* (client or server) with the same public name that an
7596 error should be diagnosed.
7598 9. When Exim looked up a host name for an IP address, but failed to find the
7599 original IP address when looking up the host name (a safety check), it
7600 output the message "<ip address> does not match any IP for NULL", which was
7601 confusing, to say the least. The bug was that the host name should have
7602 appeared instead of "NULL".
7604 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
7605 user that is built into the binary, and the -C or -D options is used, root
7606 privilege is dropped before the configuration file is read. In addition,
7607 logging is switched to stderr instead of the normal log files. If the
7608 configuration then re-defines the Exim user, the unprivileged environment
7609 is probably not what is expected, so Exim logs a panic warning message (but
7612 However, if deliver_drop_privilege is set, the unprivileged state may well
7613 be exactly what is intended, so the warning has been cut out in that case,
7614 and Exim is allowed to try to write to its normal log files.
7620 1. smtp_return_error_details was not giving details for temporary sender
7621 or receiver verification errors.
7623 2. Diagnose a configuration error if two authenticators have the same public
7626 3. Exim used not to create the message log file for a message until the first
7627 delivery attempt. This could be confusing when incoming messages were held
7628 for policy or load reasons. The message log file is now created at the time
7629 the message is received, and an initial "Received" line is written to it.
7631 4. The automatically generated man page for command line options had a minor
7632 bug that caused no ill effects; however, a more serious problem was that
7633 the procedure for building the man page automatically didn't always
7634 operate. Consequently, release 4.20 contains an out-of-date version. This
7635 shouldn't happen again.
7637 5. When building Exim with embedded Perl support, the script that builds the
7638 Makefile was calling 'perl' to find its compile-time parameters, ignoring
7639 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
7641 6. The freeze_tell option was not being used for messages that were frozen on
7642 arrival, either by an ACL or by local_scan().
7644 7. Added the smtp_incomplete_transaction log selector.
7646 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
7647 was accepting AUTH without a new EHLO.
7649 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
7650 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
7653 10. Logging of TCP/IP connections (when configured) now happens in the main
7654 daemon process instead of the child process, so that the TCP/IP connection
7655 count is more accurate (but it can never be perfect).
7657 11. The use of "drop" in a nested ACL was not being handled correctly in the
7658 outer ACL. Now, if condition failure induced by the nested "drop" causes
7659 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
7660 or "require"), the connection is dropped.
7662 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
7663 that yield "discard" can now be used with an "accept" or a "discard" verb,
7664 but an error is generated for any others (because I can't see a useful way
7665 to define what should happen).
7667 13. When an ACL is read dynamically from a file (or anywhere else), the lines
7668 are now processed in the same way as lines in the Exim configuration file.
7669 In particular, continuation lines are supported.
7671 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
7673 15. Added -ti meaning -t -i.
7675 16. Check for letters, digits, hyphens, and dots in the names of dnslist
7676 domains, and warn by logging if others are found.
7678 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
7679 returned from GIFCONF when Exim is trying to find the list of interfaces on
7680 a host. The code in os.c has been modified to copy each ifreq to an aligned
7681 structure in all cases.
7683 Also, in some cases, the returned ifreq's were being copied to a 'struct
7684 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
7685 means the last couple of bytes of an IPv6 address could be chopped if the
7686 ifreq contained only a normal sockaddr (14 bytes storage).
7688 18. Named domain lists were not supported in the hosts_treat_as_local option.
7689 An entry such as +xxxx was not recognized, and was treated as a literal
7692 19. Ensure that header lines added by a DATA ACL are included in the reject log
7693 if the ACL subsequently rejects the message.
7695 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
7696 MD5 (which is deprecated).
7698 21. When testing a filter file using -bf, Exim was writing a message when it
7699 took the sender from a "From " line in the message, but it was not doing so
7700 when it took $return_path from a Return-Path: header line. It now does.
7702 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
7703 with a valid header line field name (a series of printing characters
7704 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
7706 23. Changed "disc" in the source to "disk" to conform to the documentation and
7707 the book and for uniformity.
7709 24. Ignore Sendmail's -Ooption=value command line item.
7711 25. When execve() failed while trying to run a command in a pipe transport,
7712 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
7713 could be confused with a return value of 69 from the command itself. This
7714 has been changed to 127, the value the shell returns if it is asked to run
7715 a non-existent command. The wording for the related log line suggests a
7716 non-existent command as the problem.
7718 26. If received_header_text expands to an empty string, do not add a Received:
7719 header line to the message. (Well, it adds a token one on the spool, but
7720 marks it "old" so that it doesn't get used or transmitted.)
7722 27. Installed eximstats 1.28 (addition of -nt option).
7724 28. There was no check for failure on the call to getsockname() in the daemon
7725 code. This can fail if there is a shortage of resources on the system, with
7726 ENOMEM, for example. A temporary error is now given on failure.
7728 29. Contrary to the C standard, it seems that in some environments, the
7729 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
7730 program. Exim now does this explicitly; it affects the formatting of
7731 timestamps using strftime().
7733 30. If exiqsumm was given junk data, it threw up some uninitialized variable
7734 complaints. I've now initialized all the variables, to avoid this.
7736 32. Header lines added by a system filter were not being "seen" during
7737 transport-time rewrites.
7739 33. The info_callback() function passed to OpenSSL is set up with type void
7740 (*)(SSL *, int, int), as described somewhere. However, when calling the
7741 function (actually a macro) that sets it up, the type void(*)() is
7742 expected. I've put in a cast to prevent warnings from picky compilers.
7744 34. If a DNS black list lookup found a CNAME record, but there were no A
7745 records associated with the domain it pointed at, Exim crashed.
7747 35. If a DNS black list lookup returned more than one A record, Exim ignored
7748 all but the first. It now scans all returned addresses if a particular IP
7749 value is being sought. In this situation, the contents of the
7750 $dnslist_value variable are a list of all the addresses, separated by a
7753 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
7754 to accept a host name and all its aliases if the forward lookup for any of
7755 them yielded the IP address of the incoming connection. Now it accepts only
7756 those names whose forward lookup yields the correct IP address. Any other
7757 names are discarded. This closes a loophole whereby a rogue DNS
7758 administrator could create reverse DNS records to break through a
7759 wildcarded host restriction in an ACL.
7761 37. If a user filter or a system filter that ran in a subprocess used any of
7762 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
7763 the wrong values were passed to the pipe command ($thisaddress had the
7764 value of $0, $0 had the value of $1, etc). This bug was introduced by
7765 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
7767 38. Improved the line breaking for long SMTP error messages from ACLs.
7768 Previously, if there was no break point between 40 and 75 characters, Exim
7769 left the rest of the message alone. Two changes have been made: (a) I've
7770 reduced the minimum length to 35 characters; (b) if it can't find a break
7771 point between 35 and 75 characters, it looks ahead and uses the first one
7772 that it finds. This may give the occasional overlong line, but at least the
7773 remaining text gets split now.
7775 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
7776 file descriptors might be low, and that setting 1000 would always raise it.
7777 It turns out that in some environments, the limit is already over 1000 and
7778 that lowering it causes trouble. So now Exim takes care not to decrease it.
7780 40. When delivering a message, the value of $return_path is set to $sender_
7781 address at the start of routing (routers may change the value). By an
7782 oversight, this default was not being set up when an address was tested by
7783 -bt or -bv, which affected the outcome if any router or filter referred to
7786 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
7787 log only when "message" or "log_message" are set. However, if one of the
7788 conditions was an address verification, or a call to a nested ACL, the
7789 messages generated by the underlying test were being passed through. This
7790 no longer happens. The underlying message is available in $acl_verify_
7791 message for both "message" and "log_message" expansions, so it can be
7792 passed through if needed.
7794 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
7795 new expansion $bh_ to give the encoded byte string without charset
7796 translation. Translation happens only if iconv() is available; HAVE_ICONV
7797 indicates this at build time. HEADERS_CHARSET gives the charset to
7798 translate to; headers_charset can change it in the configuration, and
7799 "headers charset" can change it in an individual filter file.
7801 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
7802 that creates RFC 2047 encoded "words" labels them as that charset instead
7803 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
7804 expansion operator; (ii) when Exim creates a From: line for a local
7805 message; (iii) when a header line is rewritten to include a "phrase" part.
7807 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
7808 buggy, causing it to skip the first lines of messages whose message ID
7809 ended in 'D'. This would not have bitten before Exim release 4.14, because
7810 message IDs were unlikely to end in 'D' before then. The effect was to have
7811 incorrect size information for certain domains.
7813 45. #include "config.h" was missing at the start of the crypt16.c module. This
7814 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
7817 46. If there was a timeout during a "random" callout check, Exim treated it as
7818 a failure of the random address, and carried on sending RSET and the real
7819 address. If the delay was just some slowness somewhere, the response to the
7820 original RCPT would be taken as a response to RSET and so on, causing
7821 mayhem of various kinds.
7823 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
7824 when I implemented it. It didn't allow for the fact that some option values
7825 may legitimately be negative (e.g. size_addition), and it didn't even do
7826 the right test for positive values.
7828 48. Domain names in DNS records are case-independent. Exim always looks them up
7829 in lower case. Some resolvers return domain names in exactly the case they
7830 appear in the zone file, that is, they may contain uppercase letters. Not
7831 all resolvers do this - some return always lower case. Exim was treating a
7832 change of case by a resolver as a change of domain, similar to a widening
7833 of a domain abbreviation. This triggered its re-routing code and so it was
7834 trying to route what was effectively the same domain again. This normally
7835 caused routing to fail (because the router wouldn't handle the domain
7836 twice). Now Exim checks for this case specially, and just changes the
7837 casing of the domain that it ultimately uses when it transmits the message
7840 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
7843 50. If a filter generated a file delivery with a non-absolute name (possible if
7844 no home directory exists for the router), the forbid_file option was not
7847 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
7848 the existing equality matching.
7850 52. Exim was using ints instead of ino_t variables in some places where it was
7851 dealing with inode numbers.
7853 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
7854 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
7855 called TMPDIR, and if it finds it is different, it changes its value.
7857 54. The smtp_printf() function is now made available to local_scan() so
7858 additional output lines can be written before returning. There is also an
7859 smtp_fflush() function to enable the detection of a dropped connection.
7860 The variables smtp_input and smtp_batched_input are exported to
7863 55. Changed the default runtime configuration: the message "Unknown user"
7864 has been removed from the ACL, and instead placed on the localuser router,
7865 using the cannot_route_message feature. This means that any verification
7866 failures that generate their own messages won't get overridden. Similarly,
7867 the "Unrouteable address" message that was in the ACL for unverifiable
7868 relay addresses has also been removed.
7870 56. Added hosts_avoid_esmtp to the smtp transport.
7872 57. The exicyclog script was not checking for the esoteric option
7873 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7874 will work only if exicyclog is run under the appropriate euid.
7876 58. Following a discussion on the list, the rules by which Exim recognises line
7877 endings on incoming messages have been changed. The -dropcr and drop_cr
7878 options are now no-ops, retained only for backwards compatibility. The
7879 following line terminators are recognized: LF CRLF CR. However, special
7880 processing applies to CR:
7882 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7883 nor a local message in the state where . is a terminator.
7885 (ii) If a bare CR is encountered in a header line, an extra space is added
7886 after the line terminator so as not to end the header. The reasoning
7887 behind this is that bare CRs in header lines are most likely either
7888 to be mistakes, or people trying to play silly games.
7890 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7891 was being incorrectly given as 18 bytes larger than it should have been.
7892 This is a VOB (very old bug).
7894 60. This may never have affected anything current, but just in case it has:
7895 When the local host is found other than at the start of a list of hosts,
7896 the local host, those with the same MX, and any that follow, are discarded.
7897 When the list in question was part of a longer list of hosts, the following
7898 hosts (not currently being processed) were also being discarded. This no
7899 longer happens. I'm not sure if this situation could ever has previously
7902 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7905 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7906 header. This is something that is recommended in a new Internet Draft, and
7907 is something that is documented as being done by Sendmail. There are two
7908 possible values. For messages generated by the autoreply transport, Exim
7911 Auto-Submitted: auto-replied
7913 whereas for all other generated messages (e.g. bounces) it adds
7915 Auto-Submitted: auto-generated
7917 63. The "personal" condition in filters now includes a test for the
7918 Auto-Submitted: header. If it contains the string "auto-" the message it
7919 not considered personal.
7921 64. Added rcpt_include_affixes as a generic transport option.
7923 65. Added queue_only_override (default true).
7925 66. Added the syslog_duplication option.
7927 67. If what should have been the first header line of a message consisted of
7928 a space followed by a colon, Exim was mis-interpreting it as a header line.
7929 It isn't of course - it is syntactically invalid and should therefore be
7930 treated as the start of the message body. The misbehaviour could have
7931 caused a number of strange effects, including loss of data in subsequent
7932 header lines, and spool format errors.
7934 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7935 client host had authenticated. This control can now be exercised by an ACL
7936 for more flexibility.
7938 69. By default, callouts do not happen when testing with -bh. There is now a
7939 variant, -bhc, which does actually run the callout code, including
7940 consulting and updating the callout cache.
7942 70. Added support for saslauthd authentication, courtesy of Alexander
7945 71. If statvfs() failed on the spool or log directories while checking their
7946 size for availability, Exim confusingly gave the error "space shortage".
7947 Furthermore, in debugging mode it crashed with a floating point exception.
7948 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7949 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7950 serious problem, Exim now writes to the main and panic logs when this
7951 happens, with details of the failure. It then refuses to accept the
7952 incoming message, giving the message "spool directory problem" or "log
7953 directory problem" with a 421 code for SMTP messages.
7955 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7956 0, 1, and 2 exist, because some OS complain for execs without them (see
7957 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7958 descriptors. However, the code omitted to check that the open succeeded,
7959 causing mysterious errors if for some reason the permissions on /dev/null
7960 got screwed. Now Exim writes a message to the main and panic logs, and
7961 bombs out if it can't open /dev/null.
7963 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7964 interact so that it is all more flexible. It is supposed to remain
7965 backwards compatible. Also added extra_local_interfaces.
7967 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7968 to bomb out with an assertion failure - to the client this appears as a
7969 connection drop. This problem occurs in the part of the code that was taken
7970 from the Samba project. Fortunately, the assertion is in a very simple
7971 function, so I have fixed this by reproducing the function inline in the
7972 one place where it is called, and arranging for authentication to fail
7973 instead of killing the process with assert().
7975 75. The SPA client code was not working when the server requested OEM rather
7976 than Unicode encoding.
7978 76. Added code to make require_files with a specific uid setting more usable in
7979 the case where statting the file as root fails - usually a non-root-mounted
7980 NFS file system. When this happens and the failure is EACCES, Exim now
7981 forks a subprocess and does the per-uid checking as the relevant uid.
7983 77. Added process_log_path.
7985 78. If log_file_path was not explicitly set, a setting of check_log_space or
7986 check_log_inodes was ignored.
7988 79. If a space check for the spool or log partitions fails, the incident is now
7989 logged. Of course, in the latter case the data may get lost...
7991 80. Added the %p formatting code to string_format() so that it can be used to
7992 print addresses in debug_print(). Adjusted all the address printing in the
7993 debugging in store.c to use %p rather than %d.
7995 81. There was a concern that a line of code in smtp_in.c could overflow a
7996 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7997 initially expressed, the concern was not well-founded, because trailing
7998 spaces are removed early. However, if the trailing spaces were followed by
7999 a NULL, they did not get removed, so the overflow was possible. Two fixes
8002 (a) I re-wrote the offending code in a cleaner fashion.
8003 (b) If an incoming SMTP command contains a NULL character, it is rejected
8006 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
8007 runs initgroups(), so that if the Exim user is in any additional groups,
8008 they will be used during message reception.
8014 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob