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.
229 JH/48 Use a less bogus-looking filename for a temporary used for DH-parameters
230 for GnuTLS. Previously the name started "%s" which, while not a bug,
231 looked as if if might be one.
233 JH/49 Bug 2710: when using SOCKS for additional messages after the first (a
234 "continued connection") make the $proxy_* variables available. Previously
235 the information was not passed across the exec() call for subsequent
236 transport executions. This also mean that the log lines for the
237 messages can show the proxy information.
244 JH/01 Avoid costly startup code when not strictly needed. This reduces time
245 for some exim process initialisations. It does mean that the logging
246 of TLS configuration problems is only done for the daemon startup.
248 JH/02 Early-pipelining support code is now included unless disabled in Makefile.
250 JH/03 DKIM verification defaults no long accept sha1 hashes, to conform to
251 RFC 8301. They can still be enabled, using the dkim_verify_hashes main
254 JH/04 Support CHUNKING from an smtp transport using a transport_filter, when
255 DKIM signing is being done. Previously a transport_filter would always
256 disable CHUNKING, falling back to traditional DATA.
258 JH/05 Regard command-line recipients as tainted.
260 JH/06 Bug 340: Remove the daemon pid file on exit, when due to SIGTERM.
262 JH/07 Bug 2489: Fix crash in the "pam" expansion condition. It seems that the
263 PAM library frees one of the arguments given to it, despite the
264 documentation. Therefore a plain malloc must be used.
266 JH/08 Bug 2491: Use tainted buffers for the transport smtp context. Previously
267 on-stack buffers were used, resulting in a taint trap when DSN information
268 copied from a received message was written into the buffer.
270 JH/09 Bug 2493: Harden ARC verify against Outlook, whick has been seen to mix
271 the ordering of its ARC headers. This caused a crash.
273 JH/10 Bug 2492: Use tainted memory for retry record when needed. Previously when
274 a new record was being constructed with information from the peer, a trap
277 JH/11 Bug 2494: Unset the default for dmarc_tld_file. Previously a naiive
278 installation would get error messages from DMARC verify, when it hit the
279 nonexistent file indicated by the default. Distros wanting DMARC enabled
280 should both provide the file and set the option.
281 Also enforce no DMARC verification for command-line sourced messages.
283 JH/12 Fix an uninitialised flag in early-pipelining. Previously connections
284 could, depending on the platform, hang at the STARTTLS response.
286 JH/13 Bug 2498: Reset a counter used for ARC verify before handling another
287 message on a connection. Previously if one message had ARC headers and
288 the following one did not, a crash could result when adding an
289 Authentication-Results: header.
291 JH/14 Bug 2500: Rewind some of the common-coding in string handling between the
292 Exim main code and Exim-related utities. The introduction of taint
293 tracking also did many adjustments to string handling. Since then, eximon
294 frequently terminated with an assert failure.
296 JH/15 When PIPELINING, synch after every hundred or so RCPT commands sent and
297 check for 452 responses. This slightly helps the inefficieny of doing
298 a large alias-expansion into a recipient-limited target. The max_rcpt
299 transport option still applies (and at the current default, will override
300 the new feature). The check is done for either cause of synch, and forces
301 a fast-retry of all 452'd recipients using a new MAIL FROM on the same
302 connection. The new facility is not tunable at this time.
304 JH/16 Fix the variables set by the gsasl authenticator. Previously a pointer to
305 library live data was being used, so the results became garbage. Make
306 copies while it is still usable.
308 JH/17 Logging: when the deliver_time selector ise set, include the DT= field
309 on delivery deferred (==) and failed (**) lines (if a delivery was
310 attemtped). Previously it was only on completion (=>) lines.
312 JH/18 Authentication: the gsasl driver not provides the $authN variables in time
313 for the expansion of the server_scram_iter and server_scram_salt options.
315 WB/01 SPF: DNS lookups for the obsolete SPF RR type done by the libspf2 library
316 are now specifically given a NO_DATA response without hitting the system
317 resolver. The library goes on to do the now-standard TXT lookup.
318 Use of dnsdb lookups is not affected.
320 JH/19 Bug 2507: Modules: on handling a dynamic-module (lookups) open failure,
321 only retrieve the errormessage once. Previously two calls to dlerror()
322 were used, and the second one (for mainlog/paniclog) retrieved null
325 JH/20 Taint checking: disallow use of tainted data for
326 - the appendfile transport file and directory options
327 - the pipe transport command
328 - the autoreply transport file, log and once options
329 - file names used by the redirect router (including filter files)
331 - paths used by single-key lookups
332 Previously this was permitted.
334 JH/21 Bug 2501: Fix init call in the heimdal authenticator. Previously it
335 adjusted the size of a major service buffer; this failed because the
336 buffer was in use at the time. Change to a compile-time increase in the
337 buffer size, when this authenticator is compiled into exim.
339 JH/22 Taint-checking: move to safe-mode taint checking on all platforms. The
340 previous fast-mode was untenable in the face of glibs using mmap to
341 support larger malloc requests.
343 PP/01 Update the openssl_options possible values through OpenSSL 1.1.1c.
344 New values supported, if defined on system where compiled:
345 allow_no_dhe_kex, cryptopro_tlsext_bug, enable_middlebox_compat,
346 no_anti_replay, no_encrypt_then_mac, prioritize_chacha, tlsext_padding
348 JH/23 Performance improvement in the initial phase of a two-pass queue run. By
349 running a limited number of proceses in parallel, a benefit is gained. The
350 amount varies with the platform hardware and load. The use of the option
351 queue_run_in_order means we cannot do this, as ordering becomes
354 JH/24 Bug 2524: fix the cyrus_sasl auth driver gssapi usage. A previous fix
355 had introduced a string-copy (for ensuring NUL-termination) which was not
356 appropriate for that case, which can include embedded NUL bytes in the
357 block of data. Investigation showed the copy to actually be needless, the
358 data being length-specified.
360 JH/25 Fix use of concurrent TLS connections under GnuTLS. When a callout was
361 done during a receiving connection, and both used TLS, global info was
362 used rather than per-connection info for tracking the state of data
363 queued for transmission. This could result in a connection hang.
365 JH/26 Fix use of the SIZE parameter on MAIL commands, on continued connections.
366 Previously, when delivering serveral messages down a single connection
367 only the first would provide a SIZE. This was due to the size information
368 not being properly tracked.
370 JH/27 Bug 2530: When operating in a timezone with sub-minute offset, such as
371 TAI (at 37 seconds currently), pretend to be in UTC for time-related
372 expansion and logging. Previously, spurious values such as a future
373 minute could be seen.
375 JH/28 Bug 2533: Fix expansion of ${tr } item. When called in some situations
376 it could crash from a null-deref. This could also affect the
377 ${addresses: } operator and ${readsock } item.
379 JH/29 Bug 2537: Fix $mime_part_count. When a single connection had a non-mime
380 message following a mime one, the variable was not reset.
382 JH/30 When an pipelined-connect fails at the first response, assume incorrect
383 cached capability (perhaps the peer reneged?) and immediately retry in
386 JH/31 Fix spurious detection of timeout while writing to transport filter.
388 JH/32 Bug 2541: Fix segfault on bad cmdline -f (sender) argument. Previously
389 an attempt to copy the string was made before checking it.
391 JH/33 Fix the dsearch lookup to return an untainted result. Previously the
392 taint of the lookup key was maintained; we now regard the presence in the
393 filesystem as sufficient validation.
395 JH/34 Fix the readsocket expansion to not segfault when an empty "options"
396 argument is supplied.
398 JH/35 The dsearch lookup now requires that the directory is an absolute path.
399 Previously this was not checked, and nonempty relative paths made an
400 access under Exim's current working directory.
402 JH/36 Bug 2554: Fix msg:defer event for the hosts_max_try_hardlimit case.
403 Previously no event was raised.
405 JH/37 Bug 2552: Fix the check on spool space during reception to use the SIZE
406 parameter supplied by the sender MAIL FROM command. Previously it was
407 ignored, and only the check_spool_space option value for the required
410 JH/38 Fix $dkim_key_length. This should, after a DKIM verification, present
411 the size of the signing public-key. Previously it was instead giving
412 the size of the signature hash.
414 JH/39 DKIM verification: the RFC 8301 restriction on sizes of RSA keys is now
415 the default. See the (new) dkim_verify_min_keysizes option.
417 JH/40 Fix a memory-handling bug: when a connection carried multiple messages
418 and an ACL use a lookup for checking either the local_part or domain,
419 stale data could be accessed. Ensure that variable references are
420 dropped between messages.
422 JH/41 Bug 2571: Fix SPA authenticator. Running as a server, an offset supplied
423 by the client was not checked as pointing within response data before
424 being used. A malicious client could thus cause an out-of-bounds read and
425 possibly gain authentication. Fix by adding the check.
427 JH/42 Internationalisation: change the default for downconversion in the smtp
428 transport to be "if needed". Previously it was "as previously set" for
429 the message, which usually meant "if needed" for message-submission but
430 "no" for everything else. However, MTAs have been seen using SMTPUTF8
431 even when the envelope addresses did not need it, resulting in forwarding
432 failures to non-supporting MTAs. A downconvert in such cases will be
433 a no-op on the addresses, merely dropping the use of SMTPUTF8 by the
434 transport. The change does mean that addresses needing conversion will
435 be converted when previously a delivery failure would occur.
437 JH/43 Fix possible long line in DSN. Previously when a very long SMTP error
438 response was received it would be used unchecked in a fail-DSN, violating
439 standards on line-length limits. Truncate if needed.
441 HS/01 Remove parameters of the link to www.open-spf.org. The linked form
442 doesn't work. (Additionally add a new main config option to configure the
449 JH/01 OpenSSL: With debug enabled output keying information sufficient, server
450 side, to decode a TLS 1.3 packet capture.
452 JH/02 OpenSSL: Suppress the sending of (stateful) TLS1.3 session tickets.
453 Previously the default library behaviour applied, sending two, each in
456 JH/03 Debug output for ACL now gives the config file name and line number for
459 JH/04 The default received_header_text now uses the RFC 8314 tls cipher clause.
461 JH/05 DKIM: ensure that dkim_domain elements are lowercased before use.
463 JH/06 Fix buggy handling of autoreply bounce_return_size_limit, and a possible
464 buffer overrun for (non-chunking) other transports.
466 JH/07 GnuTLS: Our use of late (post-handshake) certificate verification, under
467 TLS1.3, means that a server rejecting a client certificate is not visible
468 to the client until the first read of encrypted data (typically the
469 response to EHLO). Add detection for that case and treat it as a failed
470 TLS connection attempt, so that the normal retry-in-clear can work (if
471 suitably configured).
473 JB/01 Bug 2375: fix expansions of 822 addresses having comments in local-part
474 and/or domain. Found and fixed by Jason Betts.
476 JH/08 Add hardening against SRV & TLSA lookups the hit CNAMEs (a nonvalid
477 configuration). If a CNAME target was not a wellformed name pattern, a
480 JH/09 Logging: Fix initial listening-on line for multiple ports for an IP when
481 the OS reports them interleaved with other addresses.
483 JH/10 OpenSSL: Fix aggregation of messages. Previously, when PIPELINING was
484 used both for input and for a verify callout, both encrypted, SMTP
485 responses being sent by the server could be lost. This resulted in
486 dropped connections and sometimes bounces generated by a peer sending
489 JH/11 Harden plaintext authenticator against a badly misconfigured client-send
490 string. Previously it was possible to cause undefined behaviour in a
491 library routine (usually a crash). Found by "zerons".
493 JH/12 Bug 2384: fix "-bP smtp_receive_timeout". Previously it returned no
496 JH/13 Bug 2386: Fix builds with Dane under LibreSSL 2.9.0 onward. Some old
497 API was removed, so update to use the newer ones.
499 JH/14 Bug 1891: Close the log file if receiving a non-smtp message, without
500 any timeout set, is taking a long time. Previously we would hang on to a
501 rotated logfile "forever" if the input was arriving with long gaps
502 (a previous attempt to fix addressed lack, for a long time, of initial
505 HS/01 Bug 2390: Use message_id for tempfile creation to avoid races in a
506 shared (NFS) environment. The length of the tempfile name is now
507 4 + 16 ("hdr.$message_exim_id") which might break on file
508 systems which restrict the file name length to lower values.
511 HS/02 Bug 2390: Use message_id for tempfile creation to avoid races in a
512 shared (NFS) environment.
514 HS/03 Bug 2392: exigrep does case sensitive *option* processing (as it
515 did for all versions <4.90). Notably -M, -m, --invert, -I may be
518 JH/15 Use unsigned when creating bitmasks in macros, to avoid build errors
519 on some platforms for bit 31.
521 JH/16 GnuTLS: rework ciphersuite strings under recent library versions. Thanks
522 to changes apparently associated with TLS1.3 handling some of the APIs
523 previously used were either nonfunctional or inappropriate. Strings
524 like TLS1.3:ECDHE_SECP256R1__RSA_PSS_RSAE_SHA256__AES_256_GCM__AEAD:256
525 and TLS1.2:ECDHE_SECP256R1__RSA_SHA256__AES_128_CBC__SHA256:128 replace
526 the previous TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256 .
527 This affects log line X= elements, the $tls_{in,out}_cipher variables,
528 and the use of specific cipher names in the encrypted= ACL condition.
530 JH/17 OpenSSL: the default openssl_options now disables ssl_v3.
532 JH/18 GnuTLS: fix $tls_out_ocsp under hosts_request_ocsp. Previously the
533 verification result was not updated unless hosts_require_ocsp applied.
535 JH/19 Bug 2398: fix listing of a named-queue. Previously, even with the option
536 queue_list_requires_admin set to false, non-admin users were denied the
539 JH/20 Bug 2389: fix server advertising of usable certificates, under GnuTLS in
540 directory-of-certs mode. Previously they were advertised despite the
543 JH/21 The smtp transport option "hosts_noproxy_tls" is now unset by default.
544 A single TCP connection by a client will now hold a TLS connection open
545 for multiple message deliveries, by default. Previously the default was to
548 JH/22 The smtp transport option "hosts_try_dane" now enables all hosts by
549 default. If built with the facility, DANE will be used. The facility
550 SUPPORT_DANE is now enabled in the prototype build Makefile "EDITME".
552 JH/23 The build default is now for TLS to be included; the SUPPORT_TLS define
553 is replaced with DISABLE_TLS. Either USE_GNUTLS or (the new) USE_OPENSSL
554 must be defined and you must still, unless you define DISABLE_TLS, manage
555 the the include-dir and library-file requirements that go with that
556 choice. Non-TLS builds are still supported.
558 JH/24 Fix duplicated logging of peer name/address, on a transport connection-
561 JH/25 The smtp transport option "hosts_try_fastopen" now enables all hosts by
562 default. If the platform supports and has the facility enabled, it will
563 be requested on all coneections.
565 JH/26 The PIPE_CONNECT facility is promoted from experimental status and is now
566 controlled by the build-time option SUPPORT_PIPE_CONNECT.
568 PP/01 Unbreak heimdal_gssapi, broken in 4.92.
570 JH/27 Bug 2404: Use the main-section configuration option "dsn_from" for
571 success-DSN messages. Previously the From: header was always the default
572 one for these; the option was ignored.
574 JH/28 Fix the timeout on smtp response to apply to the whole response.
575 Previously it was reset for every read, so a teergrubing peer sending
576 single bytes within the time limit could extend the connection for a
577 long time. Credit to Qualsys Security Advisory Team for the discovery.
579 JH/29 Fix DSN Final-Recipient: field. Previously it was the post-routing
580 delivery address, which leaked information of the results of local
581 forwarding. Change to the original envelope recipient address, per
584 JH/30 Bug 2411: Fix DSN generation when RFC 3461 failure notification is
585 requested. Previously not bounce was generated and a log entry of
586 error ignored was made.
588 JH/31 Avoid re-expansion in ${sort } expansion. (CVE-2019-13917)
590 JH/32 Introduce a general tainting mechanism for values read from the input
591 channel, and values derived from them. Refuse to expand any tainted
592 values, to catch one form of exploit.
594 JH/33 Bug 2413: Fix dkim_strict option. Previously the expansion result
595 was unused and the unexpanded text used for the test. Found and
596 fixed by Ruben Jenster.
598 JH/34 Fix crash after TLS shutdown. When the TCP/SMTP channel was left open,
599 an attempt to use a TLS library read routine dereffed a nul pointer,
602 JH/35 Bug 2409: filter out-of-spec chars from callout response before using
603 them in our smtp response.
605 JH/36 Have the general router option retry_use_local_part default to true when
606 any of the restrictive preconditions are set (to anything). Previously it
607 was only for check_local user. The change removes one item of manual
608 configuration which is required for proper retries when a remote router
609 handles a subset of addresses for a domain.
611 JH/37 Appendfile: when evaluating quota use (non-quota_size_regex) take the file
612 link count into consideration.
614 HS/04 Fix handling of very log lines in -H files. If a -<key> <value> line
615 caused the extension of big_buffer, the following lines were ignored.
617 JH/38 Bug 1395: Teach the DNS negative-cache about TTL value from the SOA in
618 accordance with RFC 2308. Previously there was no expiry, so a longlived
619 receive process (eg. due to ACL delays) versus a short SOA value could
622 HS/05 Handle trailing backslash gracefully. (CVE-2019-15846)
624 JH/39 Promote DMARC support to mainline.
626 JH/40 Bug 2452: Add a References: header to DSNs.
628 JH/41 With GnuTLS 3.6.0 (and later) do not attempt to manage Diffie-Hellman
629 parameters. The relevant library call is documented as "Deprecated: This
630 function is unnecessary and discouraged on GnuTLS 3.6.0 or later. Since
631 3.6.0, DH parameters are negotiated following RFC7919."
633 HS/06 Change the default of dnssec_request_domains to "*"
635 JH/42 Bug 2545: Fix CHUNKING for all RCPT commands rejected. Previously we
636 carried on and emitted a BDAT command, even when PIPELINING was not
639 JH/43 Bug 2465: Fix taint-handling in dsearch lookup. Previously a nontainted
640 buffer was used for the filename, resulting in a trap when tainted
641 arguments (eg. $domain) were used.
643 JH/44 With OpenSSL 1.1.1 (onwards) disable renegotiation for TLS1.2 and below;
644 recommended to avoid a possible server-load attack. The feature can be
645 re-enabled via the openssl_options main cofiguration option.
647 JH/45 local_scan API: documented the current smtp_printf() call. This changed
648 for version 4.90 - adding a "more data" boolean to the arguments.
649 Bumped the ABI version number also, this having been missed previously;
650 release versions 4.90 to 4.92.3 inclusive were effectively broken in
651 respect of usage of smtp_printf() by either local_scan code or libraries
652 accessed via the ${dlfunc } expansion item. Both will need coding
653 adjustment for any calls to smtp_printf() to match the new function
654 signature; a FALSE value for the new argument is always safe.
656 JH/46 FreeBSD: fix use of the sendfile() syscall. The shim was not updating
657 the file-offset (which the Linux syscall does, and exim expects); this
658 resulted in an indefinite loop.
660 JH/47 ARC: fix crash in signing, triggered when a configuration error failed
661 to do ARC verification. The Authentication-Results: header line added
662 by the configuration then had no ARC item.
668 JH/01 Remove code calling the customisable local_scan function, unless a new
669 definition "HAVE_LOCAL_SCAN=yes" is present in the Local/Makefile.
671 JH/02 Bug 1007: Avoid doing logging from signal-handlers, as that can result in
672 non-signal-safe functions being used.
674 JH/03 Bug 2269: When presented with a received message having a stupidly large
675 number of DKIM-Signature headers, disable DKIM verification to avoid
676 a resource-consumption attack. The limit is set at twenty.
678 JH/04 Add variables $arc_domains, $arc_oldest_pass for ARC verify. Fix the
679 report of oldest_pass in ${authres } in consequence, and separate out
680 some descriptions of reasons for verification fail.
682 JH/05 Bug 2273: Cutthrough delivery left a window where the received messsage
683 files in the spool were present and unlocked. A queue-runner could spot
684 them, resulting in a duplicate delivery. Fix that by doing the unlock
685 after the unlink. Investigation by Tim Stewart. Take the opportunity to
686 add more error-checking on spoolfile handling while that code is being
689 PP/01 Refuse to open a spool data file (*-D) if it's a symlink.
690 No known attacks, no CVE, this is defensive hardening.
692 JH/06 Bug 2275: The MIME ACL unlocked the received message files early, and
693 a queue-runner could start a delivery while other operations were ongoing.
694 Cutthrough delivery was a common victim, resulting in duplicate delivery.
695 Found and investigated by Tim Stewart. Fix by using the open message data
696 file handle rather than opening another, and not locally closing it (which
697 releases a lock) for that case, while creating the temporary .eml format
698 file for the MIME ACL. Also applies to "regex" and "spam" ACL conditions.
700 JH/07 Bug 177: Make a random-recipient callout success visible in ACL, by setting
701 $sender_verify_failure/$recipient_verify_failure to "random".
703 JH/08 When generating a selfsigned cert, use serial number 1 since zero is not
706 JH/09 Bug 2274: Fix logging of cmdline args when starting in an unlinked cwd.
707 Previously this would segfault.
709 JH/10 Fix ARC signing for case when DKIM signing failed. Previously this would
712 JH/11 Bug 2264: Exim now only follows CNAME chains one step by default. We'd
713 like zero, since the resolver should be doing this for us, But we need one
714 as a CNAME but no MX presence gets the CNAME returned; we need to check
715 that doesn't point to an MX to declare it "no MX returned" rather than
716 "error, loop". A new main option is added so the older capability of
717 following some limited number of chain links is maintained.
719 JH/12 Add client-ip info to non-pass iprev ${authres } lines.
721 JH/13 For receent Openssl versions (1.1 onward) use modern generic protocol
722 methods. These should support TLS 1.3; they arrived with TLS 1.3 and the
723 now-deprecated earlier definitions used only specified the range up to TLS
724 1.2 (in the older-version library docs).
726 JH/14 Bug 2284: Fix DKIM signing for body lines starting with a pair of dots.
728 JH/15 Rework TLS client-side context management. Stop using a global, and
729 explicitly pass a context around. This enables future use of TLS for
730 connections to service-daemons (eg. malware scanning) while a client smtp
731 connection is using TLS; with cutthrough connections this is quite likely.
733 JH/16 Fix ARC verification to do AS checks in reverse order.
735 JH/17 Support a "tls" option on the ${readsocket } expansion item.
737 JH/18 Bug 2287: Fix the protocol name (eg utf8esmtp) for multiple messages
738 using the SMTPUTF8 option on their MAIL FROM commands, in one connection.
739 Previously the "utf8" would be re-prepended for every additional message.
741 JH/19 Reject MAIL FROM commands with SMTPUTF8 when the facility was not advertised.
742 Previously thery were accepted, resulting in issues when attempting to
743 forward messages to a non-supporting MTA.
745 PP/02 Let -n work with printing macros too, not just options.
747 JH/20 Bug 2296: Fix cutthrough for >1 address redirection. Previously only
748 one parent address was copied, and bogus data was used at delivery-logging
749 time. Either a crash (after delivery) or bogus log data could result.
750 Discovery and analysis by Tim Stewart.
752 PP/03 Make ${utf8clean:} expansion operator detect incomplete final character.
753 Previously if the string ended mid-character, we did not insert the
754 promised '?' replacement.
756 PP/04 Documentation: current string operators work on bytes, not codepoints.
758 JH/21 Change as many as possible of the global flags into one-bit bitfields; these
759 should pack well giving a smaller memory footprint so better caching and
760 therefore performance. Group the declarations where this can't be done so
761 that the byte-sized flag variables are not interspersed among pointer
762 variables, giving a better chance of good packing by the compiler.
764 JH/22 Bug 1896: Fix the envelope from for DMARC forensic reports to be possibly
765 non-null, to avoid issues with sites running BATV. Previously reports were
766 sent with an empty envelope sender so looked like bounces.
768 JH/23 Bug 2318: Fix the noerror command within filters. It wasn't working.
769 The ignore_error flag wasn't being returned from the filter subprocess so
770 was not set for later routers. Investigation and fix by Matthias Kurz.
772 JH/24 Bug 2310: Raise a msg:fail:internal event for each undelivered recipient,
773 and a msg:complete for the whole, when a message is manually removed using
774 -Mrm. Developement by Matthias Kurz, hacked on by JH.
776 JH/25 Avoid fixed-size buffers for pathnames in DB access. This required using
777 a "Gnu special" function, asprintf() in the DB utility binary builds; I
778 hope that is portable enough.
780 JH/26 Bug 2311: Fix DANE-TA verification under GnuTLS. Previously it was also
781 requiring a known-CA anchor certificate; make it now rely entirely on the
782 TLSA as an anchor. Checking the name on the leaf cert against the name
783 on the A-record for the host is still done for TA (but not for EE mode).
785 JH/27 Fix logging of proxy address. Previously, a pointless "PRX=[]:0" would be
786 included in delivery lines for non-proxied connections, when compiled with
787 SUPPORT_SOCKS and running with proxy logging enabled.
789 JH/28 Bug 2314: Fire msg:fail:delivery event even when error is being ignored.
790 Developement by Matthias Kurz, tweaked by JH. While in that bit of code,
791 move the existing event to fire before the normal logging of message
792 failure so that custom logging is bracketed by normal logging.
794 JH/29 Bug 2322: A "fail" command in a non-system filter (file) now fires the
795 msg:fail:internal event. Developement by Matthias Kurz.
797 JH/30 Bug 2329: Increase buffer size used for dns lookup from 2k, which was
798 far too small for todays use of crypto signatures stored there. Go all
799 the way to the max DNS message size of 64kB, even though this might be
800 overmuch for IOT constrained device use.
802 JH/31 Fix a bad use of a copy function, which could be used to pointlessly
803 copy a string over itself. The library routine is documented as not
804 supporting overlapping copies, and on MacOS it actually raised a SIGABRT.
806 JH/32 For main options check_spool_space and check_inode_space, where the
807 platform supports 64b integers, support more than the previous 2^31 kB
808 (i.e. more than 2 TB). Accept E, P and T multipliers in addition to
809 the previous G, M, k.
811 JH/33 Bug 2338: Fix the cyrus-sasl authenticator to fill in the
812 $authenticated_fail_id variable on authentication failure. Previously
815 JH/34 Increase RSA keysize of autogen selfsign cert from 1024 to 2048. RHEL 8.0
816 OpenSSL didn't want to use such a weak key. Do for GnuTLS also, and for
817 more-modern GnuTLS move from GNUTLS_SEC_PARAM_LOW to
818 GNUTLS_SEC_PARAM_MEDIUM.
820 JH/35 OpenSSL: fail the handshake when SNI processing hits a problem, server
821 side. Previously we would continue as if no SNI had been received.
823 JH/36 Harden the handling of string-lists. When a list consisted of a sole
824 "<" character, which should be a list-separator specification, we walked
825 off past the nul-terimation.
827 JH/37 Bug 2341: Send "message delayed" warning MDNs (restricted to external
828 causes) even when the retry time is not yet met. Previously they were
829 not, meaning that when (say) an account was over-quota and temp-rejecting,
830 and multiple senders' messages were queued, only one sender would get
831 notified on each configured delay_warning cycle.
833 JH/38 Bug 2351: Log failures to extract envelope addresses from message headers.
835 JH/39 OpenSSL: clear the error stack after an SSL_accept(). With anon-auth
836 cipher-suites, an error can be left on the stack even for a succeeding
837 accept; this results in impossible error messages when a later operation
840 AM/01 Bug 2359: GnuTLS: repeat lowlevel read and write operations while they
841 return error codes indicating retry. Under TLS1.3 this becomes required.
843 JH/40 Fix the feature-cache refresh for EXPERIMENTAL_PIPE_CONNECT. Previously
844 it only wrote the new authenticators, resulting in a lack of tracking of
845 peer changes of ESMTP extensions until the next cache flush.
847 JH/41 Fix the loop reading a message header line to check for integer overflow,
848 and more-often against header_maxsize. Previously a crafted message could
849 induce a crash of the recive process; now the message is cleanly rejected.
851 JH/42 Bug 2366: Fix the behaviour of the dkim_verify_signers option. It had
852 been totally disabled for all of 4.91. Discovery and fix by "Mad Alex".
858 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
859 When redis_servers is set to a list of > 1 element, and the Redis servers
860 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
861 case of MOVED into a DEFER case instead, thus moving the query onto the
862 next server in the list. For a cluster of N elements, all N servers must
863 be defined in redis_servers.
865 GF/02 Catch and remove uninitialized value warning in exiqsumm
866 Check for existence of @ARGV before looking at $ARGV[0]
868 JH/01 Replace the store_release() internal interface with store_newblock(),
869 which internalises the check required to safely use the old one, plus
870 the allocate and data copy operations duplicated in both (!) of the
871 extant use locations.
873 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
874 modifier. This matches the restriction on the commandline.
876 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
877 Previously only the last row was returned.
879 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
880 we assumed that tags in the header were well-formed, and parsed the
881 element content after inspecting only the first char of the tag.
882 Assumptions at that stage could crash the receive process on malformed
885 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
886 While running the DKIM ACL we operate on the Permanent memory pool so that
887 variables created with "set" persist to the DATA ACL. Also (at any time)
888 DNS lookups that fail create cache records using the Permanent pool. But
889 expansions release any allocations made on the current pool - so a dnsdb
890 lookup expansion done in the DKIM ACL releases the memory used for the
891 DNS negative-cache, and bad things result. Solution is to switch to the
892 Main pool for expansions.
893 While we're in that code, add checks on the DNS cache during store_reset,
894 active in the testsuite.
895 Problem spotted, and debugging aided, by Wolfgang Breyha.
897 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
898 When none of the hosts presented to a transport match an already-open
899 connection, close it and proceed with the list. Previously we would
900 queue the message. Spotted by Lena with Yahoo, probably involving
903 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
904 Previously a spurious "250 OK id=" response was appended to the proper
907 JH/08 The "support for" informational output now, which built with Content
908 Scanning support, has a line for the malware scanner interfaces compiled
909 in. Interface can be individually included or not at build time.
911 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
912 by the template makefile "src/EDITME". The "STREAM" support for an older
913 ClamAV interface method is removed.
915 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
916 rows affected is given instead).
918 JH/11 The runtime Berkeley DB library version is now additionally output by
919 "exim -d -bV". Previously only the compile-time version was shown.
921 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
922 SMTP connection. Previously, when one had more recipients than the
923 first, an abortive onward connection was made. Move to full support for
924 multiple onward connections in sequence, handling cutthrough connection
925 for all multi-message initiating connections.
927 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
928 routers. Previously, a multi-recipient message would fail to match the
929 onward-connection opened for the first recipient, and cause its closure.
931 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
932 a timeout on read on a GnuTLS initiating connection, resulting in the
933 initiating connection being dropped. This mattered most when the callout
934 was marked defer_ok. Fix to keep the two timeout-detection methods
937 JH/15 Relax results from ACL control request to enable cutthrough, in
938 unsupported situations, from error to silently (except under debug)
939 ignoring. This covers use with PRDR, frozen messages, queue-only and
942 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
944 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
945 metadata, resulting in a crash in free().
947 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
948 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
949 Broken also in d185889f4, with init system revamp.
951 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
952 Previously we abruptly closed the connection after reading a malware-
953 found indication; now we go on to read the "scan ok" response line,
956 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
957 ACL. Previously, a crash would result.
959 JH/19 Speed up macro lookups during configuration file read, by skipping non-
960 macro text after a replacement (previously it was only once per line) and
961 by skipping builtin macros when searching for an uppercase lead character.
963 JH/20 DANE support moved from Experimental to mainline. The Makefile control
964 for the build is renamed.
966 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
967 was allocated for every new TLS startup, meaning one per message. Fix
968 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
970 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
971 reported the original. Fix to report (as far as possible) the ACL
972 result replacing the original.
974 JH/23 Fix memory leak during multi-message connections using STARTTLS under
975 OpenSSL. Certificate information is loaded for every new TLS startup,
976 and the resources needed to be freed.
978 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
980 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
983 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
984 DATA response info to the (existing) per-recipient response info for
985 the "C=" log element. It can have useful tracking info from the
986 destination system. Patch from Simon Arlott.
988 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
989 length value. Previously this would segfault.
991 HS/02 Support Avast multiline protoocol, this allows passing flags to
992 newer versions of the scanner.
994 JH/28 Ensure that variables possibly set during message acceptance are marked
995 dead before release of memory in the daemon loop. This stops complaints
996 about them when the debug_store option is enabled. Discovered specifically
997 for sender_rate_period, but applies to a whole set of variables.
998 Do the same for the queue-runner and queue-list loops, for variables set
999 from spool message files. Do the same for the SMTP per-message loop, for
1000 certain variables indirectly set in ACL operations.
1002 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
1003 as a multi-recipient message from a mailinglist manager). The coding had
1004 an arbitrary cutoff number of characters while checking for more input;
1005 enforced by writing a NUL into the buffer. This corrupted long / fast
1006 input. The problem was exposed more widely when more pipelineing of SMTP
1007 responses was introduced, and one Exim system was feeding another.
1008 The symptom is log complaints of SMTP syntax error (NUL chars) on the
1009 receiving system, and refused recipients seen by the sending system
1010 (propating to people being dropped from mailing lists).
1011 Discovered and pinpointed by David Carter.
1013 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
1014 replaced by the ${authresults } expansion.
1016 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
1018 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
1019 allows proper process termination in container environments.
1021 JH/32 Bug 2258: Fix spool_wireformat in combination with LMTP transport.
1022 Previously the "final dot" had a newline after it; ensure it is CR,LF.
1024 JH/33 SPF: remove support for the "spf" ACL condition outcome values "err_temp"
1025 and "err_perm", deprecated since 4.83 when the RFC-defined words
1026 "temperror" and "permerror" were introduced.
1028 JH/34 Re-introduce enforcement of no cutthrough delivery on transports having
1029 transport-filters or DKIM-signing. The restriction was lost in the
1030 consolidation of verify-callout and delivery SMTP handling.
1031 Extend the restriction to also cover ARC-signing.
1033 JH/35 Cutthrough: for a final-dot response timeout (and nonunderstood responses)
1034 in defer=pass mode supply a 450 to the initiator. Previously the message
1037 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
1038 tls_require_ciphers is used as before.
1040 HS/03 Malware Avast: Better match the Avast multiline protocol. Add
1041 "pass_unscanned". Only tmpfails from the scanner are written to
1042 the paniclog, as they may require admin intervention (permission
1043 denied, license issues). Other scanner errors (like decompression
1044 bombs) do not cause a paniclog entry.
1046 JH/36 Fix reinitialisation of DKIM logging variable between messages.
1047 Previously it was possible to log spurious information in receive log
1050 JH/37 Bug 2255: Revert the disable of the OpenSSL session caching. This
1051 triggered odd behaviour from Outlook Express clients.
1053 PP/03 Add util/renew-opendmarc-tlds.sh script for safe renewal of public
1056 JH/38 DKIM: accept Ed25519 pubkeys in SubjectPublicKeyInfo-wrapped form,
1057 since the IETF WG has not yet settled on that versus the original
1058 "bare" representation.
1060 JH/39 Fix syslog logging for syslog_timestamp=no and log_selector +millisec.
1061 Previously the millisecond value corrupted the output.
1062 Fix also for syslog_pid=no and log_selector +pid, for which the pid
1063 corrupted the output.
1069 JH/01 Rework error string handling in TLS interface so that the caller in
1070 more cases is responsible for logging. This permits library-sourced
1071 string to be attached to addresses during delivery, and collapses
1072 pairs of long lines into single ones.
1074 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
1075 during configuration. Wildcards are allowed and expanded.
1077 JH/02 Rework error string handling in DKIM to pass more info back to callers.
1078 This permits better logging.
1080 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
1081 do not close it down and have the child transport start it up again on
1082 the passed-on TCP connection. Instead, proxy the child (and any
1083 subsequent ones) for TLS via a unix-domain socket channel. Logging is
1084 affected: the continued delivery log lines do not have any DNSSEC, TLS
1085 Certificate or OCSP information. TLS cipher information is still logged.
1087 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
1088 identical IP addresses on different listening ports. Will also affect
1091 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
1092 add noisy ifdef guards to special-case this sillyness.
1093 Patch from Bernd Kuhls.
1095 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
1096 than 255 are no longer allowed.
1098 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
1099 Disable the session-cache too, which might reduce our load. Since we
1100 currrectly use a new context for every connection, both as server and
1101 client, there is no benefit for these.
1102 GnuTLS appears to not support tickets server-side by default (we don't
1103 call gnutls_session_ticket_enable_server()) but client side is enabled
1104 by default on recent versions (3.1.3 +) unless the PFS priority string
1107 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
1108 <https://reproducible-builds.org/specs/source-date-epoch/>.
1110 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
1111 the check for any unsuccessful recipients did not notice the limit, and
1112 erroneously found still-pending ones.
1114 JH/08 Pipeline CHUNKING command and data together, on kernels that support
1115 MSG_MORE. Only in-clear (not on TLS connections).
1117 JH/09 Avoid using a temporary file during transport using dkim. Unless a
1118 transport-filter is involved we can buffer the headers in memory for
1119 creating the signature, and read the spool data file once for the
1120 signature and again for transmission.
1122 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
1123 4.77 as the kernel support then wasn't solid, having issues in 64bit
1124 mode. Now, it's been long enough. Add support for FreeBSD also.
1126 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
1127 case where the routing stage had gathered several addresses to send to
1128 a host before calling the transport for the first, we previously failed
1129 to close down TLS in the old transport process before passing the TCP
1130 connection to the new process. The new one sent a STARTTLS command
1131 which naturally failed, giving a failed delivery and bloating the retry
1132 database. Investigation and fix prototype from Wolfgang Breyha.
1134 JH/12 Fix check on SMTP command input synchronisation. Previously there were
1135 false-negatives in the check that the sender had not preempted a response
1136 or prompt from Exim (running as a server), due to that code's lack of
1137 awareness of the SMTP input buffering.
1139 PP/04 Add commandline_checks_require_admin option.
1140 Exim drops privileges sanely, various checks such as -be aren't a
1141 security problem, as long as you trust local users with access to their
1142 own account. When invoked by services which pass untrusted data to
1143 Exim, this might be an issue. Set this option in main configuration
1144 AND make fixes to the calling application, such as using `--` to stop
1147 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
1148 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
1149 responses to those, into a single TLS record each way (this usually means
1150 a single packet). As a side issue, smtp_enforce_sync now works on TLS
1153 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
1154 affects you only if you're dancing at the edge of the param size limits.
1155 If you are, and this message makes sense to you, then: raise the
1156 configured limit or use OpenSSL 1.1. Nothing we can do for older
1159 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
1160 cmdline element to get the documented default one. Previously it was
1163 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
1164 are unsuable for non-message-specific reasons - eg. connection timeout,
1165 banner-time rejection.
1167 JH/16 Fix logging of delivery remote port, when specified by router, under
1170 PP/06 Repair manualroute's ability to take options in any order, even if one
1171 is the name of a transport.
1174 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
1176 JH/17 Change the list-building routines interface to use the expanding-string
1177 triplet model, for better allocation and copying behaviour.
1179 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
1180 Previously it was constructed the first time a possibly-matching string
1181 was met in the configuration file input during startup; now it is done
1184 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
1185 compatible one, to avoid the (poorly documented) possibility of a config
1186 file in the working directory redirecting the DB files, possibly correpting
1187 some existing file. CVE-2017-10140 assigned for BDB.
1189 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
1190 cache-hot. Previously, although the result was properly cached, the
1191 initial verify call returned a defer.
1193 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
1194 the main verify for receipient in uncached-mode.
1196 JH/22 Retire historical build files to an "unsupported" subdir. These are
1197 defined as "ones for which we have no current evidence of testing".
1199 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
1200 if present. Previously it was ignored.
1202 JH/24 Start using specified-initialisers in C structure init coding. This is
1203 a C99 feature (it's 2017, so now considered safe).
1205 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
1206 if was a fixed-sized field and bitmask ops via macros; it is now more
1209 PP/07 GitHub PR 56: Apply MariaDB build fix.
1210 Patch provided by Jaroslav Škarvada.
1212 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
1213 during Coverity cleanups [4.87 JH/47]
1214 Diagnosis and fix provided by Michael Fischer v. Mollard.
1216 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
1217 the right size to place the terminating semicolon on its own folded
1218 line, the header hash was calculated to an incorrect value thanks to
1219 the (relaxed) space the fold became.
1221 HS/02 Fix Bug 2130: large writes from the transport subprocess were chunked
1222 and confused the parent.
1224 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
1225 which could crash as a result. This could lead to undeliverable messages.
1227 JH/28 Logging: "next input sent too soon" now shows where input was truncated
1230 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
1231 matters on fast-turnover and PID-randomising systems, which were getting
1232 out-of-order delivery.
1234 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
1235 a possibly-overlapping copy. The symptom was that "Remote host closed
1236 connection in response to HELO" was logged instead of the actual 4xx
1239 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
1240 Previously only that bufferd was discarded, resulting in SYMTP command
1243 JH/32 DKIM: when a message has multiple signatures matching an identity given
1244 in dkim_verify_signers, run the dkim acl once for each. Previously only
1245 one run was done. Bug 2189.
1247 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
1248 "panic the current process" to "deliberately defer". The panic log is
1249 still written with the problem list name; the mail and reject logs now
1250 get a temp-reject line for the message that was being handled, saying
1251 something like "domains check lookup or other defer". The SMTP 451
1252 message is still "Temporary local problem".
1254 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
1255 A crafted sequence of BDAT commands could result in in-use memory beeing
1256 freed. CVE-2017-16943.
1258 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
1259 from SMTP input. Previously it was always done; now only done for DATA
1260 and not BDAT commands. CVE-2017-16944.
1262 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
1263 to the message (such as an overlong header line). Previously this was
1264 not done and we did not exit BDAT mode. Followon from the previous item
1265 though a different problem.
1271 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
1272 than -2003 did; needs libidn2 in addition to libidn.
1274 JH/02 The path option on a pipe transport is now expanded before use.
1276 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
1277 Patch provided by "Björn", documentation fix added too.
1279 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
1280 missing a wire-to-host endian conversion.
1282 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
1283 close after a BDAT command line could be taken as a following command,
1284 giving a synch failure. Fix by only checking for synch immediately
1285 before acknowledging the chunk.
1287 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
1288 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
1289 macro. Patches provided by Josh Soref.
1291 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
1292 Previously we did not; the RFC seems ambiguous and VRFY is not listed
1293 by IANA as a service extension. However, John Klensin suggests that we
1296 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
1297 the dkim code may be unix-mode line endings rather than smtp wire-format
1298 CRLF, so prepend a CR to any bare LF.
1300 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
1301 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
1303 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
1304 the first were themselves being wrongly included in the feed into dkim
1305 processing; with most chunk sizes in use this resulted in an incorrect
1306 body hash calculated value.
1308 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
1309 DKIM signature block, for verification. Although advised against by
1310 standards it is specifically not ruled illegal.
1312 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
1314 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
1315 missing a body hash (the bh= tag).
1317 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
1318 It seems that HAProxy sends the Proxy Protocol information in clear and
1319 only then does a TLS startup, so do the same.
1321 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
1322 TCP connections (such as for Spamd) unless the daemon successfully set
1323 Fast Open mode on its listening sockets. This fixes breakage seen on
1324 too-old kernels or those not configured for Fast Open, at the cost of
1325 requiring both directions being enabled for TFO, and TFO never being used
1326 by non-daemon-related Exim processes.
1328 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
1329 endings, at least on the first header line. Try to canonify any that get
1330 past that check, despite the cost.
1332 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
1333 now limited to an arbitrary five deep, while parsing addresses with the
1334 strip_excess_angle_brackets option enabled.
1336 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
1337 instead leave the unprompted TLS handshake in socket buffer for the
1338 TLS library to consume.
1340 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
1342 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
1344 JH/16 Drop variables when they go out of scope. Memory management drops a whole
1345 region in one operation, for speed, and this leaves assigned pointers
1346 dangling. Add checks run only under the testsuite which checks all
1347 variables at a store-reset and panics on a dangling pointer; add code
1348 explicitly nulling out all the variables discovered. Fixes one known
1349 bug: a transport crash, where a dangling pointer for $sending_ip_address
1350 originally assigned in a verify callout, is re-used.
1352 PP/06 Drop '.' from @INC in various Perl scripts.
1354 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
1356 PP/08 Reduce a number of compilation warnings under clang; building with
1357 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
1358 should be warning-free.
1360 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
1362 HS/01 Fix portability problems introduced by PP/08 for platforms where
1363 realloc(NULL) is not equivalent to malloc() [SunOS et al].
1365 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
1366 chunk. This allows us to accept broken chunked messages. We need a more
1367 general solution here.
1369 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
1370 already-broken messages in the queue.
1372 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
1374 JH/19 Fix reference counting bug in routing-generated-address tracking.
1380 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
1381 supports it and a size is available (ie. the sending peer gave us one).
1383 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
1384 years of being deprecated). The replacements are the ACLs
1385 acl_smtp_mime and acl_not_smtp_mime.
1387 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
1388 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
1389 or even an in-clear connection were permitted. Now, if the host lookup
1390 was dnssec and dane was requested then the host is only used if the
1391 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
1392 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
1393 if one fails this test.
1394 This means that a poorly-configured remote DNS will make it incommunicado;
1395 but it protects against a DNS-interception attack on it.
1397 JH/04 Bug 1810: make continued-use of an open smtp transport connection
1398 non-noisy when a race steals the message being considered.
1400 JH/05 If main configuration option tls_certificate is unset, generate a
1401 self-signed certificate for inbound TLS connections.
1403 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
1404 in rewrites and routers.
1406 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
1407 and logged a warning sing 4.83; now they are a configuration file error.
1409 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
1410 (lacking @domain). Apply the same qualification processing as RCPT.
1412 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
1414 JH/10 Support ${sha256:} applied to a string (as well as the previous
1417 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
1418 a cutthrough deliver is pending, as we always want to make a connection.
1419 This also avoids re-routing the message when later placing the cutthrough
1420 connection after a verify cache hit.
1421 Do not update it with the verify result either.
1423 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
1424 when routing results in more than one destination address.
1426 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
1427 signing (which inhibits the cutthrough capability). Previously only
1428 the presence of an option was tested; now an expansion evaluating as
1429 empty is permissible (obviously it should depend only on data available
1430 when the cutthrough connection is made).
1432 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
1433 the relevant preceding SMTP command did not note the pipelining mode.
1435 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
1436 Previously they were not counted.
1438 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
1439 as one having no matching records. Previously we deferred the message
1440 that needed the lookup.
1442 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
1443 distinguished as "(=".
1445 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
1446 for missing MX records. Previously it only worked for missing A records.
1448 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
1450 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
1451 after the data-go-ahead and data-ack. Patch from Jason Betts.
1453 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
1454 even for a "none" policy. Patch from Tony Meyer.
1456 JH/22 Fix continued use of a connection for further deliveries. If a port was
1457 specified by a router, it must also match for the delivery to be
1460 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
1461 When one of the recipients of a message was unsuitable for the connection
1462 (has no matching addresses), we lost track of needing to mark it
1463 deferred. As a result mail would be lost.
1465 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
1467 JH/25 Decoding ACL controls is now done using a binary search; the source code
1468 takes up less space and should be simpler to maintain. Merge the ACL
1469 condition decode tables also, with similar effect.
1471 JH/26 Fix problem with one_time used on a redirect router which returned the
1472 parent address unchanged. A retry would see the parent address marked as
1473 delivered, so not attempt the (identical) child. As a result mail would
1476 JH/27 Fix a possible security hole, wherein a process operating with the Exim
1477 UID can gain a root shell. Credit to http://www.halfdog.net/ for
1478 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
1481 JH/28 Enable {spool,log} filesystem space and inode checks as default.
1482 Main config options check_{log,spool}_{inodes,space} are now
1483 100 inodes, 10MB unless set otherwise in the configuration.
1485 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
1486 Previously it only applied to the main-section connection policy
1489 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
1491 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
1492 by me. Added RFC7919 DH primes as an alternative.
1494 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
1495 are not in the system include path.
1497 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
1498 GnuTLS, when a session startup failed (eg because the client disconnected)
1499 Exim did stdio operations after fclose. This was exposed by a recent
1500 change which nulled out the file handle after the fclose.
1502 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
1503 signed directly by the cert-signing cert, rather than an intermediate
1504 OCSP-signing cert. This is the model used by LetsEncrypt.
1506 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
1508 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
1509 an incoming connection.
1511 HS/02 Bug 1802: Do not half-close the connection after sending a request
1514 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
1515 fallback to "prime256v1".
1517 JH/34 SECURITY: Use proper copy of DATA command in error message.
1518 Could leak key material. Remotely exploitable. CVE-2016-9963.
1524 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
1525 and 3.4.4 - once the server is enabled to respond to an OCSP request
1526 it does even when not requested, resulting in a stapling non-aware
1527 client dropping the TLS connection.
1529 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
1530 support variable-length bit vectors. No functional change.
1532 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
1533 The I= interface field on outgoing lines is now after the H= remote
1534 host field, same as incoming lines. There is a separate
1535 outgoing_interface log selector which allows you to disable the
1538 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
1539 If not running log_selector +smtp_connection the mainlog would be held
1540 open indefinitely after a "too many connections" event, including to a
1541 deleted file after a log rotate. Leave the per net connection logging
1542 leaving it open for efficiency as that will be quickly detected by the
1543 check on the next write.
1545 HS/01 Bug 1671: Fix post transport crash.
1546 Processing the wait-<transport> messages could crash the delivery
1547 process if the message IDs didn't exist for some reason. When
1548 using 'split_spool_directory=yes' the construction of the spool
1549 file name failed already, exposing the same netto behaviour.
1551 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
1552 mime_regex ACL conditions.
1554 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
1555 to DSN fail messages (bounces): remote IP, remote greeting, remote response
1556 to HELO, local diagnostic string.
1558 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
1559 log line to debug. Even when configured with a tls authenticator many
1560 client connections are expected to not authenticate in this way, so
1561 an authenticate fail is not an error.
1563 HS/02 Add the Exim version string to the process info. This way exiwhat
1564 gives some more detail about the running daemon.
1566 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
1567 matter for fast-change records such as DNSBLs.
1569 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
1570 retry record, even if constant. There may be multiple transports with
1571 different interface settings and the retry behaviour needs to be kept
1574 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
1576 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
1578 JH/10 Bug 840: fix log_defer_output option of pipe transport
1580 JH/11 Bug 830: use same host for all RCPTS of a message, even under
1581 hosts_randomize. This matters a lot when combined with mua_wrapper.
1583 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
1584 ${quote_pgsql:<string>} operator.
1586 JH/13 Bug 1708: avoid misaligned access in cached lookup.
1588 JH/14 Change header file name for freeradius-client. Relevant if compiling
1589 with Radius support; from the Gentoo tree and checked under Fedora.
1591 JH/15 Bug 1712: Introduce $prdr_requested flag variable
1593 JH/16 Bug 1714: Permit an empty string as expansion result for transport
1594 option transport_filter, meaning no filtering.
1596 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
1598 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
1599 defaults to "*" (all hosts). The variable is now available when not built
1600 with TLS, default unset, mainly to enable keeping the testsuite sane.
1601 If a server certificate is not supplied (via tls_certificate) an error is
1602 logged, and clients will find TLS connections fail on startup. Presumably
1603 they will retry in-clear.
1604 Packagers of Exim are strongly encouraged to create a server certificate
1605 at installation time.
1607 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
1608 with the $config_file variable.
1610 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
1611 in transport context, after the attempt, and per-recipient. The latter type
1612 is per host attempted. The event data is the error message, and the errno
1613 information encodes the lookup type (A vs. MX) used for the (first) host,
1614 and the trailing two digits of the smtp 4xx response.
1616 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
1617 to write to mainlog (or rejectlog, paniclog) in the window between file
1618 creation and permissions/ownership being changed. Particularly affects
1619 installations where exicyclog is run as root, rather than exim user;
1620 result is that the running daemon panics and dies.
1622 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
1624 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
1625 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
1626 "pri" and "weight". Note that the previous implicit priority given by the
1627 list order is no longer honoured.
1629 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
1630 for DKIM processing.
1632 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
1633 by defining SUPPORT_SOCKS.
1635 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
1636 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
1637 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
1638 variables are renamed to proxy_{local,external}_{address,port}.
1640 JH/27 Move Internationalisation support from Experimental to mainline, enabled
1641 for a build by defining SUPPORT_I18N
1643 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
1644 of the query string, and make ${quote_redis:} do that quoting.
1646 JH/29 Move Events support from Experimental to mainline, enabled by default
1647 and removable for a build by defining DISABLE_EVENT.
1649 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
1651 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
1652 cached by the daemon.
1654 JH/32 Move Redis support from Experimental to mainline, enabled for a build
1655 by defining LOOKUP_REDIS. The libhiredis library is required.
1657 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
1658 keys are given for lookup.
1660 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
1661 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
1662 only supported when built with TLS support. The PolarSSL SHA routines
1663 are still used when the TLS library is too old for convenient support.
1665 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
1666 openssl_options), for security. OpenSSL forces this from version 1.1.0
1667 server-side so match that on older versions.
1669 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
1670 allocation for $value could be released as the expansion processing
1671 concluded, but leaving the global pointer active for it.
1673 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
1674 and to use the domains and local_parts ACL conditions.
1676 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
1677 incorrectly not doubled on cutthrough transmission, hence seen as a
1678 body-termination at the receiving system - resulting in truncated mails.
1679 Commonly the sender saw a TCP-level error, and retransmitted the message
1680 via the normal store-and-forward channel. This could result in duplicates
1681 received - but deduplicating mailstores were liable to retain only the
1682 initial truncated version.
1684 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
1686 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
1688 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
1689 we're in there, support oversigning also; bug 1309.
1691 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
1693 HS/04 Add support for keep_environment and add_environment options.
1695 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
1696 either intentional arithmetic overflow during PRNG, or testing config-
1699 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
1700 delivery resulted in actual delivery. Cancel cutthrough before DATA
1703 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
1704 rejects a recipient: pass the reject to the originator.
1706 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
1707 Many were false-positives and ignorable, but it's worth fixing the
1710 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
1711 for the new environment-manipulation done at startup. Move the routines
1712 from being local to tls.c to being global via the os.c file.
1714 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
1715 an extract embedded as result-arg for a map, the first arg for extract
1716 is unavailable so we cannot tell if this is a numbered or keyed
1717 extraction. Accept either.
1723 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
1726 JH/02 The smtp transport option "multi_domain" is now expanded.
1728 JH/03 The smtp transport now requests PRDR by default, if the server offers
1731 JH/04 Certificate name checking on server certificates, when exim is a client,
1732 is now done by default. The transport option tls_verify_cert_hostnames
1733 can be used to disable this per-host. The build option
1734 EXPERIMENTAL_CERTNAMES is withdrawn.
1736 JH/05 The value of the tls_verify_certificates smtp transport and main options
1737 default to the word "system" to access the system default CA bundle.
1738 For GnuTLS, only version 3.0.20 or later.
1740 JH/06 Verification of the server certificate for a TLS connection is now tried
1741 (but not required) by default. The verification status is now logged by
1742 default, for both outbound TLS and client-certificate supplying inbound
1745 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
1748 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
1749 Status Notification (bounce) messages are now MIME format per RFC 3464.
1750 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
1751 under the control of the dsn_advertise_hosts option, and routers may
1752 have a dsn_lasthop option.
1754 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
1755 default, modifiable by a malware= option. The list separator for
1756 the options can now be changed in the usual way. Bug 68.
1758 JH/10 The smtp_receive_timeout main option is now expanded before use.
1760 JH/11 The incoming_interface log option now also enables logging of the
1761 local interface on delivery outgoing connections.
1763 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
1764 if the interface and destination host and port all match.
1766 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
1769 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
1770 Patch from Andrew Lewis.
1772 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
1773 now supports optional time-restrictions, weighting, and priority
1774 modifiers per server. Patch originally by <rommer@active.by>.
1776 JH/16 The spamd_address main option now supports a mixed list of local
1777 and remote servers. Remote servers can be IPv6 addresses, and
1778 specify a port-range.
1780 JH/17 Bug 68: The spamd_address main option now supports an optional
1781 timeout value per server.
1783 JH/18 Bug 1581: Router and transport options headers_add/remove can
1784 now have the list separator specified.
1786 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
1789 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
1792 JH/21 Support for the A6 type of dns record is withdrawn.
1794 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
1795 rather than the verbs used.
1797 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
1798 from 255 to 1024 chars.
1800 JH/24 Verification callouts now attempt to use TLS by default.
1802 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
1803 are generic router options now. The defaults didn't change.
1805 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
1806 Original patch from Alexander Shikoff, worked over by JH.
1808 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
1809 files if ZCAT_COMMAND is not executable.
1811 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
1813 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
1815 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
1816 Normally benign, it bites when the pair was led to by a CNAME;
1817 modern usage is to not canonicalize the domain to a CNAME target
1818 (and we were inconsistent anyway for A-only vs AAAA+A).
1820 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
1822 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
1823 when evaluating $sender_host_dnssec.
1825 JH/31 Check the HELO verification lookup for DNSSEC, adding new
1826 $sender_helo_dnssec variable.
1828 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
1830 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
1832 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
1834 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
1835 documented as working, but never had. Support all but $spam_report.
1837 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
1838 added for tls authenticator.
1840 HS/03 Add perl_taintmode main config option
1846 TL/01 When running the test suite, the README says that variables such as
1847 no_msglog_check are global and can be placed anywhere in a specific
1848 test's script, however it was observed that placement needed to be near
1849 the beginning for it to behave that way. Changed the runtest perl
1850 script to read through the entire script once to detect and set these
1851 variables, reset to the beginning of the script, and then run through
1852 the script parsing/test process like normal.
1854 TL/02 The BSD's have an arc4random API. One of the functions to induce
1855 adding randomness was arc4random_stir(), but it has been removed in
1856 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
1857 function when detected.
1859 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
1860 cause callback expansion.
1862 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
1863 syntax errors in an expansion can be treated as a string instead of
1864 logging or causing an error, due to the internal use of bool_lax
1865 instead of bool when processing it.
1867 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
1868 server certificates when making smtp deliveries.
1870 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
1872 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
1874 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1876 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1877 Merged patch from Sebastian Wiedenroth.
1879 JH/05 Fix results-pipe from transport process. Several recipients, combined
1880 with certificate use, exposed issues where response data items split
1881 over buffer boundaries were not parsed properly. This eventually
1882 resulted in duplicates being sent. This issue only became common enough
1883 to notice due to the introduction of connection certificate information,
1884 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1886 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1887 size buffer was used, resulting in syntax errors when an expansion
1890 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1891 version 3.3.6 or later.
1893 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1894 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1895 both become "event_action", the variables become $event_name, $event_data
1896 and $event_defer_errno. There is a new variable $verify_mode, usable in
1897 routers, transports and related events. The tls:cert event is now also
1898 raised for inbound connections, if the main configuration event_action
1901 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1902 early OCSP support, but no stapling (appears to be less than 1.0.0).
1904 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1905 server certificate names available under the smtp transport option
1906 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1909 JH/10 Time-related extraction expansions from certificates now use the main
1910 option "timezone" setting for output formatting, and are consistent
1911 between OpenSSL and GnuTLS compilations. Bug 1541.
1913 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1914 encoded parameter in the incoming message. Bug 1558.
1916 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1917 include certificate info, eximon was claiming there were spoolfile
1920 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1922 JH/14 Log delivery-related information more consistently, using the sequence
1923 "H=<name> [<ip>]" wherever possible.
1925 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1926 are problematic for Debian distribution, omit them from the release
1929 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1931 JH/16 Fix string representation of time values on 64bit time_t architectures.
1934 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1935 output list separator was used.
1940 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1941 checkers that were complaining about end of non-void function with no
1944 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1945 This was a regression introduced in 4.83 by another bugfix.
1947 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1949 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1950 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1956 TF/01 Correctly close the server side of TLS when forking for delivery.
1958 When a message was received over SMTP with TLS, Exim failed to clear up
1959 the incoming connection properly after forking off the child process to
1960 deliver the message. In some situations the subsequent outgoing
1961 delivery connection happened to have the same fd number as the incoming
1962 connection previously had. Exim would try to use TLS and fail, logging
1963 a "Bad file descriptor" error.
1965 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1966 utilities have not been installed.
1968 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1969 temporary space as the ACL may create new global variables.
1971 TL/01 LDAP support uses per connection or global context settings, depending
1972 upon the detected version of the libraries at build time.
1974 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1975 to extract and use the src ip:port in logging and expansions as if it
1976 were a direct connection from the outside internet. PPv2 support was
1977 updated based on HAProxy spec change in May 2014.
1979 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1981 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1982 Properly escape header and check for NULL return.
1984 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1987 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1989 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1990 characters in header names, implemented as a verify condition.
1991 Contributed by Michael Fischer v. Mollard.
1993 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1994 results permerror and temperror. Previous values are deprecated but
1995 still accepted. In a future release, err_perm and err_temp will be
1996 completely removed, which will be a backward incompatibility if the
1997 ACL tests for either of these two old results. Patch contributed by
1998 user bes-internal on the mailing list.
2000 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
2002 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
2003 selectors, in both main and reject logs.
2005 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
2008 JH/07 Add malware type "sock" for talking to simple daemon.
2010 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
2012 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
2013 routers/transports under cutthrough routing.
2015 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
2016 numbers. Touch up "bool" conditional to keep the same definition.
2018 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
2020 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
2022 JH/12 Expand items in router/transport headers_add or headers_remove lists
2023 individually rather than the list as a whole. Bug 1452.
2025 Required for reasonable handling of multiple headers_ options when
2026 they may be empty; requires that headers_remove items with embedded
2027 colons must have them doubled (or the list-separator changed).
2029 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
2030 view the policy declared in the DMARC record. Currently, $dmarc_status
2031 is a combined value of both the record presence and the result of the
2034 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
2036 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
2037 dnslookup router and the smtp transport (applying to the forward
2040 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
2041 of ldap servers used for a specific lookup. Patch provided by Heiko
2044 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
2045 New variable $lookup_dnssec_authenticated for observability.
2047 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
2048 Patch submitted by Lars Timman.
2050 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
2052 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
2053 Requires trusted mode and valid format message id, aborts otherwise.
2054 Patch contributed by Heiko Schlichting.
2056 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
2057 certextract with support for various fields. Bug 1358.
2059 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
2060 is requested by default, modifiable by smtp transport option
2063 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
2064 operate on certificate variables to give certificate fingerprints
2065 Also new ${sha256:cert_variable}.
2067 JH/23 The PRDR feature is moved from being Experimental into the mainline.
2069 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
2070 Christian Aistleitner.
2072 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
2074 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
2075 file. Patch from Wolfgang Breyha.
2077 JH/25 Expand the coverage of the delivery $host and $host_address to
2078 client authenticators run in verify callout. Bug 1476.
2080 JH/26 Port service names are now accepted for tls_on_connect_ports, to
2081 align with daemon_smtp_ports. Bug 72.
2083 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
2084 support and error reporting did not work properly.
2086 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
2087 and is readable. Patch from Andrew Colin Kissa.
2089 TL/14 Enhance documentation of ${run expansion and how it parses the
2090 commandline after expansion, particularly in the case when an
2091 unquoted variable expansion results in an empty value.
2093 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
2095 PP/02 Fix internal collision of T_APL on systems which support RFC3123
2096 by renaming away from it. Addresses GH issue 15, reported by
2099 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
2101 TL/15 SECURITY: prevent double expansion in math comparison functions
2102 (can expand unsanitized data). Not remotely exploitable.
2109 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
2111 PP/02 Make -n do something, by making it not do something.
2112 When combined with -bP, the name of an option is not output.
2114 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
2117 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
2118 $sender_host_name and config options to manage this, and basic check
2121 PP/05 DSCP support for outbound connections and control modifier for inbound.
2123 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
2124 (Only plugin which currently uses this is kerberos4, which nobody should
2125 be using, but we should make it available and other future plugins might
2126 conceivably use it, even though it would break NAT; stuff *should* be
2127 using channel bindings instead).
2129 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
2130 name; added for Sendmail compatibility; requires admin caller.
2131 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
2132 just ignore it); requires trusted caller.
2133 Also parse but ignore: -Ac -Am -X<logfile>
2136 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
2138 TL/02 Add +smtp_confirmation as a default logging option.
2140 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
2141 Patch by Magnus Holmgren from 2007-02-20.
2143 TL/04 Bugzilla 1281 - Spec typo.
2144 Bugzilla 1283 - Spec typo.
2145 Bugzilla 1290 - Spec grammar fixes.
2147 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
2149 TL/06 Add Experimental DMARC support using libopendmarc libraries.
2151 TL/07 Fix an out of order global option causing a segfault. Reported to dev
2152 mailing list by by Dmitry Isaikin.
2154 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
2156 JH/02 Support "G" suffix to numbers in ${if comparisons.
2158 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
2160 NM/01 Bugzilla 1197 - Spec typo
2161 Bugzilla 1196 - Spec examples corrections
2163 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
2165 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
2166 gnutls_enable_pkcs11, but renamed to more accurately indicate its
2169 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
2170 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
2172 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
2173 "acl {{name}{arg}...}", and optional args on acl condition
2176 JH/05 Permit multiple router/transport headers_add/remove lines.
2178 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
2180 JH/07 Avoid using a waiting database for a single-message-only transport.
2181 Performance patch from Paul Fisher. Bugzilla 1262.
2183 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
2186 JH/09 Add $headers_added variable, with content from use of ACL modifier
2187 add_header (but not yet added to the message). Bugzilla 199.
2189 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
2190 Pulled from Bugzilla 817 by Wolfgang Breyha.
2192 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
2194 (nb: this is the same fix as in Exim 4.80.1)
2196 JH/11 Add A= logging on delivery lines, and a client_set_id option on
2199 JH/12 Add optional authenticated_sender logging to A= and a log_selector
2202 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
2204 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
2205 advertise SMTP AUTH mechanism to us, instead of a generic
2206 protocol violation error. Also, make Exim more robust to bad
2207 data from the Dovecot auth socket.
2209 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
2211 When a queue runner is handling a message, Exim first routes the
2212 recipient addresses, during which it prunes them based on the retry
2213 hints database. After that it attempts to deliver the message to
2214 any remaining recipients. It then updates the hints database using
2217 So if a recipient address works intermittently, it can get repeatedly
2218 deferred at routing time. The retry hints record remains fresh so the
2219 address never reaches the final cutoff time.
2221 This is a fairly common occurrence when a user is bumping up against
2222 their storage quota. Exim had some logic in its local delivery code
2223 to deal with this. However it did not apply to per-recipient defers
2224 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
2226 This change adds a proper retry rule check during routing so that the
2227 final cutoff time is checked against the message's age. We only do
2228 this check if there is an address retry record and there is not a
2229 domain retry record; this implies that previous attempts to handle
2230 the address had the retry_use_local_parts option turned on. We use
2231 this as an approximation for the destination being like a local
2232 delivery, as in LMTP.
2234 I suspect this new check makes the old local delivery cutoff check
2235 redundant, but I have not verified this so I left the code in place.
2237 TF/02 Correct gecos expansion when From: is a prefix of the username.
2239 Test 0254 submits a message to Exim with the header
2243 When I ran the test suite under the user fanf2, Exim expanded
2244 the header to contain my full name, whereas it should have added
2245 a Resent-Sender: header. It erroneously treats any prefix of the
2246 username as equal to the username.
2248 This change corrects that bug.
2250 GF/01 DCC debug and logging tidyup
2251 Error conditions log to paniclog rather than rejectlog.
2252 Debug lines prefixed by "DCC: " to remove any ambiguity.
2254 TF/03 Avoid unnecessary rebuilds of lookup-related code.
2256 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
2257 Bug spotted by Jeremy Harris; was flawed since initial commit.
2258 Would have resulted in OCSP responses post-SNI triggering an Exim
2259 NULL dereference and crash.
2261 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
2263 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
2264 Bug detection, analysis and fix by Samuel Thibault.
2265 Bugzilla 1331, Debian bug #698092.
2267 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
2269 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
2270 Server implementation by Todd Lyons, client by JH.
2271 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
2272 config variable "prdr_enable" controls whether the server
2273 advertises the facility. If the client requests PRDR a new
2274 acl_data_smtp_prdr ACL is called once for each recipient, after
2275 the body content is received and before the acl_smtp_data ACL.
2276 The client is controlled by both of: a hosts_try_prdr option
2277 on the smtp transport, and the server advertisement.
2278 Default client logging of deliveries and rejections involving
2279 PRDR are flagged with the string "PRDR".
2281 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
2282 fclose(). Diagnosis by Todd Lyons.
2284 PP/17 Update configure.default to handle IPv6 localhost better.
2285 Patch by Alain Williams (plus minor tweaks).
2288 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
2289 This is now consistent with GnuTLS, and is now documented: the
2290 previous undocumented portable approach to treating the option as
2291 unset was to force an expansion failure. That still works, and
2292 an empty string is now equivalent.
2294 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
2295 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
2296 not performing validation itself.
2298 PP/20 Added force_command boolean option to pipe transport.
2299 Patch from Nick Koston, of cPanel Inc.
2301 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
2304 TF/04 Added udpsend ACL modifier and hexquote expansion operator
2306 PP/21 Fix eximon continuous updating with timestamped log-files.
2307 Broken in a format-string cleanup in 4.80, missed when I repaired the
2308 other false fix of the same issue.
2309 Report and fix from Heiko Schlichting.
2312 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
2313 Report from Prashanth Katuri.
2315 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
2316 It's SecureTransport, so affects any MacOS clients which use the
2317 system-integrated TLS libraries, including email clients.
2319 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
2320 using a MIME ACL for non-SMTP local injection.
2321 Report and assistance in diagnosis by Warren Baker.
2323 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
2325 JH/16 Fix comparisons for 64b. Bugzilla 1385.
2327 TL/09 Add expansion variable $authenticated_fail_id to keep track of
2328 last id that failed so it may be referenced in subsequent ACL's.
2330 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
2333 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
2334 ldap library initialization, allowing self-signed CA's to be
2335 used. Also properly sets require_cert option later in code by
2336 using NULL (global ldap config) instead of ldap handle (per
2337 session). Bug diagnosis and testing by alxgomz.
2339 TL/12 Enhanced documentation in the ratelimit.pl script provided in
2340 the src/util/ subdirectory.
2342 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
2343 renamed to Transport Post Delivery Action by Jeremy Harris, as
2346 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
2347 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
2348 redis_servers = needs to be configured which will be used by the redis
2349 lookup. Patch from Warren Baker, of The Packet Hub.
2351 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
2353 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
2354 hostname or reverse DNS when processing a host list. Used suggestions
2355 from multiple comments on this bug.
2357 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
2359 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
2360 Missed a few lines, added it to make the runtest require no keyboard
2363 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
2364 contains upper case chars. Make router use caseful_local_part.
2366 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
2367 support when GnuTLS has been built with p11-kit.
2373 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
2375 This, or similar/improved, will also be change PP/11 of 4.82.
2381 PP/01 Handle short writes when writing local log-files.
2382 In practice, only affects FreeBSD (8 onwards).
2383 Bugzilla 1053, with thanks to Dmitry Isaikin.
2385 NM/01 Bugzilla 949 - Documentation tweak
2387 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
2390 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
2392 PP/02 Implemented gsasl authenticator.
2394 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
2396 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
2397 `pkg-config foo` for cflags/libs.
2399 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
2400 with rest of GSASL and with heimdal_gssapi.
2402 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
2403 `pkg-config foo` for cflags/libs for the TLS implementation.
2405 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
2406 properties get this fed in as external SSF. A number of robustness
2407 and debugging improvements to the cyrus_sasl authenticator.
2409 PP/08 cyrus_sasl server now expands the server_realm option.
2411 PP/09 Bugzilla 1214 - Log authentication information in reject log.
2412 Patch by Jeremy Harris.
2414 PP/10 Added dbmjz lookup type.
2416 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
2418 PP/12 MAIL args handles TAB as well as SP, for better interop with
2419 non-compliant senders.
2420 Analysis and variant patch by Todd Lyons.
2422 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
2423 Bug report from Lars Müller <lars@samba.org> (via SUSE),
2424 Patch from Dirk Mueller <dmueller@suse.com>
2426 PP/13 tls_peerdn now print-escaped for spool files.
2427 Observed some $tls_peerdn in wild which contained \n, which resulted
2428 in spool file corruption.
2430 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
2431 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
2432 or write after TLS renegotiation, which otherwise led to messages
2435 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
2436 as a tracking header (ie: a signed header comes before the signature).
2437 Patch from Wolfgang Breyha.
2439 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
2440 comma-sep list; embedded commas doubled.
2442 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
2444 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
2446 Report and patch from Dmitry Banschikov.
2448 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
2449 Removed SSL_clear() after SSL_new() which led to protocol negotiation
2450 failures. We appear to now support TLS1.1+ with Exim.
2452 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
2453 lets Exim select keys and certificates based upon TLS SNI from client.
2454 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
2455 before an outbound SMTP session. New log_selector, +tls_sni.
2457 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
2458 NULL dereference. Report and patch from Alun Jones.
2460 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
2461 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
2462 Not seeing resolver debug output on NetBSD, but suspect this is a
2463 resolver implementation change.
2465 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
2466 Left warnings. Added "eximon gdb" invocation mode.
2468 PP/21 Defaulting "accept_8bitmime" to true, not false.
2470 PP/22 Added -bw for inetd wait mode support.
2472 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
2473 locate the relevant includes and libraries. Made this the default.
2475 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
2476 Bugzilla 1246, report and most of solution from Tomasz Kusy.
2478 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
2479 This may cause build issues on older platforms.
2481 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
2482 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
2483 gnutls_require_mac & gnutls_require_protocols (no longer supported).
2484 Added SNI support via GnuTLS too.
2485 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
2487 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
2489 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
2490 Applied second patch from Janne, implementing suggestion to default
2491 multiple-strings-in-record handling to match SPF spec.
2493 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
2495 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
2496 read-only, out of scope).
2497 Patch from Wolfgang Breyha, report from Stuart Northfield.
2499 PP/29 Fix three issues highlighted by clang analyser static analysis.
2500 Only crash-plausible issue would require the Cambridge-specific
2501 iplookup router and a misconfiguration.
2502 Report from Marcin Mirosław.
2504 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
2506 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
2507 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
2508 As part of this, removing so much warning spew let me fix some minor
2509 real issues in debug logging.
2511 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
2512 assignment on my part. Fixed.
2514 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
2515 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
2516 Janne Snabb (who went above and beyond: thank you).
2518 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
2519 string otherwise requires a connection and a bunch more work and it's
2520 relatively easy to get wrong. Should also expose TLS library linkage
2523 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
2524 64-bit ${eval} (JH/03).
2526 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
2527 GNU libc to support some of the 64-bit stuff, should not lead to
2528 conflicts. Defined before os.h is pulled in, so if a given platform
2529 needs to override this, it can.
2531 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
2532 protection layer was required, which is not implemented.
2533 Bugzilla 1254, patch from Wolfgang Breyha.
2535 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
2536 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
2537 tls_dhparam take prime identifiers. Also unbreak combination of
2538 OpenSSL+DH_params+TLSSNI.
2540 PP/39 Disable SSLv2 by default in OpenSSL support.
2546 PP/01 Solaris build fix for Oracle's LDAP libraries.
2547 Bugzilla 1109, patch from Stephen Usher.
2549 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
2551 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
2554 TF/02 Fix a couple more cases where we did not log the error message
2555 when unlink() failed. See also change 4.74-TF/03.
2557 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
2558 lock up or crash if it happened to be inside a call to libc when it
2559 got a SIGUSR1 from exiwhat.
2561 The SIGUSR1 handler appends the current process status to the process
2562 log which is later printed by exiwhat. It used to use the general
2563 purpose logging code to do this, but several functions it calls are
2564 not safe for signals.
2566 The new output code in the SIGUSR1 handler is specific to the process
2567 log, and simple enough that it's easy to inspect for signal safety.
2568 Removing some special cases also simplifies the general logging code.
2569 Removing the spurious timestamps from the process log simplifies
2572 TF/04 Improved ratelimit ACL condition.
2574 The /noupdate option has been deprecated in favour of /readonly which
2575 has clearer semantics. The /leaky, /strict, and /readonly update modes
2576 are mutually exclusive. The update mode is no longer included in the
2577 database key; it just determines when the database is updated. (This
2578 means that when you upgrade Exim will forget old rate measurements.)
2580 Exim now checks that the per_* options are used with an update mode that
2581 makes sense for the current ACL. For example, when Exim is processing a
2582 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
2583 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
2584 must specify per_mail/readonly. If you omit the update mode it defaults to
2585 /leaky where that makes sense (as before) or /readonly where required.
2587 The /noupdate option is now undocumented but still supported for
2588 backwards compatibility. It is equivalent to /readonly except that in
2589 ACLs where /readonly is required you may specify /leaky/noupdate or
2590 /strict/noupdate which are treated the same as /readonly.
2592 A useful new feature is the /count= option. This is a generalization
2593 of the per_byte option, so that you can measure the throughput of other
2594 aggregate values. For example, the per_byte option is now equivalent
2595 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
2597 The per_rcpt option has been generalized using the /count= mechanism
2598 (though it's more complicated than the per_byte equivalence). When it is
2599 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
2600 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
2601 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
2602 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
2603 Note that using per_rcpt with a non-readonly update mode in more than
2604 one ACL will cause the recipients to be double-counted. (The per_mail
2605 and per_byte options don't have this problem.)
2607 The handling of very low rates has changed slightly. If the computed rate
2608 is less than the event's count (usually one) then this event is the first
2609 after a long gap. In this case the rate is set to the same as this event's
2610 count, so that the first message of a spam run is counted properly.
2612 The major new feature is a mechanism for counting the rate of unique
2613 events. The new per_addr option counts the number of different
2614 recipients that someone has sent messages to in the last time period. It
2615 behaves like per_rcpt if all the recipient addresses are different, but
2616 duplicate recipient addresses do not increase the measured rate. Like
2617 the /count= option this is a general mechanism, so the per_addr option
2618 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
2619 example, measure the rate that a client uses different sender addresses
2620 with the options per_mail/unique=$sender_address. There are further
2621 details in the main documentation.
2623 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
2625 TF/06 Removed a few PCRE remnants.
2627 TF/07 Automatically extract Exim's version number from tags in the git
2628 repository when doing development or release builds.
2630 PP/02 Raise smtp_cmd_buffer_size to 16kB.
2631 Bugzilla 879. Patch from Paul Fisher.
2633 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
2634 Heavily based on revision 40f9a89a from Simon Arlott's tree.
2637 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
2639 PP/05 Variable $av_failed, true if the AV scanner deferred.
2640 Bugzilla 1078. Patch from John Horne.
2642 PP/06 Stop make process more reliably on build failure.
2643 Bugzilla 1087. Patch from Heiko Schlittermann.
2645 PP/07 Make maildir_use_size_file an _expandable_ boolean.
2646 Bugzilla 1089. Patch from Heiko Schlittermann.
2648 PP/08 Handle ${run} returning more data than OS pipe buffer size.
2649 Bugzilla 1131. Patch from Holger Weiß.
2651 PP/09 Handle IPv6 addresses with SPF.
2652 Bugzilla 860. Patch from Wolfgang Breyha.
2654 PP/10 GnuTLS: support TLS 1.2 & 1.1.
2656 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
2659 PP/11 match_* no longer expand right-hand-side by default.
2660 New compile-time build option, EXPAND_LISTMATCH_RHS.
2661 New expansion conditions, "inlist", "inlisti".
2663 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
2665 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
2667 PP/14 fix log_write() format string regression from TF/03.
2668 Bugzilla 1152. Patch from Dmitry Isaikin.
2674 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
2676 PP/02 Harmonised TLS library version reporting; only show if debugging.
2677 Layout now matches that introduced for other libraries in 4.74 PP/03.
2679 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
2681 PP/04 New "dns_use_edns0" global option.
2683 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
2686 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
2687 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
2689 TK/01 Updated PolarSSL code to 0.14.2.
2690 Bugzilla 1097. Patch from Andreas Metzler.
2692 PP/07 Catch divide-by-zero in ${eval:...}.
2693 Fixes bugzilla 1102.
2695 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
2698 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
2699 format-string attack -- SECURITY: remote arbitrary code execution.
2701 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
2702 time unintentionally subject to list matching rules, letting the header
2703 cause arbitrary Exim lookups (of items which can occur in lists, *not*
2704 arbitrary string expansion). This allowed for information disclosure.
2706 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
2707 INT_MIN/-1 -- value coerced to INT_MAX.
2713 NM/01 Workaround for PCRE version dependency in version reporting
2716 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
2717 This fixes portability to compilers other than gcc, notably
2718 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
2720 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
2721 makefiles for portability to HP-UX and POSIX correctness.
2723 PP/01 Permit LOOKUP_foo enabling on the make command-line.
2724 Also via indented variable definition in the Makefile.
2725 (Debugging by Oliver Heesakkers).
2727 PP/02 Restore caching of spamd results with expanded spamd_address.
2728 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
2730 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
2731 Improves build reliability. Fix from: Frank Elsner
2733 NM/02 Fix wide character breakage in the rfc2047 coding
2734 Fixes bug 1064. Patch from Andrey N. Oktyabrski
2736 NM/03 Allow underscore in dnslist lookups
2737 Fixes bug 1026. Patch from Graeme Fowler
2739 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
2740 Code patches from Adam Ciarcinski of NetBSD.
2742 NM/04 Fixed exiqgrep to cope with mailq missing size issue
2745 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
2746 is logged, to avoid truncation. Patch from John Horne.
2748 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
2749 Patch from Jakob Hirsch.
2751 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
2752 SQL string expansion failure details.
2753 Patch from Andrey Oktyabrski.
2755 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
2756 Patch from Simon Arlott.
2758 PP/09 New lookups functionality failed to compile on old gcc which rejects
2759 extern declarations in function scope.
2760 Patch from Oliver Fleischmann
2762 PP/10 Use sig_atomic_t for flags set from signal handlers.
2763 Check getgroups() return and improve debugging.
2764 Fixed developed for diagnosis in bug 927 (which turned out to be
2767 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
2768 Patch from Mark Zealey.
2770 PP/12 Bugzilla 1056: Improved spamd server selection.
2771 Patch from Mark Zealey.
2773 PP/13 Bugzilla 1086: Deal with maildir quota file races.
2774 Based on patch from Heiko Schlittermann.
2776 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
2777 Patch from Uwe Doering, sign-off by Michael Haardt.
2779 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
2780 variable declaration deep within a block. Bug and patch from
2783 PP/15 lookups-Makefile IRIX compatibility coercion.
2785 PP/16 Make DISABLE_DKIM build knob functional.
2787 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
2788 Patch by Simon Arlott
2790 TF/03 Fix valgrind.h portability to C89 compilers that do not support
2791 variable argument macros. Our copy now differs from upstream.
2797 TF/01 Failure to get a lock on a hints database can have serious
2798 consequences so log it to the panic log.
2800 TF/02 Log LMTP confirmation messages in the same way as SMTP,
2801 controlled using the smtp_confirmation log selector.
2803 TF/03 Include the error message when we fail to unlink a spool file.
2805 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
2806 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
2807 for maintaining out-of-tree patches for some time.
2809 PP/01 Bugzilla 139: Documentation and portability issues.
2810 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
2811 Handle per-OS dynamic-module compilation flags.
2813 PP/02 Let /dev/null have normal permissions.
2814 The 4.73 fixes were a little too stringent and complained about the
2815 permissions on /dev/null. Exempt it from some checks.
2816 Reported by Andreas M. Kirchwitz.
2818 PP/03 Report version information for many libraries, including
2819 Exim version information for dynamically loaded libraries. Created
2820 version.h, now support a version extension string for distributors
2821 who patch heavily. Dynamic module ABI change.
2823 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
2824 privilege escalation vulnerability whereby the Exim run-time user
2825 can cause root to append content of the attacker's choosing to
2828 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
2831 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
2832 If dropping privileges for untrusted macros, we disabled normal logging
2833 on the basis that it would fail; for the Exim run-time user, this is not
2834 the case, and it resulted in successful deliveries going unlogged.
2835 Fixed. Reported by Andreas Metzler.
2841 PP/01 Date: & Message-Id: revert to normally being appended to a message,
2842 only prepend for the Resent-* case. Fixes regression introduced in
2843 Exim 4.70 by NM/22 for Bugzilla 607.
2845 PP/02 Include check_rfc2047_length in configure.default because we're seeing
2846 increasing numbers of administrators be bitten by this.
2848 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
2850 PP/03 Bugzilla 994: added openssl_options main configuration option.
2852 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
2854 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
2856 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
2858 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
2859 without a peer certificate, leading to a segfault because of an
2860 assumption that peers always have certificates. Be a little more
2861 paranoid. Problem reported by Martin Tscholak.
2863 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
2864 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
2865 NB: ClamAV planning to remove STREAM in "middle of 2010".
2866 CL also introduces -bmalware, various -d+acl logging additions and
2867 more caution in buffer sizes.
2869 PP/09 Implemented reverse_ip expansion operator.
2871 PP/10 Bugzilla 937: provide a "debug" ACL control.
2873 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2875 PP/12 Bugzilla 973: Implement --version.
2877 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2879 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2881 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2883 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2884 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2885 ignore trailing whitespace.
2887 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2889 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2892 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2893 Notification from Dr Andrew Aitchison.
2895 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2896 ExtendedDetectionInfo response format.
2897 Notification from John Horne.
2899 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2902 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2903 XSL and documented dependency on system catalogs, with examples of how
2906 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2909 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2910 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2911 configuration file which is writeable by the Exim user or group.
2913 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2914 of configuration files to cover files specified with the -C option if
2915 they are going to be used with root privileges, not just the default
2918 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2919 option (effectively making it always true).
2921 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2922 files to be used while preserving root privileges.
2924 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2925 that rogue child processes cannot use them.
2927 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2928 run-time user, instead of root.
2930 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2931 Exim run-time user without dropping privileges.
2933 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2934 result string, instead of calling string_vformat() twice with the same
2937 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2938 for other users. Others should always drop root privileges if they use
2939 -C on the command line, even for a whitelisted configure file.
2941 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2943 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2949 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2950 $data_path, and $header_path variables; fixed documentation bugs and
2953 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2954 exipick to access non-standard spools, including the "frozen" queue
2957 NM/01 Bugzilla 965: Support mysql stored procedures.
2958 Patch from Alain Williams
2960 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2962 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2963 Patch from Andreas Metzler
2965 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2966 Patch from Kirill Miazine
2968 NM/05 Bugzilla 671: Added umask to procmail example.
2970 JJ/03 installed exipick 20100323.0, fixing doc bug
2972 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2973 directory. Notification and patch from Dan Rosenberg.
2975 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2977 TK/02 Improve log output when DKIM signing operation fails.
2979 MH/01 Treat the transport option dkim_domain as a colon separated
2980 list, not as a single string, and sign the message with each element,
2981 omitting multiple occurences of the same signer.
2983 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2984 Bugzilla 985, 986. Patch by Simon Arlott
2986 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2987 Patch by Simon Arlott
2989 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2990 MBX locking. Notification from Dan Rosenberg.
2996 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2998 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
3000 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
3002 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
3004 NM/04 Bugzilla 915: Flag broken perl installation during build.
3010 TK/01 Added patch by Johannes Berg that expands the main option
3011 "spamd_address" if it starts with a dollar sign.
3013 TK/02 Write list of recipients to X-Envelope-Sender header when building
3014 the mbox-format spool file for content scanning (suggested by Jakob
3017 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
3018 (http://www.dcc-servers.net/) support via dccifd. Activated by
3019 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
3021 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
3022 by Mark Daniel Reidel <mr@df.eu>.
3024 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
3025 When building exim an external PCRE library is now needed -
3026 PCRE is a system library on the majority of modern systems.
3027 See entry on PCRE_LIBS in EDITME file.
3029 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
3030 conversation. Added nologin parameter to request.
3031 Patch contributed by Kirill Miazine.
3033 TF/01 Do not log submission mode rewrites if they do not change the address.
3035 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
3037 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
3038 log files in place. Contributed by Roberto Lima.
3040 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
3042 TF/03 Bugzilla 615: When checking the local_parts router precondition
3043 after a local_part_suffix or local_part_prefix option, Exim now
3044 does not use the address's named list lookup cache, since this
3045 contains cached lookups for the whole local part.
3047 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
3048 Robert Millan. Documentation is in experimental-spec.txt.
3050 TF/04 Bugzilla 668: Fix parallel build (make -j).
3052 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
3054 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
3055 Patch provided by Jan Srzednicki.
3057 TF/05 Leading white space used to be stripped from $spam_report which
3058 wrecked the formatting. Now it is preserved.
3060 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
3061 that they are available at delivery time.
3063 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
3065 TF/08 TLS error reporting now respects the incoming_interface and
3066 incoming_port log selectors.
3068 TF/09 Produce a more useful error message if an SMTP transport's hosts
3069 setting expands to an empty string.
3071 NM/06 Bugzilla 744: EXPN did not work under TLS.
3072 Patch provided by Phil Pennock.
3074 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
3075 Patch provided by Richard Godbee.
3077 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
3078 acl_smtp_notquit, added index entry.
3080 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
3081 Patch provided by Eugene Bujak.
3083 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
3084 accept(). Patch provided by Maxim Dounin.
3086 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
3087 Patch provided by Phil Pennock.
3089 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
3091 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
3092 Patch provided by Brad "anomie" Jorsch.
3094 NM/14 Bugzilla 622: Added timeout setting to transport filter.
3095 Patch provided by Dean Brooks.
3097 TK/05 Add native DKIM support (does not depend on external libraries).
3099 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
3100 Patch provided by Graeme Fowler.
3102 NM/16 Bugzilla 851: Documentation example syntax fix.
3104 NM/17 Changed NOTICE file to remove references to embedded PCRE.
3106 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
3109 NM/19 Bugzilla 745: TLS version reporting.
3110 Patch provided by Phil Pennock.
3112 NM/20 Bugzilla 167: bool: condition support.
3113 Patch provided by Phil Pennock.
3115 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
3116 clients. Patch provided by Phil Pennock.
3118 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
3119 Patch provided by Brad "anomie" Jorsch.
3121 NM/23 Bugzilla 687: Fix misparses in eximstats.
3122 Patch provided by Heiko Schlittermann.
3124 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
3125 Patch provided by Heiko Schlittermann.
3127 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
3128 plus update to original patch.
3130 NM/26 Bugzilla 799: Documentation correction for ratelimit.
3132 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
3133 Patch provided by David Brownlee.
3135 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
3137 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
3139 NM/30 Bugzilla 888: TLS documentation bugfixes.
3141 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
3143 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
3144 Unlike the original bugzilla I have changed all shell scripts in src tree.
3146 NM/33 Bugzilla 898: Transport filter timeout fix.
3147 Patch by Todd Rinaldo.
3149 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
3150 Patch by Serge Demonchaux.
3152 NM/35 Bugzilla 39: Base64 decode bug fixes.
3153 Patch by Jakob Hirsch.
3155 NM/36 Bugzilla 909: Correct connect() call in dcc code.
3157 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
3159 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
3161 NM/39 Bugzilla 911: Fixed MakeLinks build script.
3167 TK/01 Add preliminary DKIM support. Currently requires a forked version of
3168 ALT-N's libdkim that I have put here:
3169 http://duncanthrax.net/exim-experimental/
3171 Note to Michael Haardt: I had to rename some vars in sieve.c. They
3172 were called 'true' and it seems that C99 defines that as a reserved
3173 keyword to be used with 'bool' variable types. That means you could
3174 not include C99-style headers which use bools without triggering
3175 build errors in sieve.c.
3177 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
3178 as mailq or other aliases. Changed the --help handling significantly
3179 to do whats expected. exim_usage() emits usage/help information.
3181 SC/01 Added the -bylocaldomain option to eximstats.
3183 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
3185 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
3187 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
3193 PH/01 Another patch from the Sieve maintainer.
3195 PH/02 When an IPv6 address is converted to a string for single-key lookup
3196 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
3197 dots are used instead of colons so that keys in lsearch files need not
3198 contain colons. This was done some time before quoting was made available
3199 in lsearch files. However, iplsearch files do require colons in IPv6 keys
3200 (notated using the quote facility) so as to distinguish them from IPv4
3201 keys. This meant that lookups for IP addresses in host lists did not work
3202 for iplsearch lookups.
3204 This has been fixed by arranging for IPv6 addresses to be expressed with
3205 colons if the lookup type is iplsearch. This is not incompatible, because
3206 previously such lookups could never work.
3208 The situation is now rather anomalous, since one *can* have colons in
3209 ordinary lsearch keys. However, making the change in all cases is
3210 incompatible and would probably break a number of configurations.
3212 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
3215 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
3216 conversion specification without a maximum field width, thereby enabling
3217 a rogue spamd server to cause a buffer overflow. While nobody in their
3218 right mind would setup Exim to query an untrusted spamd server, an
3219 attacker that gains access to a server running spamd could potentially
3220 exploit this vulnerability to run arbitrary code as the Exim user.
3222 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
3223 $primary_hostname instead of what libspf2 thinks the hosts name is.
3225 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
3226 a directory entry by the name of the lookup key. Previously, if a
3227 symlink pointed to a non-existing file or a file in a directory that
3228 Exim lacked permissions to read, a lookup for a key matching that
3229 symlink would fail. Now it is enough that a matching directory entry
3230 exists, symlink or not. (Bugzilla 503.)
3232 PH/03 The body_linecount and body_zerocount variables are now exported in the
3235 PH/04 Added the $dnslist_matched variable.
3237 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
3238 This means they are set thereafter only if the connection becomes
3241 PH/06 Added the client_condition to authenticators so that some can be skipped
3242 by clients under certain conditions.
3244 PH/07 The error message for a badly-placed control=no_multiline_responses left
3245 "_responses" off the end of the name.
3247 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
3249 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
3250 (without spaces) instead of just copying the configuration text.
3252 PH/10 Added the /noupdate option to the ratelimit ACL condition.
3254 PH/11 Added $max_received_linelength.
3256 PH/12 Added +ignore_defer and +include_defer to host lists.
3258 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
3259 way in which PCRE > 7.0 is built.
3261 PH/14 Implemented queue_only_load_latch.
3263 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
3264 MAIL command. The effect was to mangle the value on 64-bit systems.
3266 PH/16 Another patch from the Sieve maintainer.
3268 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
3270 PH/18 If a system quota error occurred while trying to create the file for
3271 a maildir delivery, the message "Mailbox is full" was not appended to the
3272 bounce if the delivery eventually timed out. Change 4.67/27 below applied
3273 only to a quota excession during the actual writing of the file.
3275 PH/19 It seems that peer DN values may contain newlines (and other non-printing
3276 characters?) which causes problems in log lines. The DN values are now
3277 passed through string_printing() before being added to log lines.
3279 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
3280 and InterBase are left for another time.)
3282 PH/21 Added message_body_newlines option.
3284 PH/22 Guard against possible overflow in moan_check_errorcopy().
3286 PH/23 POSIX allows open() to be a macro; guard against that.
3288 PH/24 If the recipient of an error message contained an @ in the local part
3289 (suitably quoted, of course), incorrect values were put in $domain and
3290 $local_part during the evaluation of errors_copy.
3296 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
3297 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
3300 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
3301 issue a MAIL command.
3303 PH/02 In an ACL statement such as
3305 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
3307 if a client was not listed at all, or was listed with a value other than
3308 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
3309 the condition was not true (as it should be), so access was not denied.
3310 The bug was that the ! inversion was incorrectly passed on to the second
3311 item. This has been fixed.
3313 PH/03 Added additional dnslists conditions == and =& which are different from
3314 = and & when the dns lookup returns more than one IP address.
3316 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
3317 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
3319 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
3320 FSYNC, which compiles an option called disable_fsync that allows for
3321 bypassing fsync(). The documentation is heavily laced with warnings.
3323 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
3325 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
3326 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
3327 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
3328 including adding "make clean"; (3) Added -fPIC when compiling the test
3329 dynamically loaded module, to get rid of a warning.
3331 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
3332 message fails, move_frozen_messages = true and ignore_bounce_errors_after
3333 = 0s. The bug is otherwise harmless.
3335 PH/07 There was a bug in the dovecot authenticator such that the value of
3336 $auth1 could be overwritten, and so not correctly preserved, after a
3337 successful authentication. This usually meant that the value preserved by
3338 the server_setid option was incorrect.
3340 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
3342 PH/09 Installed PCRE release 7.0.
3344 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
3345 run for batched SMTP input. It is now run at the start of every message
3346 in the batch. While fixing this I discovered that the process information
3347 (output by running exiwhat) was not always getting set for -bs and -bS
3348 input. This is fixed, and it now also says "batched" for BSMTP.
3350 PH/11 Added control=no_pipelining.
3352 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
3353 patch, slightly modified), and move the expansion of helo_data till after
3354 the connection is made in the smtp transport (so it can use these
3357 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
3359 PH/14 Added log_selector = +pid.
3361 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
3363 PH/16 Add ${if forany and ${if forall.
3365 PH/17 Added dsn_from option to vary the From: line in DSNs.
3367 PH/18 Flush SMTP output before performing a callout, unless control =
3368 no_callout_flush is set.
3370 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
3371 was true (the default) a successful delivery failed to delete the retry
3372 item, thus causing premature timeout of the address. The bug is now
3375 PH/20 Added hosts_avoid_pipelining to the smtp transport.
3377 PH/21 Long custom messages for fakedefer and fakereject are now split up
3378 into multiline responses in the same way that messages for "deny" and
3379 other ACL rejections are.
3381 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
3382 with slight modification.
3384 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
3385 draft, changing the syntax and factoring some duplicate code".
3387 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
3388 for deliveries of the second and subsequent messages over the same SMTP
3391 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
3392 ${reduce, with only minor "tidies".
3394 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
3396 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
3397 expansion side effects.
3399 PH/27 When a message times out after an over-quota error from an Exim-imposed
3400 quota, the bounce message says "mailbox is full". This message was not
3401 being given when it was a system quota that was exceeded. It now should
3404 MH/03 Made $recipients available in local_scan(). local_scan() already has
3405 better access to the recipient list through recipients_list[], but
3406 $recipients can be useful in postmaster-provided expansion strings.
3408 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
3409 in the case of a MAIL command with additional options following the
3410 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
3411 were accidentally chopped off.
3413 PH/29 SMTP synchronization checks are implemented when a command is read -
3414 there is a check that no more input is waiting when there shouldn't be
3415 any. However, for some commands, a delay in an ACL can mean that it is
3416 some time before the response is written. In this time, more input might
3417 arrive, invalidly. So now there are extra checks after an ACL has run for
3418 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
3419 pipelining has not been advertised.
3421 PH/30 MH's patch to allow iscntrl() characters to be list separators.
3423 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
3424 returned in the SMTP response when smtp_return_error_details was false.
3425 This has been fixed.
3427 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
3428 instead of the C I/O that was originally supplied, because problems were
3429 reported on Solaris.
3431 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
3432 Exim which did not show up earlier: it was assuming that a call to
3433 SSL_CTX_set_info_callback() might give an error value. In fact, there is
3434 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
3435 was a macro that became an assignment, so it seemed to work. This has
3436 changed to a proper function call with a void return, hence the compile
3437 error. Exim's code has been fixed.
3439 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
3442 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
3444 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
3446 PH/37 If a message is not accepted after it has had an id assigned (e.g.
3447 because it turns out to be too big or there is a timeout) there is no
3448 "Completed" line in the log. When some messages of this type were
3449 selected by exigrep, they were listed as "not completed". Others were
3450 picked up by some special patterns. I have improved the selection
3451 criteria to be more general.
3453 PH/38 The host_find_failed option in the manualroute router can now be set
3454 to "ignore", to completely ignore a host whose IP address cannot be
3455 found. If all hosts are ignored, the behaviour is controlled by the new
3456 host_all_ignored option.
3458 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
3459 homing or because of multiple MX records with /mx) generated more than
3460 one IP address, and the following item turned out to be the local host,
3461 all the secondary addresses of the first item were incorrectly removed
3462 from the list, along with the local host and any following hosts (which
3463 is what is supposed to happen).
3465 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
3466 whoever called Exim into the -H file. In the case of the daemon it was
3467 behaving confusingly. When first started, it used values for whoever
3468 started the daemon, but after a SIGHUP it used the Exim user (because it
3469 calls itself on a restart). I have changed the code so that it now always
3472 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
3473 message are rejected with the same error (e.g. no authentication or bad
3474 sender address), and a DATA command is nevertheless sent (as can happen
3475 with PIPELINING or a stupid MUA), the error message that was given to the
3476 RCPT commands is included in the rejection of the DATA command. This is
3477 intended to be helpful for MUAs that show only the final error to their
3480 PH/42 Another patch from the Sieve maintainer.
3482 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
3483 Eximstats - Fixed some broken HTML links and added missing column headers
3485 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
3486 columns for Rejects, Temp Rejects, Ham, and Spam rows.
3488 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
3490 PH/43 Yet another patch from the Sieve maintainer.
3492 PH/44 I found a way to check for a TCP/IP connection going away before sending
3493 the response to the final '.' that terminates a message, but only in the
3494 case where the client has not sent further data following the '.'
3495 (unfortunately, this is allowed). However, in many cases there won't be
3496 any further data because there won't be any more messages to send. A call
3497 to select() can be used: if it shows that the input is "ready", there is
3498 either input waiting, or the socket has been closed. An attempt to read
3499 the next input character can distinguish the two cases. Previously, Exim
3500 would have sent an OK response which the client would never have see.
3501 This could lead to message repetition. This fix should cure that, at
3502 least in a lot of common cases.
3504 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
3505 advertised in response to EHLO.
3511 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
3512 fixed by 4.65/MH/01 (is this a record?) are fixed:
3514 (i) An empty string was always treated as zero by the numeric comparison
3515 operators. This behaviour has been restored.
3517 (ii) It is documented that the numeric comparison operators always treat
3518 their arguments as decimal numbers. This was broken in that numbers
3519 starting with 0 were being interpreted as octal.
3521 While fixing these problems I realized that there was another issue that
3522 hadn't been noticed. Values of message_size_limit (both the global option
3523 and the transport option) were treated as octal if they started with 0.
3524 The documentation was vague. These values are now always treated as
3525 decimal, and I will make that clear in the documentation.
3531 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
3532 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
3535 MH/01 Don't check that the operands of numeric comparison operators are
3536 integers when their expansion is in "skipping" mode (fixes bug
3537 introduced by 4.64-PH/07).
3539 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
3540 child addresses, Exim now panics and dies. Previously, because the count
3541 is held in a short int, deliveries were likely to be lost. As such a
3542 large number of recipients for a single message is ridiculous
3543 (performance will be very, very poor), I have chosen to impose a limit
3544 rather than extend the field.
3550 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
3551 leftover -K file (the existence of which was triggered by #402).
3552 While we were at it, introduced process PID as part of the -K
3553 filename. This should rule out race conditions when creating
3556 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
3557 processing considerably. Previous code took too long for large mails,
3558 triggering a timeout which in turn triggers #401.
3560 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
3561 in the DK code in transports.c. sendfile() is not really portable,
3562 hence the _LINUX specificness.
3564 TF/01 In the add_headers option to the mail command in an Exim filter,
3565 there was a bug that Exim would claim a syntax error in any
3566 header after the first one which had an odd number of characters
3569 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
3570 callout verification, Exim cached a "reject" for the entire domain. This
3571 is correct for most verifications, but it is not correct for a recipient
3572 verification with use_sender or use_postmaster set, because in that case
3573 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
3574 case of MAIL FROM:<> rejection from other early rejections (e.g.
3575 rejection of HELO). When verifying a recipient using a non-null MAIL
3576 address, the cache is ignored if it shows MAIL FROM:<> rejection.
3577 Whatever the result of the callout, the value of the domain cache is
3578 left unchanged (for any other kind of callout, getting as far as trying
3579 RCPT means that the domain itself is ok).
3581 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
3584 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
3585 manifest itself as EPIPE rather than ECONNECT. When tidying away a
3586 session, the daemon ignores ECONNECT errors and logs others; it now
3587 ignores EPIPE as well.
3589 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
3590 (quoted-printable decoding).
3592 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
3593 later the small subsequent patch to fix an introduced bug.
3595 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
3597 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
3599 PH/08 An error is now given if message_size_limit is specified negative.
3601 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
3602 to be given (somewhat) arbitrary names.
3604 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
3607 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
3608 miscellaneous code fixes
3610 PH/10 Added the log_reject_target ACL modifier to specify where to log
3613 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
3614 hostname. This is wrong, because it relates to the incoming message (and
3615 probably the interface on which it is arriving) and not to the outgoing
3616 callout (which could be using a different interface). This has been
3617 changed to use the value of the helo_data option from the smtp transport
3618 instead - this is what is used when a message is actually being sent. If
3619 there is no remote transport (possible with a router that sets up host
3620 addresses), $smtp_active_hostname is used.
3622 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
3623 tweaks were necessary in order to get it to work (see also 21 below):
3624 (a) The code assumed that strncpy() returns a negative number on buffer
3625 overflow, which isn't the case. Replaced with Exim's string_format()
3627 (b) There were several signed/unsigned issues. I just did the minimum
3628 hacking in of casts. There is scope for a larger refactoring.
3629 (c) The code used strcasecmp() which is not a standard C function.
3630 Replaced with Exim's strcmpic() function.
3631 (d) The code set only $1; it now sets $auth1 as well.
3632 (e) A simple test gave the error "authentication client didn't specify
3633 service in request". It would seem that Dovecot has changed its
3634 interface. Fortunately there's a specification; I followed it and
3635 changed what the client sends and it appears to be working now.
3637 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
3640 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
3641 address A is aliased to B and C, where B exists and C does not. Without
3642 -v the output is "A verified" because verification stops after a
3643 successful redirection if more than one address is generated. However,
3644 with -v the child addresses are also verified. Exim was outputting "A
3645 failed to verify" and then showing the successful verification for C,
3646 with its parentage. It now outputs "B failed to verify", showing B's
3647 parentage before showing the successful verification of C.
3649 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
3650 look up a TXT record in a specific list after matching in a combined
3653 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
3654 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
3655 they consult the DNS. I had assumed they would set it the way they
3656 wanted; and indeed my experiments on Linux seem to show that in some
3657 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
3658 To be on the safe side, however, I have now made the interface to
3659 host_find_byname() similar to host_find_bydns(), with an argument
3660 containing the DNS resolver options. The host_find_byname() function now
3661 sets these options at its start, just as host_find_bydns() does. The smtp
3662 transport options dns_qualify_single and dns_search_parents are passed to
3663 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
3664 of host_find_byname() use the default settings of RES_DEFNAMES
3665 (qualify_single) but not RES_DNSRCH (search_parents).
3667 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
3668 spool_read_header() do less string testing, by means of a preliminary
3669 switch on the second character of optional "-foo" lines. (This is
3670 overdue, caused by the large number of possibilities that now exist.
3671 Originally there were few.) While I was there, I also converted the
3672 str(n)cmp tests so they don't re-test the leading "-" and the first
3673 character, in the hope this might squeeze out yet more improvement.
3675 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
3676 flag allowing group syntax was set by the header_syntax check but not
3677 turned off, possible causing trouble later; (2) The flag was not being
3678 set at all for the header_verify test, causing "group"-style headers to
3679 be rejected. I have now set it in this case, and also caused header_
3680 verify to ignore an empty address taken from a group. While doing this, I
3681 came across some other cases where the code for allowing group syntax
3682 while scanning a header line wasn't quite right (mostly, not resetting
3683 the flag correctly in the right place). These bugs could have caused
3684 trouble for malformed header lines. I hope it is now all correct.
3686 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
3687 with the "reply" argument non-NULL. The code, however (which originally
3688 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
3689 but it didn't always do it. This confused somebody who was copying the
3690 code for some other use. I have removed all the tests.
3692 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
3693 feature that was used to support insecure browsers during the U.S. crypto
3694 embargo. It requires special client support, and Exim is probably the
3695 only MTA that supported it -- and would never use it because real RSA is
3696 always available. This code has been removed, because it had the bad
3697 effect of slowing Exim down by computing (never used) parameters for the
3698 RSA_EXPORT functionality.
3700 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
3701 authenticator to fail if there's a tab character in the incoming data
3702 (there should never be unless someone is messing about, as it's supposed
3703 to be base64-encoded). Also added, on Timo's advice, the "secured" option
3704 if the connection is using TLS or if the remote IP is the same as the
3705 local IP, and the "valid-client-cert option" if a client certificate has
3708 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
3709 authenticators. This can be used for authorization after authentication
3710 succeeds. (In the case of plaintext, it servers for both authentication
3713 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
3714 if any retry times were supplied.
3716 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
3717 connection, where there is no client IP address to check. In this
3718 situation, the verify now always succeeds.
3720 PH/25 Applied John Jetmore's -Mset patch.
3722 PH/26 Added -bem to be like -Mset, but loading a message from a file.
3724 PH/27 In a string expansion for a processed (not raw) header when multiple
3725 headers of the same name were present, leading whitespace was being
3726 removed from all of them, but trailing whitespace was being removed only
3727 from the last one. Now trailing whitespace is removed from each header
3728 before concatenation. Completely empty headers in a concatenation (as
3729 before) are ignored.
3731 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
3732 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
3734 PH/29 [Removed. This was a change that I later backed out, and forgot to
3735 correct the ChangeLog entry (that I had efficiently created) before
3736 committing the later change.]
3738 PH/30 Exim was sometimes attempting to deliver messages that had suffered
3739 address errors (4xx response to RCPT) over the same connection as other
3740 messages routed to the same hosts. Such deliveries are always "forced",
3741 so retry times are not inspected. This resulted in far too many retries
3742 for the affected addresses. The effect occurred only when there were more
3743 hosts than the hosts_max_try setting in the smtp transport when it had
3744 the 4xx errors. Those hosts that it had tried were not added to the list
3745 of hosts for which the message was waiting, so if all were tried, there
3746 was no problem. Two fixes have been applied:
3748 (i) If there are any address or message errors in an SMTP delivery, none
3749 of the hosts (tried or untried) are now added to the list of hosts
3750 for which the message is waiting, so the message should not be a
3751 candidate for sending over the same connection that was used for a
3752 successful delivery of some other message. This seems entirely
3753 reasonable: after all the message is NOT "waiting for some host".
3754 This is so "obvious" that I'm not sure why it wasn't done
3755 previously. Hope I haven't missed anything, but it can't do any
3756 harm, as the worst effect is to miss an optimization.
3758 (ii) If, despite (i), such a delivery is accidentally attempted, the
3759 routing retry time is respected, so at least it doesn't keep
3760 hammering the server.
3762 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
3763 in ${readsocket because some servers need this prod.
3765 PH/32 Added some extra debug output when updating a wait-xxx database.
3767 PH/33 The hint "could be header name not terminated by colon", which has been
3768 given for certain expansion errors for a long time, was not being given
3769 for the ${if def:h_colon_omitted{... case.
3771 PH/34 The spec says: "With one important exception, whenever a domain list is
3772 being scanned, $domain contains the subject domain." There was at least
3773 one case where this was not true.
3775 PH/35 The error "getsockname() failed: connection reset by peer" was being
3776 written to the panic log as well as the main log, but it isn't really
3777 panic-worthy as it just means the connection died rather early on. I have
3778 removed the panic log writing for the ECONNRESET error when getsockname()
3781 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
3782 runs only) independently of the message's sender address. This meant
3783 that, if the 4xx error was in fact related to the sender, a different
3784 message to the same recipient with a different sender could confuse
3785 things. In particular, this can happen when sending to a greylisting
3786 server, but other circumstances could also provoke similar problems.
3787 I have changed the default so that the retry time for these errors is now
3788 based a combination of the sender and recipient addresses. This change
3789 can be overridden by setting address_retry_include_sender=false in the
3792 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
3793 remote server are returned as part of bounce messages. This was not
3794 happening for LMTP over a pipe (the lmtp transport), but now it is the
3795 same for both kinds of LMTP.
3797 PH/38 Despite being documented as not happening, Exim was rewriting addresses
3798 in header lines that were in fact CNAMEs. This is no longer the case.
3800 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
3801 and queue runs started by the daemon processed all messages. This has
3802 been fixed so that -R and -S can now usefully be given with -q<time>.
3804 PH/40 Import PCRE release 6.7 (fixes some bugs).
3806 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
3808 PH/42 Give an error if -q is specified more than once.
3810 PH/43 Renamed the variables $interface_address and $interface_port as
3811 $received_ip_address and $received_port, to make it clear that these
3812 values apply to message reception, and not to the outgoing interface when
3813 a message is delivered. (The old names remain recognized, of course.)
3815 PH/44 There was no timeout on the connect() call when using a Unix domain
3816 socket in the ${readsocket expansion. There now is.
3818 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
3819 be meaningful with "accept".
3821 SC/01 Eximstats V1.43
3822 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
3824 SC/02 Eximstats V1.44
3825 Use a glob alias rather than an array ref in the generated
3826 parser. This improves both readability and performance.
3828 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
3829 Collect SpamAssassin and rejection statistics.
3830 Don't display local sender or destination tables unless
3831 there is data to show.
3832 Added average volumes into the top table text output.
3834 SC/04 Eximstats V1.46
3835 Collect data on the number of addresses (recipients)
3836 as well as the number of messages.
3838 SC/05 Eximstats V1.47
3839 Added 'Message too big' to the list of mail rejection
3840 reasons (thanks to Marco Gaiarin).
3842 SC/06 Eximstats V1.48
3843 Mainlog lines which have GMT offsets and are too short to
3844 have a flag are now skipped.
3846 SC/07 Eximstats V1.49 (Alain Williams)
3847 Added the -emptyok flag.
3849 SC/08 Eximstats V1.50
3850 Fixes for obtaining the IP address from reject messages.
3852 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
3853 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
3854 whitespace changes from 4.64-PH/27
3856 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
3859 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
3860 are found, allow negative numbers in numeric criteria)
3862 JJ/06 exipick.20061117.2, added new $message_body_missing variable
3864 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
3865 to match changes made in 4.64-PH/43
3867 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
3869 PH/47 Put in an explicit test for a DNS lookup of an address record where the
3870 "domain" is actually an IP address, and force a failure. This locks out
3871 those revolvers/nameservers that support "A-for-A" lookups, in
3872 contravention of the specifications.
3874 PH/48 When a host name was looked up from an IP address, and the subsequent
3875 forward lookup of the name timed out, the host name was left in
3876 $sender_host_name, contrary to the specification.
3878 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3879 restricted to single-key lookups, Exim was not diagnosing an error if
3880 * or *@ was used with a query-style lookup.
3882 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3884 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3885 long ago, but noone interested enough thought of it. Let's just say that
3886 the "1.1" means that there are some new functions that weren't there at
3887 some point in the past.
3889 PH/51 Error processing for expansion failure of helo_data from an smtp
3890 transport during callout processing was broken.
3892 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3893 tested/used via the -bh/-bhc/-bs options.
3895 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3896 bug, fixed in subsequent PCRE releases).
3898 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3899 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3901 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3907 SC/01 Use a glob alias rather than an array ref in eximstats generated
3908 parser. This improves both readability and performance.
3910 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3911 Don't display local sender or destination tables in eximstats unless
3912 there is data to show.
3913 Added average volumes into the eximstats top table text output.
3915 SC/03 Collect data on the number of addresses (recipients) as well
3916 as the number of messages in eximstats.
3918 TF/01 Correct an error in the documentation for the redirect router. Exim
3919 does (usually) call initgroups() when daemonizing.
3921 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3922 with consistent privilege compared to when running as a daemon.
3924 TF/03 Note in the spec that $authenticated_id is not set for local
3925 submissions from trusted users.
3927 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3928 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3930 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3931 by adding some example configuration directives to the default
3932 configuration file. A little bit of work is required to uncomment the
3933 directives and define how usernames and passwords are checked, but
3934 there is now a framework to start from.
3936 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3937 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3938 without this. I don't know how relevant this is to other LDAP libraries.
3940 PH/02 Add the verb name to the "unknown ACL verb" error.
3942 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3944 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3946 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3947 directory not expanded when it should be if an expanded home directory
3948 was set for the address (which is overridden by the transport).
3950 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3953 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3954 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3955 because it is too late at that time, and has no effect.
3957 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3958 security issue with \' (bugzilla #107). I could not use the
3959 PQescapeStringConn() function, because it needs a PGconn value as one of
3962 PH/08 When testing addresses using -bt, indicate those final addresses that
3963 are duplicates that would not cause an additional delivery. At least one
3964 person was confused, thinking that -bt output corresponded to deliveries.
3965 (Suppressing duplicates isn't a good idea as you lose the information
3966 about possibly different redirections that led to the duplicates.)
3968 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3969 systems where poll() doesn't work, in particular OS X.
3971 PH/10 Added more information to debugging output for retry time not reached.
3973 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3974 operations in malware.c.
3976 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3979 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3980 syslog_duplication set false, log lines that would normally be written
3981 both the the main log and to the reject log were not written to syslog at
3984 PH/14 In the default configuration, change the use of "message" in ACL warn
3985 statements to "add_header".
3987 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3988 not followed by a command (e.g. "seen endif").
3990 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3991 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3994 PH/17 Added extra conditions to the default value of delay_warning_condition
3998 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3999 { match{$h_precedence:}{(?i)bulk|list|junk} } \
4000 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
4003 The Auto-Submitted: and various List- headers are standardised, whereas I
4004 don't think Precedence: ever was.
4006 PH/18 Refactored debugging code in route_finduser() to show more information,
4007 in particular, the error code if getpwnam() issues one.
4009 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
4010 This is apparently needed in addition to the PH/07 change above to avoid
4011 any possible encoding problems.
4013 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
4014 but not after initializing Perl.
4016 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
4017 output them only if debugging. By default they are written stderr,
4018 apparently, which is not desirable.
4020 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
4023 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
4026 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
4028 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
4029 authenticated or an ident call has been made. Suppress the default
4030 values for $authenticated_id and $authenticated_sender (but permit -oMai
4031 and -oMas) when testing with -bh.
4033 PH/24 Re-jigged the order of the tests in the default configuration so that the
4034 tests for valid domains and recipients precede the DNS black list and CSA
4035 tests, on the grounds that those ones are more expensive.
4037 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
4038 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
4039 This bug exists in every version of Exim that I still have, right back to
4042 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
4043 However, an attempt to turn on case-sensitivity in a regex key by
4044 including (?-i) didn't work because the subject string was already
4045 lowercased, and the effects were non-intuitive. It turns out that a
4046 one-line patch can be used to allow (?-i) to work as expected.
4052 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
4053 other effects) broke the use of negated acl sub-conditions.
4055 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
4058 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
4059 "Deny" causes Exim to reject the incoming connection with a 554 error.
4060 Unfortunately, if there is a major crisis, such as a disk failure,
4061 tcp-wrappers gives "deny", whereas what one would like would be some
4062 kind of temporary error. A kludge has been added to help with this.
4063 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
4064 554 error is used if errno is still zero or contains ENOENT (which occurs
4065 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
4068 PH/03 Add -lutil to the default FreeBSD LIBS setting.
4070 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
4071 errors. Otherwise a message that provokes a temporary error (when other
4072 messages do not) can cause a whole host to time out.
4074 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
4075 addresses were routed directly to files or pipes from a redirect router.
4076 File deliveries just didn't batch; pipe deliveries might have suffered
4079 PH/06 A failure to get a lock for a hints database would erroneously always say
4080 "Failed to get write lock", even when it was really a read lock.
4082 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
4083 of 0600. This has been changed to use the value of the lockfile_mode
4084 option (which defaults to 0600).
4086 PH/08 Applied small patch from the Sieve maintainer.
4088 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
4089 folder from quota calculations, a direct delivery into this folder messed
4090 up the contents of the maildirsize file. This was because the regex was
4091 used only to exclude .Trash (or whatever) when the size of the mailbox
4092 was calculated. There was no check that a delivery was happening into an
4093 excluded directory. This bug has been fixed by ignoring all quota
4094 processing for deliveries into excluded directories.
4096 PH/10 Added the maildirfolder_create_regex option to appendfile.
4102 PH/01 The code for finding all the local interface addresses on a FreeBSD
4103 system running IPv6 was broken. This may well have applied to all BSD
4104 systems, as well as to others that have similar system calls. The broken
4105 code found IPv4 interfaces correctly, but gave incorrect values for the
4106 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
4107 that it would not match correctly against @[] and not recognize the IPv6
4110 PH/02 The ipliteral router was not recognizing addresses of the form user@
4111 [ipv6:....] because it didn't know about the "ipv6:" prefix.
4113 PH/03 Added disable_ipv6.
4115 PH/04 Changed $reply_address to use the raw form of the headers instead of the
4116 decoded form, because it is most often used to construct To: headers
4117 lines in autoreplies, and the decoded form may well be syntactically
4118 invalid. However, $reply_address has leading white space removed, and all
4119 newlines turned into spaces so that the autoreply transport does not
4122 PH/05 If group was specified without a user on a router, and no group or user
4123 was specified on a transport, the group from the router was ignored.
4125 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
4126 for visible compile-time settings that can be used to change these
4127 numbers, for those that want even more. Backwards compatibility with old
4128 spool files has been maintained. However, going back to a previous Exim
4129 release will lost any variables that are in spool files.
4131 PH/07 Two small changes when running in the test harness: increase delay when
4132 passing a TCP/IP connection to a new process, in case the original
4133 process has to generate a bounce, and remove special handling of
4134 127.0.0.2 (sic), which is no longer necessary.
4136 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
4137 be the same on different OS.
4139 PH/09 Moved a debug statement in filter processing to avoid a race problem when
4142 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
4143 whether --show-vars was specified or not
4145 JJ/02 exipick: Added support for new ACL variable spool format introduced
4148 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
4149 syntactically invalid From: or Reply-to: line, and a filter used this to
4150 generate an autoreply, and therefore failed to obtain an address for the
4151 autoreply, Exim could try to deliver to a non-existent relative file
4152 name, causing unrelated and misleading errors. What now happens is that
4153 it logs this as a hard delivery error, but does not attempt to create a
4156 PH/11 The exinext utility has a -C option for testing purposes, but although
4157 the given file was scanned by exinext itself; it wasn't being passed on
4158 when Exim was called.
4160 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
4161 an end-of-file indication when reading a command response.
4163 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
4164 compiled. In many other places in Exim, IPv6 addresses are always
4165 recognized, so I have changed this. It also means that IPv4 domain
4166 literals of the form [IPV4:n.n.n.n] are now always recognized.
4168 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
4169 used if the router is not running as root, for example, when verifying at
4170 ACL time, or when using -bh. The debugging output from this situation was
4171 non-existent - all you got was a failure to exec. I have made two
4174 (a) Failures to set uid/gid, the current directory, or a process leader
4175 in a subprocess such as that created by queryprogram now generate
4176 suitable debugging output when -d is set.
4178 (b) The queryprogram router detects when it is not running as root,
4179 outputs suitable debugging information if -d is set, and then runs
4180 the subprocess without attempting to change uid/gid.
4182 PH/15 Minor change to Makefile for building test_host (undocumented testing
4185 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
4186 additional section of a DNS packet that returns MX or SRV records.
4187 Instead, it always explicitly searches for A/AAAA records. This avoids
4188 major problems that occur when a DNS server includes only records of one
4189 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
4190 fixed another bug: if SRV records were looked up and the corresponding
4191 address records were *not* found in the additional section, the port
4192 values from the SRV records were lost.
4194 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
4195 using the correct key (the original address) when searching the retry
4196 rules in order to find which one to use for generating the retry hint.
4198 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
4199 adding the default one. Similarly, if it contains a Reply-To: header, the
4200 errors_reply_to option, if set, is not used.
4202 PH/19 When calculating a retry time, Exim used to measure the "time since
4203 failure" by looking at the "first failed" field in the retry record. Now
4204 it does not use this if it is later than than the arrival time of the
4205 message. Instead it uses the arrival time. This makes for better
4206 behaviour in cases where some deliveries succeed, thus re-setting the
4207 "first failed" field. An example is a quota failure for a huge message
4208 when small messages continue to be delivered. Without this change, the
4209 "time since failure" will always be short, possible causing more frequent
4210 delivery attempts for the huge message than are intended.
4211 [Note: This change was subsequently modified - see PH/04 for 4.62.]
4213 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
4214 $1, $2, $3) because the numerical variables can be reset during some
4215 expansion items (e.g. "match"), thereby losing the authentication data.
4217 PH/21 Make -bV show the size of off_t variables so that the test suite can
4218 decide whether to run tests for quotas > 2G.
4220 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
4221 mailbox_size, and mailbox_filecount in the appendfile transport. If a
4222 filecount value is greater than 2G or if a quota value is greater than 2G
4223 on a system where the size of off_t is not greater than 4, a panic error
4226 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
4227 never match. The debug and -bh output now contains an explicit error
4228 message indicating a malformed IPv4 address or mask.
4230 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
4231 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
4232 PH/23 above applies.
4234 PH/25 Do not write to syslog when running in the test harness. The only
4235 occasion when this arises is a failure to open the main or panic logs
4236 (for which there is an explicit test).
4238 PH/26 Added the /no_tell option to "control=freeze".
4240 PH/27 If a host name lookup failed very early in a connection, for example, if
4241 the IP address matched host_lookup and the reverse lookup yielded a name
4242 that did not have a forward lookup, an error message of the form "no IP
4243 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
4244 could be logged. Now it outputs the IP address instead of "NULL".
4246 PH/28 An enabling patch from MH: add new function child_open_exim2() which
4247 allows the sender and the authenticated sender to be set when
4248 submitting a message from within Exim. Since child_open_exim() is
4249 documented for local_scan(), the new function should be too.
4251 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
4252 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
4253 results in an empty string is now treated as unset.
4255 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
4257 PH/31 Added sender_verify_fail logging option.
4259 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
4260 needed by RFC 822 but not by RFC 2822 was commented out. I have now
4261 tidied the source and removed it altogether.
4263 PH/33 When a queue run was abandoned because the load average was too high, a
4264 log line was always written; now it is written only if the queue_run log
4265 selector is set. In addition, the log line for abandonment now contains
4266 information about the queue run such as the pid. This is always present
4267 in "start" and "stop" lines but was omitted from the "abandon" line.
4269 PH/34 Omit spaces between a header name and the colon in the error message that
4270 is given when verify = headers_syntax fails (if there are lots of them,
4271 the message gets confusing).
4273 PH/35 Change the default for dns_check_names_pattern to allow slashes within
4274 names, as there are now some PTR records that contain slashes. This check
4275 is only to protect against broken name servers that fall over on strange
4276 characters, so the fact that it applies to all lookups doesn't matter.
4278 PH/36 Now that the new test suite is complete, we can remove some of the
4279 special code in Exim that was needed for the old test suite. For example,
4280 sorting DNS records because real resolvers return them in an arbitrary
4281 order. The new test suite's fake resolver always returns records in the
4284 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
4285 bounces) except when queue_only is set, to avoid logging races between
4286 the different processes.
4288 PH/38 Panic-die if .include specifies a non-absolute path.
4290 PH/39 A tweak to the "H" retry rule from its user.
4292 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
4293 a label. They prevented compilation on older perls.
4295 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
4296 a warning to be raised on newish perls.
4298 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
4299 on queue. Changes to match documented behaviour of showing count of
4300 messages matching specified criteria.
4302 PH/40 Changed the default ident timeout from 30s to 5s.
4304 PH/41 Added support for the use of login_cap features, on those BSD systems
4305 that have them, for controlling the resources used by pipe deliveries.
4307 PH/42 The content-scanning code uses fopen() to create files in which to put
4308 message data. Previously it was not paying any attention to the mode of
4309 the files. Exim runs with umask(0) because the rest of the code creates
4310 files with open(), and sets the required mode explicitly. Thus, these
4311 files were ending up world-writeable. This was not a big issue, because,
4312 being within the spool directory, they were not world-accessible. I have
4313 created a function called modefopen, which takes an additional mode
4314 argument. It sets umask(777), creates the file, chmods it to the required
4315 mode, then resets the umask. All the relevant calls to fopen() in the
4316 content scanning code have been changed to use this function.
4318 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
4319 to 24 hours. This avoids potential overflow problems when processing G
4320 and H retry rules. I suspect nobody ever tinkers with this value.
4322 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
4324 PH/45 When the plaintext authenticator is running as a client, the server's
4325 challenges are checked to ensure they are valid base64 strings. By
4326 default, the authentication attempt is cancelled if an invalid string is
4327 received. Setting client_ignore_invalid_base64 true ignores these errors.
4328 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
4329 they are received. Thus, the responses can be made to depend on the
4330 challenges. If an invalid string is ignored, an empty string is placed in
4333 PH/46 Messages that are created by the autoreply transport now contains a
4334 References: header, in accordance with RFCs 2822 and 3834.
4336 PH/47 Added authenticated_sender_force to the smtp transport.
4338 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
4340 PH/49 Installed latest patch from the Sieve maintainer.
4342 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
4343 was also set, the appendfile transport was unnecessarily scanning a
4344 directory of message files (e.g. for maildir delivery) to find the count
4345 of files (along with the size), even though it did not need this
4346 information. It now does the scan only if it needs to find either the
4347 size of the count of files.
4349 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
4351 PH/52 Two bugs concerned with error handling when the smtp transport is
4354 (i) Exim was not creating retry information for temporary errors given
4355 for individual recipients after the DATA command when the smtp transport
4356 was used in LMTP mode. This meant that they could be retried too
4357 frequently, and not timed out correctly.
4359 (ii) Exim was setting the flag that allows error details to be returned
4360 for LMTP errors on RCPT commands, but not for LMTP errors for individual
4361 recipients that were returned after the DATA command.
4363 PH/53 This is related to PH/52, but is more general: for any failing address,
4364 when detailed error information was permitted to be returned to the
4365 sender, but the error was temporary, then after the final timeout, only
4366 "retry timeout exceeded" was returned. Now it returns the full error as
4367 well as "retry timeout exceeded".
4369 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
4370 do this, and (what is worse) MTAs that accept it.
4372 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
4373 will now be deprecated.
4375 PH/56 New os.c-cygwin from the Cygwin maintainer.
4377 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
4378 formats (previously only available in exim formats via -bpr, -bpru,
4379 and -bpra. Now also available in native and exiqgrep formats)
4381 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
4382 with very large, slow to parse queues
4384 JJ/08 exipick: added ! as generic prefix to negate any criteria format
4386 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
4388 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
4389 responses to authentication challenges, though it was showing the
4390 challenges; (ii) I've removed the CR characters from the debug output for
4393 PH/58 Allow for the insertion of a newline as well as a space when a string
4394 is turned into more than one encoded-word during RFC 2047 encoding. The
4395 Sieve code now uses this.
4397 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
4398 data_4xx, lost_connection, tls_required.
4400 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
4401 message was being sent as an SMTP response.
4403 PH/61 Add -l and -k options to exicyclog.
4405 PH/62 When verifying, if an address was redirected to one new address, so that
4406 verification continued, and the new address failed or deferred after
4407 having set something in $address_data, the value of $address_data was not
4408 passed back to the ACL. This was different to the case when no
4409 redirection occurred. The value is now passed back in both cases.
4411 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
4412 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
4413 use login_cap.h, so on its own it isn't the distinguishing feature. The
4414 new name refers directly to the setclassresources() function.
4416 PH/65 Added configuration files for NetBSD3.
4418 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
4420 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
4421 is preferred over IPv4.
4423 PH/68 The bounce_return_message and bounce_return_body options were not being
4424 honoured for bounces generated during the reception of non-SMTP messages.
4425 In particular, this applied to messages rejected by the ACL. This bug has
4426 been fixed. However, if bounce_return_message is true and bounce_return_
4427 body is false, the headers that are returned for a non-SMTP message
4428 include only those that have been read before the error was detected.
4429 (In the case of an ACL rejection, they have all been read.)
4431 PH/69 The HTML version of the specification is now built in a directory called
4432 spec_html instead of spec.html, because the latter looks like a path with
4433 a MIME-type, and this confuses some software.
4435 PH/70 Catch two compiler warnings in sieve.c.
4437 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
4438 function verify_get_ident() calls ip_connect() to connect a socket, but
4439 if the "connect()" function timed out, ip_connect() used to close the
4440 socket. However, verify_get_ident() also closes the socket later, and in
4441 between Exim writes to the log, which may get opened at this point. When
4442 the socket was closed in ip_connect(), the log could get the same file
4443 descriptor number as the socket. This naturally causes chaos. The fix is
4444 not to close the socket in ip_connect(); the socket should be closed by
4445 the function that creates it. There was only one place in the code where
4446 this was missing, in the iplookup router, which I don't think anybody now
4447 uses, but I've fixed it anyway.
4449 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
4450 well as to direct DNS lookups. Otherwise the handling of names in host
4451 lists is inconsistent and therefore confusing.
4457 PH/01 Two changes to the default runtime configuration:
4459 (1) Move the checks for relay_from_hosts and authenticated clients from
4460 after to before the (commented out) DNS black list checks.
4462 (2) Add control=submission to the relay_from_hosts and authenticated
4463 clients checks, on the grounds that messages accepted by these
4464 statements are most likely to be submissions.
4466 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
4468 (1) Generate an error if the third argument for the ${prvs expansion is
4471 (2) Treat a missing third argument of ${prvscheck as if it were an empty
4474 (3) Reset the variables that are obtained from the first argument of
4475 ${prvscheck and used in the second argument before leaving the code,
4476 because their memory is reclaimed, so using them afterwards may do
4479 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
4480 one (it's much easier than Tom thought :-).
4482 (5) Because of (4), we can now allow for the use of $prvscheck_result
4483 inside the third argument.
4485 PH/03 For some reason, the default setting of PATH when running a command from
4486 a pipe transport was just "/usr/bin". I have changed it to
4489 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
4490 anything to be listed in the output from -bV.
4492 PH/05 When a filter generated an autoreply, the entire To: header line was
4493 quoted in the delivery log line, like this:
4495 => >A.N.Other <ano@some.domain> <original@ddress> ...
4497 This has been changed so that it extracts the operative address. There
4498 may be more than one such address. If so, they are comma-separated, like
4501 => >ano@some.domain,ona@other.domain <original@ddress> ...
4503 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
4504 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
4505 not being looked up in the rDNS to get a host name, Exim was showing the
4506 IP address twice in Received: lines, even though the IP addresses were
4507 identical. For example:
4509 Received: from [1.2.3.4] (helo=[1.2.3.4])
4511 However, if the real host name was known, it was omitting the HELO data
4512 if it matched the actual IP address. This has been tidied up so that it
4513 doesn't show the same IP address twice.
4515 PH/07 When both +timestamp and +memory debugging was on, the value given by
4516 $tod_xxx expansions could be wrong, because the tod_stamp() function was
4517 called by the debug printing, thereby overwriting the timestamp buffer.
4518 Debugging no longer uses the tod_stamp() function when +timestamp is set.
4520 PH/08 When the original message was included in an autoreply transport, it
4521 always said "this is a copy of the message, including all the headers",
4522 even if body_only or headers_only was set. It now gives an appropriate
4525 PH/09 Applied a patch from the Sieve maintainer which:
4527 o fixes some comments
4528 o adds the (disabled) notify extension core
4529 o adds some debug output for the result of if/elsif tests
4530 o points to the current vacation draft in the documentation
4531 and documents the missing references header update
4535 o fixes a bug in processing the envelope test (when testing
4536 multiple envelope elements, the last element determined the
4539 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
4540 Electronic Mail") by including:
4542 Auto-submitted: auto-generated
4544 in the messages that it generates (bounce messages and others, such as
4545 warnings). In the case of bounce messages for non-SMTP messages, there was
4546 also a typo: it was using "Auto_submitted" (underscore instead of
4547 hyphen). Since every message generated by Exim is necessarily in response
4548 to another message, thes have all been changed to:
4550 Auto-Submitted: auto-replied
4552 in accordance with these statements in the RFC:
4554 The auto-replied keyword:
4556 - SHOULD be used on messages sent in direct response to another
4557 message by an automatic process,
4559 - MUST NOT be used on manually-generated messages,
4561 - MAY be used on Delivery Status Notifications (DSNs) and Message
4562 Disposition Notifications (MDNs),
4564 - MUST NOT be used on messages generated by automatic or periodic
4565 processes, except for messages which are automatic responses to
4568 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
4569 to the default Received: header definition.
4571 PH/12 Added log selector acl_warn_skipped (default on).
4573 PH/13 After a successful wildlsearch lookup, discard the values of numeric
4574 variables because (a) they are in the wrong storage pool and (b) even if
4575 they were copied, it wouldn't work properly because of the caching.
4577 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
4578 checking when decoding. Apparently there are clients that generate
4579 overlong encoded strings. Why am I not surprised?
4581 PH/15 If the first argument of "${if match_address" was not empty, but did not
4582 contain an "@" character, Exim crashed. Now it writes a panic log message
4583 and treats the condition as false.
4585 PH/16 In autoreply, treat an empty string for "once" the same as unset.
4587 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
4588 extension "envelope-auth". The code is finished and in agreement with
4589 other implementations, but there is no documentation so far and in fact,
4590 nobody wrote the draft yet. This extension is currently #undef'ed, thus
4591 not changing the active code.
4593 Print executed "if" and "elsif" statements when debugging is used. This
4594 helps a great deal to understand what a filter does.
4596 Document more things not specified clearly in RFC3028. I had all this
4597 sorted out, when out of a sudden new issues came to my mind. Oops."
4599 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
4602 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
4603 canonical form (as documented). However, after a host name lookup from
4604 the IP address, check_host() was doing a simple string comparison with
4605 addresses acquired from the DNS when checking that the found name did
4606 have the original IP as one of its addresses. Since any found IPv6
4607 addresses are likely to be in abbreviated form, the comparison could
4608 fail. Luckily, there already exists a function for doing the comparison
4609 by converting both addresses to binary, so now that is used instead of
4610 the text comparison.
4612 PH/20 There was another similar case to PH/19, when a complete host name was
4613 given in a host list; looking up its IP address could give an abbreviated
4614 form, whereas the current host's name might or might not be abbreviated.
4615 The same fix has been applied.
4621 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
4622 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
4625 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
4626 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
4628 PH/03 Typo: missing ".o" in src/pcre/Makefile.
4630 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
4631 header line, restrict the check to what is listed in RFCs 2369 and 2929.
4632 Also, for "Auto-Submitted", treat anything other than "no" as
4633 non-personal, in accordance with RFC 3834. (Previously it treated
4634 anything starting "auto-" as non-personal.)
4636 TF/01 The control=submission/name=... option had a problem with syntax
4637 errors if the name included a slash character. The /name= option
4638 now slurps the rest of the string, so it can include any characters
4639 but it must come last in the list of options (after /sender_retain
4642 PH/05 Some modifications to the interface to the fake nameserver for the new
4650 TK/01 Added the "success_on_redirect" address verification option. See
4651 NewStuff for rationale and an example.
4653 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
4655 PH/02 Patch to exigrep to allow it to work on syslog lines.
4657 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
4658 fread() to skip over the body file's header line, because in Cygwin the
4659 header line is locked and is inaccessible.
4661 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
4662 co-exist for some time) to make it clear that it is the Exim ID that is
4663 referenced, not the Message-ID: header line.
4665 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
4666 string_format() function, because snprintf() does not exist on all
4669 PH/06 The use of forbid_filter_existstest now also locks out the use of the
4670 ${stat: expansion item.
4672 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
4673 protocol synchronization error", to keep the pedants happy.
4675 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
4676 well as for IRIX systems, when gcc is being used. See the host.c source
4679 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
4681 PH/10 Named domain lists were not working if used in a queue_smtp_domains
4684 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
4685 transport and to the smtp transport in LMTP mode.
4687 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
4689 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
4690 run a filter in a subprocess. This could lead to confusion in subsequent
4691 lookups in the parent process. There should also be a search_tidyup() at
4692 the end of the subprocess.
4694 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
4695 only if the host matched helo_try_verify_hosts, which caused the
4696 verification to occur when the EHLO/HELO command was issued. The ACL just
4697 tested the remembered result. Now, if a previous verification attempt has
4698 not happened, "verify = helo" does it there and then.
4700 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
4702 TK/03 Fix log output including CR from clamd.
4704 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
4705 exist provoked a memory error which could cause a segfault.
4707 PH/15 Installed PCRE 6.2
4709 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
4711 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
4712 of the problem. Specifically, suggested +O2 rather than +O1 for the
4715 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
4717 PH/20 If a delivery was routed to a non-standard port by means of an SRV
4718 record, the port was not correctly logged when the outgoing_port log
4719 selector was set (it logged the transort's default port).
4721 PH/21 Added support for host-specific ports to manualroute, queryprogram,
4722 fallback_hosts, and "hosts" in the smtp transport.
4724 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
4725 host errors such as "Connection refused".
4727 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
4728 authentication with radiusclient 0.4.9:
4730 - Error returned from rc_read_config was caught wrongly
4731 - Username/password not passed on to radius server due to wrong length.
4733 The presumption is that some radiusclient API changes for 4.51/PH/17
4734 were not taken care of correctly. The code is still untested by me (my
4735 Linux distribution still has 0.3.2 of radiusclient), but it was
4736 contributed by a Radius user.
4738 PH/24 When doing a callout, the value of $domain wasn't set correctly when
4739 expanding the "port" option of the smtp transport.
4741 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
4742 while reading a MIME header. Thanks to Tom Hughes for a patch.
4744 PH/24 Include config.h inside local_scan.h so that configuration settings are
4747 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
4748 that in an ACL for RCPT (for example), you can examine exactly what was
4751 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
4752 commands, but it was not correctly comparing the address with the actual
4753 client host address. Thus, it would show the EHLO address in Received:
4754 header lines when this was not necessary.
4756 PH/27 Added the % operator to ${eval:}.
4758 PH/28 Exim tries to create and chdir to its spool directory when it starts;
4759 it should be ignoring failures (because with -C, for example, it has lost
4760 privilege). It wasn't ignoring creation failures other than "already
4763 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
4766 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
4767 because an input line was too long, either on its own, or by virtue of
4768 too many continuations, the temporary file was not being removed, and the
4769 return code was incorrect.
4771 PH/31 Missing "BOOL" in function definition in filtertest.c.
4773 PH/32 Applied Sieve patches from the maintainer.
4775 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
4777 PH/33 Added "verify = not_blind".
4779 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
4780 Local/Makefile (with some defaults set). These are used in built scripts
4781 such as exicyclog, but they have never been used in the exim_install
4782 script (though there are many overriding facilities there). I have
4783 arranged that the exim_install script now takes note of these two
4786 PH/35 Installed configuration files for Dragonfly.
4788 PH/36 When a locally submitted message by a trusted user did not contain a
4789 From: header, and the sender address was obtained from -f or from an SMTP
4790 MAIL command, and the trusted user did not use -F to supply a sender
4791 name, $originator_name was incorrectly used when constructing a From:
4792 header. Furthermore, $originator_name was used for submission mode
4793 messages from external hosts without From: headers in a similar way,
4794 which is clearly wrong.
4796 PH/37 Added control=suppress_local_fixups.
4798 PH/38 When log_selector = +received_sender was set, and the addition of the
4799 sender made the log line's construction buffer exactly full, or one byte
4800 less than full, an overflow happened when the terminating "\n" was
4803 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
4804 when the result of a list match is failure because a DNS lookup failed.
4806 PH/40 RM_COMMAND is now used in the building process.
4808 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
4809 the "build-* directories that it finds.
4811 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
4812 address in a domain literal was a prefix of an interface address.
4814 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
4815 when verifying a sender address, unless rewrite_headers is false.
4817 PH/44 Wrote a long comment about why errors_to addresses are verified as
4818 recipients, not senders.
4820 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
4821 the ratelimit ACL was added.
4823 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
4825 PH/46 Added extra information about PostgreSQL errors to the error string.
4827 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
4828 avoiding the need to install special zones in a real server. This is
4829 backwards compatible; if it can't find the fake resolver, it drops back.
4830 Thus, both old and new test suites can be run.
4832 TF/02 Added util/ratelimit.pl
4834 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
4835 clock is set back in time.
4837 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
4838 Candler <B.Candler@pobox.com>.
4840 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
4841 OK for addresses that are the result of redirections.
4843 PH/48 A number of further additions for the benefit of the new test suite,
4844 including a fake gethostbyname() that interfaces to the fake DNS resolver
4847 TF/06 The fix for widen_domains has also been applied to qualify_single and
4848 search_parents which are the other dnslookup options that can cause
4851 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
4854 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
4856 TF/07 Exim produced the error message "an SRV record indicated no SMTP
4857 service" if it encountered an MX record with an empty target hostname.
4858 The message is now "an MX or SRV record indicated no SMTP service".
4860 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
4861 if the DNS of the sending site is misconfigured. This is quite a
4862 common situation. This change restores the behaviour of treating a
4863 helo verification defer as a failure.
4865 PH/51 If self=fail was set on a router, the bounce message did not include the
4866 actual error message.
4872 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
4874 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4875 command itself ended in error, the underlying message about the transport
4876 filter timeout was being overwritten with the pipe command error. Now the
4877 underlying error message should be appended to the second error message.
4879 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4881 PH/02 Reduce the amount of output that "make" produces by default. Full output
4882 can still be requested.
4884 PH/03 The warning log line about a condition test deferring for a "warn" verb
4885 was being output only once per connection, rather than after each
4886 occurrence (because it was using the same function as for successful
4887 "warn" verbs). This seems wrong, so I have changed it.
4889 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4890 it should not have, which might have caused a crash in the right
4891 circumstances, but probably never did.
4893 PH/04 Installed a modified version of Tony Finch's patch to make submission
4894 mode fix the return path as well as the Sender: header line, and to
4895 add a /name= option so that you can make the user's friendly name appear
4898 TF/03 Added the control = fakedefer ACL modifier.
4900 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4901 Mark Lowes for thorough testing.
4903 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4905 TK/03 Merged latest SRS patch from Miles Wilton.
4907 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4908 with the definition in sysexits.h (which is #included earlier).
4909 Fortunately, Exim does not actually use EX_OK. The code used to try to
4910 preserve the sysexits.h value, by assuming that macro definitions were
4911 scanned for macro replacements. I have been disabused of this notion,
4912 so now the code just undefines EX_OK before #including unistd.h.
4914 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4915 in the smtp transport. When a block could not be written in a single
4916 write() function, the timeout was being re-applied to each part-write.
4917 This seems wrong - if the receiver was accepting one byte at a time it
4918 would take for ever. The timeout is now adjusted when this happens. It
4919 doesn't have to be particularly precise.
4921 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4922 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4924 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4925 without a domain if the check to <postmaster@domain> fails.
4927 SC/01 Eximstats: added -xls and the ability to specify output files
4928 (patch written by Frank Heydlauf).
4930 SC/02 Eximstats: use FileHandles for outputting results.
4932 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4934 SC/04 Eximstats: fixed display of large numbers with -nvr option
4936 SC/05 Eximstats: fixed merging of reports with empty tables.
4938 SC/06 Eximstats: added the -include_original_destination flag
4940 SC/07 Eximstats: removed tabs and trailing whitespace.
4942 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4944 TK/06 MBOX spool code: Add real "From " MBOX separator line
4945 so the .eml file is really in mbox format (even though
4946 most programs do not really care). Patch from Alex Miller.
4948 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4949 The latter is generated from $received_to and is only set if the
4950 message has one envelope recipient. SA can use these headers,
4951 obviously out-of-the-box. Patch from Alex Miller.
4953 PH/08 The ${def test on a variable was returning false if the variable's
4954 value was "0", contrary to what the specification has always said!
4955 The result should be true unless the variable is empty.
4957 PH/09 The syntax error of a character other than { following "${if
4958 def:variable_name" (after optional whitespace) was not being diagnosed.
4959 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4960 accidental colon was present, for example, could give incorrect results.
4962 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4963 result is used (not including appendfile, where other changes are about
4966 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4967 This involved changing a lot of size variables from int to off_t. It
4968 should work with maildirs and everything.
4970 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4971 spamd dying while we are connected to it.
4973 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4976 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4977 function for BDB 4.3.
4979 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4981 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4982 the sources for PCRE have been re-arranged and more files are now
4985 PH/15 The code I had for printing potentially long long variables in PH/11
4986 above was not the best (it lost precision). The length of off_t variables
4987 is now inspected at build time, and an appropriate printing format (%ld
4988 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4989 to be "long long int" or "long int". This is needed for the internal
4990 formatting function string_vformat().
4992 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4993 the configuration file to be ":syslog", then the script "guesses" where
4994 the logs files are, rather than using the compiled in default. In our
4995 case the guess is not the same as the compiled default, so the script
4996 suddenly stopped working when I started to use syslog. The patch checks
4997 to see if log_file_path is "". If so, it attempts to read it from exim
4998 with no configuration file to get the compiled in version, before it
4999 falls back to the previous guessing code."
5001 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
5002 implementing BATV in an Exim configuration. See NewStuff for the gory
5005 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
5006 Makefile that are specific to HP-UX.
5008 PH/18 If the "use_postmaster" option was set for a recipient callout together
5009 with the "random" option, the postmaster address was used as the MAIL
5010 FROM address for the random test, but not for the subsequent recipient
5011 test. It is now used for both.
5013 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
5014 patch removes a few documentation additions to RFC 3028, because the
5015 latest draft now contains them. It adds the new en;ascii-case comparator
5016 and a new error check for 8bit text in MIME parts. Comparator and
5017 require names are now matched exactly. I enabled the subaddress
5018 extension, but it is not well tested yet (read: it works for me)."
5020 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
5021 rework some of the code of TK/09 above to avoid the hardwired use of
5022 "%lld" and "long long". Replaced the call to snprintf() with a call to
5025 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
5026 records point to non-existent hosts", "retry timeout exceeded", and
5027 "retry time not reached for any host after a long failure period".
5029 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
5030 experimental DomainKeys support:
5032 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
5033 (2) On an error such as an illegally used "control", the wrong name for
5034 the control was given.
5036 These problems did NOT occur unless DomainKeys support was compiled.
5038 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
5040 PH/24 Added ${if match_ip condition.
5042 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
5043 databases so that it will be absolutely obvious if a crash occurs in the
5044 DB library. This is a regular occurrence (often caused by mis-matched
5047 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
5048 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
5049 picked up on a user's system that detects such things. There doesn't seem
5050 to be a gcc warning option for this - only an attribute that has to be
5051 put on the function's prototype. It seems that in Fedora Core 4 they have
5052 set this on a number of new functions. No doubt there will be more in due
5055 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
5056 specify a transport. However, if an address that was verified by such a
5057 router was the subject of a callout, Exim crashed because it tried to
5058 read the rcpt_include_affixes from the non-existent transport. Now it
5059 just assumes that the setting of that option is false. This bug was
5060 introduced by 4.51/PH/31.
5062 PH/28 Changed -d+all to exclude +memory, because that information is very
5063 rarely of interest, but it makes the output a lot bigger. People tend to
5064 do -d+all out of habit.
5066 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
5067 code in os-type was giving problems when libc.so lives in lib64, like on
5070 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
5071 aren't the modern standard, and it seems that some systems' include files
5072 don't always have them. Exim was already checking for some of the newer
5073 ones like T_AAAA, and defining it itself. I've added checks for all the
5074 record types that Exim uses.
5076 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
5077 not automatically generating a new one, as it is supposed to. This
5078 prevented TLS from working. If the file did exist, but contained invalid
5079 data, a new version was generated, as expected. It was only the case of a
5080 non-existent file that was broken.
5082 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
5083 with a change in libdomainkeys > 0.64.
5085 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
5086 from DNS. If the selector record carries the flag, it now has
5087 precedence over the domain-wide flag.
5089 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
5091 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
5092 the use of a port name that isn't defined in /etc/services) occurred, the
5093 message was deferred as in a normal delivery, and thus remained on the
5094 spool, instead of being failed because of the mua_wrapper setting. This
5095 is now fixed, and I tidied up some of the mua_wrapper messages at the
5098 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
5099 the messages in a hash of arrays rather than using individual hashes.
5100 This is a bit cleaner and results in dramatic memory savings, albeit
5101 at a slight CPU cost.
5103 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
5104 as requested by Marc Sherman.
5106 SC/10 Eximstats: added histograms for user specified patterns as requested
5109 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
5111 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
5112 fopen() in the content-scanning modules that did not already have it.
5118 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
5119 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
5121 TK/02 Fix ACL "control" statement not being available in MIME ACL.
5123 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
5125 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
5126 to test Sieve filters that use "vacation".
5128 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
5129 that changes the way the GnuTLS parameters are stored in the cache file.
5130 The new format can be generated externally. For backward compatibility,
5131 if the data in the cache doesn't make sense, Exim assumes it has read an
5132 old-format file, and it generates new data and writes a new file. This
5133 means that you can't go back to an older release without removing the
5136 PH/03 A redirect router that has both "unseen" and "one_time" set does not
5137 work if there are any delivery delays because "one_time" forces the
5138 parent to be marked "delivered", so its unseen clone is never tried
5139 again. For this reason, Exim now forbids the simultaneous setting of
5142 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
5143 redirected to themselves ("homonym" addresses). Read the long ChangeLog
5144 entry if you want to know the details. The fix, however, neglected to
5145 consider the case when local delivery batching is involved. The test for
5146 "previously delivered" was not happening when checking to see if an
5147 address could be batched with a previous (undelivered) one; under
5148 certain circumstances this could lead to multiple deliveries to the same
5151 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
5152 in its include files, and this causes problems building Exim.
5154 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
5155 header_syntax) but Exim was just ignoring anything given after a slash.
5156 In particular, this caused confusion with an attempt to use "verify =
5157 reverse_host_lookup/defer_ok". An error is now given when options are
5158 supplied for verify items that do not have them. (Maybe reverse_host_
5159 lookup should have a defer_ok option, but that's a different point.)
5161 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
5162 defined by RFC 821) to 2048, because there were problems with some AUTH
5163 commands, and RFC 1869 says the size should be increased for extended
5164 SMTP commands that take arguments.
5166 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
5169 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
5170 "unknown" error; now it says that the functionality isn't in the binary.
5172 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
5173 an address' error message when a string expansion fails (syntax or
5174 whatever). Otherwise the password may appear in the log. Following change
5175 PH/42 below, there is no longer a chance of it appearing in a bounce
5178 PH/11 Installed exipick version 20050225.0 from John Jetmore.
5180 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
5181 of its addresses was ever tried. (Bugzilla bug #2.)
5183 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
5184 the result incorrectly in the debug output. (It correctly added a newline
5185 to what was transported.)
5187 TF/01 Added $received_time.
5189 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
5190 commented out examples of how to interface to a virus scanner and to
5191 SpamAssassin. Also added commented examples of av_scanner and
5192 spamd_address settings.
5194 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
5195 and controls are allowed in which ACLs. There were a couple of minor
5196 errors. Some of the entries in the conditions table (which is a table of
5197 where they are NOT allowed) were getting very unwieldy; rewrote them as a
5198 negation of where the condition IS allowed.
5200 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
5202 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
5203 header file does not have a version number, so I've had to invent a new
5204 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
5205 API. The code is untested by me (my Linux distribution still has 0.3.2 of
5206 radiusclient), but it was contributed by a Radius user.
5208 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
5209 files or directories, for OpenSSL.
5211 PH/19 When an Exim process that is running as root has to create an Exim log
5212 file, it does so in a subprocess that runs as exim:exim so as to get the
5213 ownership right at creation (otherwise, other Exim processes might see
5214 the file with the wrong ownership). There was no test for failure of this
5215 fork() call, which would lead to the process getting stuck as it waited
5216 for a non-existent subprocess. Forks do occasionally fail when resources
5217 run out. I reviewed all the other calls to fork(); they all seem to check
5220 PH/20 When checking for unexpected SMTP input at connect time (before writing
5221 the banner), Exim was not dealing correctly with a non-positive return
5222 from the read() function. If the client had disconnected by this time,
5223 the result was a log entry for a synchronization error with an empty
5224 string after "input=" when read() returned zero. If read() returned -1
5225 (an event I could not check), uninitialized data bytes were printed.
5226 There were reports of junk text (parts of files, etc) appearing after
5229 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
5231 PH/22 Added support for macro redefinition, and (re)definition in between
5232 driver and ACL definitions.
5234 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
5235 forgetting to use the resulting value; it was using the unexpanded value.
5237 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
5238 hadn't been configured. The fix is from Juergen Kreileder, who
5239 understands it better than I do:
5241 "Here's what I see happening with three configured cyrus_sasl
5242 authenticators configured (plain, login, cram-md5):
5244 On startup auth_cyrus_sasl_init() gets called for each of these.
5245 This means three calls to sasl_listmech() without a specified mech_list.
5246 => SASL tests which mechs of all available mechs actually work
5247 => three warnings about OTP not working
5248 => the returned list contains: plain, login, cram-md5, digest-md5, ...
5250 With the patch, sasl_listmech() also gets called three times. But now
5251 SASL's mech_list option is set to the server_mech specified in the the
5252 authenticator. Or in other words, the answer from sasl_listmech()
5253 gets limited to just the mech you're testing for (which is different
5255 => the return list contains just 'plain' or 'login', 'cram-md5' or
5256 nothing depending on the value of ob->server_mech.
5258 I've just tested the patch: Authentication still works fine,
5259 unavailable mechs specified in the exim configuration are still
5260 caught, and the auth.log warnings about OTP are gone."
5262 PH/25 When debugging is enabled, the contents of the command line are added
5263 to the debugging output, even when log_selector=+arguments is not
5266 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
5267 answer is "GNU", and only if the return is "GNU/something" is the answer
5270 PH/27 $acl_verify_message is now set immediately after the failure of a
5271 verification in an ACL, and so is available in subsequent modifiers. In
5272 particular, the message can be preserved by coding like this:
5274 warn !verify = sender
5275 set acl_m0 = $acl_verify_message
5277 Previously, $acl_verify_message was set only while expanding "message"
5278 and "log_message" when a very denied access.
5280 PH/28 Modified OS/os.c-Linux with
5282 -#ifndef OS_LOAD_AVERAGE
5283 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
5285 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
5286 nomenclature these days.)
5288 PH/29 Installed patch from the Sieve maintainer that adds the options
5289 sieve_useraddress and sieve_subaddress to the redirect router.
5291 PH/30 In these circumstances:
5292 . Two addresses routed to the same list of hosts;
5293 . First host does not offer TLS;
5294 . First host accepts first address;
5295 . First host gives temporary error to second address;
5296 . Second host offers TLS and a TLS session is established;
5297 . Second host accepts second address.
5298 Exim incorrectly logged both deliveries with the TLS parameters (cipher
5299 and peerdn, if requested) that were in fact used only for the second
5302 PH/31 When doing a callout as part of verifying an address, Exim was not paying
5303 attention to any local part prefix or suffix that was matched by the
5304 router that accepted the address. It now behaves in the same way as it
5305 does for delivery: the affixes are removed from the local part unless
5306 rcpt_include_affixes is set on the transport.
5308 PH/32 Add the sender address, as F=<...>, to the log line when logging a
5309 timeout during the DATA phase of an incoming message.
5311 PH/33 Sieve envelope tests were broken for match types other than :is. I have
5312 applied a patch sanctioned by the Sieve maintainer.
5314 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
5315 the uid or gid is negative. A case of a negative gid caused this to be
5316 noticed. The fix allows for either to be negative.
5318 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
5319 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
5322 PH/36 The change PH/12 above was broken. Fixed it.
5324 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
5325 the grounds that routing the same address twice would always produce the
5326 same answer. This might have been true once, but it is certainly no
5327 longer true now. Routing a child address may depend on the previous
5328 routing that produced that child. Some complicated redirection strategies
5329 went wrong when messages had multiple recipients, and made Exim's
5330 behaviour dependent on the order in which the addresses were given.
5332 I have moved the duplicate checking until after the routing is complete.
5333 Exim scans the addresses that are assigned to local and remote
5334 transports, and removes any duplicates. This means that more work will be
5335 done, as duplicates will always all be routed, but duplicates are
5336 presumably rare, so I don't expect this is of any significance.
5338 For deliveries to pipes, files, and autoreplies, the duplicate checking
5339 still happens during the routing process, since they are not going to be
5342 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
5343 It corrects a timeout issue with spamd. This is Ian's comment: "The
5344 background is that sometimes spamd either never reads data from a
5345 connection it has accepted, or it never writes response data. The exiscan
5346 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
5347 blindly assumes that writes won't block so it may never time out."
5349 PH/39 Allow G after quota size as well as K and M.
5351 PH/40 The value set for $authenticated_id in an authenticator may not contain
5352 binary zeroes or newlines because the value is written to log lines and
5353 to spool files. There was no check on this. Now the value is run through
5354 the string_printing() function so that such characters are converted to
5355 printable escape sequences.
5357 PH/41 $message_linecount is a new variable that contains the total number of
5358 lines in the message. Compare $body_linecount, which is the count for the
5361 PH/42 Exim no longer gives details of delivery errors for specific addresses in
5362 bounce and delay warning messages, except in certain special cases, which
5365 (a) An SMTP error message from a remote host;
5366 (b) A message specified in a :fail: redirection;
5367 (c) A message specified in a "fail" command in a system filter;
5368 (d) A message specified in a FAIL return from the queryprogram router;
5369 (e) A message specified by the cannot_route_message router option.
5371 In these cases only, Exim does include the error details in bounce and
5372 warning messages. There are also a few cases where bland messages such
5373 as "unrouteable address" or "local delivery error" are given.
5375 PH/43 $value is now also set for the "else" part of a ${run expansion.
5377 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
5378 being worked on, but at least Exim now implements the latest version to
5381 PH/45 In a pipe transport, although a timeout while waiting for the pipe
5382 process to complete was treated as a delivery failure, a timeout while
5383 writing the message to the pipe was logged, but erroneously treated as a
5384 successful delivery. Such timeouts include transport filter timeouts. For
5385 consistency with the overall process timeout, these timeouts are now
5386 treated as errors, giving rise to delivery failures by default. However,
5387 there is now a new Boolean option for the pipe transport called
5388 timeout_defer, which, if set TRUE, converts the failures into defers for
5389 both kinds of timeout. A transport filter timeout is now identified in
5392 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
5393 systems where "make" and "gmake" are different, calling "gmake" at top
5394 level broke things. I've arranged for the value of $(MAKE) to be passed
5395 from the Makefile to this script so that it can call the same version of
5399 A note about Exim versions 4.44 and 4.50
5400 ----------------------------------------
5402 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
5403 changes of various kinds. As a consequence, a big documentation update was
5404 needed. This delayed the release for rather longer than seemed good, especially
5405 in the light of a couple of (minor) security issues. Therefore, the changes
5406 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
5407 release. So 4.44 and 4.50 are in effect two different branches that both start
5410 I have left the 4.50 change log unchanged; it contains all the changes since
5411 4.43. The change log for 4.44 is below; many of its items are identical to
5412 those for 4.50. This seems to be the most sensible way to preserve the
5413 historical information.
5419 1. Minor wording change to the doc/README.SIEVE file.
5421 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
5422 computation of the current number of files was incorrect.
5424 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5425 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5428 4. Give more explanation in the error message when the command for a transport
5429 filter fails to execute.
5431 5. There are several places where Exim runs a non-Exim command in a
5432 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5433 was being done only for the command run by the queryprogram router. It is
5434 now done for all such subprocesses. The other cases are: ${run, transport
5435 filters, and the commands run by the lmtp and pipe transports.
5437 6. Added CONFIGURE_GROUP build-time option.
5439 7. Some older OS have a limit of 256 on the maximum number of file
5440 descriptors. Exim was using setrlimit() to set 1000 as a large value
5441 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5442 systems. I've change it so that if it can't get 1000, it tries for 256.
5444 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5445 was an oversight, and furthermore, ever since the addition of extra
5446 controls (e.g. 4.43/32), the checks on when to allow different forms of
5447 "control" were broken. There should now be diagnostics for all cases when a
5448 control that does not make sense is encountered.
5450 9. Added the /retain_sender option to "control=submission".
5452 10. $recipients is now available in the predata ACL (oversight).
5454 11. Tidy the search cache before the fork to do a delivery from a message
5455 received from the command line. Otherwise the child will trigger a lookup
5456 failure and thereby defer the delivery if it tries to use (for example) a
5457 cached ldap connection that the parent has called unbind on.
5459 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5460 of $address_data from the recipient verification was clobbered by the
5461 sender verification.
5463 13. The value of address_data from a sender verification is now available in
5464 $sender_address_data in subsequent conditions in the ACL statement.
5466 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
5468 15. Added a new option "connect=<time>" to callout options, to set a different
5471 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5472 was its contents. (It was OK if the option was not defined at all.)
5474 17. A "Completed" log line is now written for messages that are removed from
5475 the spool by the -Mrm option.
5477 18. New variables $sender_verify_failure and $recipient_verify_failure contain
5478 information about exactly what failed.
5480 19. Added -dd to debug only the daemon process.
5482 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
5483 handles timeouts, both on the server side and network timeouts. Renamed the
5484 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
5486 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
5487 It is now set to "smtps".
5489 22. $host_address is now set to the target address during the checking of
5490 ignore_target_hosts.
5492 23. When checking ignore_target_hosts for an ipliteral router, no host name was
5493 being passed; this would have caused $sender_host_name to have been used if
5494 matching the list had actually called for a host name (not very likely,
5495 since this list is usually IP addresses). A host name is now passed as
5498 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5499 code that specifies a non-restarting handler (typically sigaction() in
5500 modern systems) in an attempt to fix a rare and obscure crash bug.
5502 25. Narrowed the window for a race in the daemon that could cause it to ignore
5503 SIGCHLD signals. This is not a major problem, because they are used only to
5504 wake it up if nothing else does.
5506 26. A malformed maildirsize file could cause Exim to calculate negative values
5507 for the mailbox size or file count. Odd effects could occur as a result.
5508 The maildirsize information is now recalculated if the size or filecount
5511 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5512 support for a long time. Removed HAVE_SYS_VFS_H.
5514 28. Installed the latest version of exipick from John Jetmore.
5516 29. In an address list, if the pattern was not a regular expression, an empty
5517 subject address (from a bounce message) matched only if the pattern was an
5518 empty string. Non-empty patterns were not even tested. This was the wrong
5519 because it is perfectly reasonable to use an empty address as part of a
5520 database query. An empty address is now tested by patterns that are
5521 lookups. However, all the other forms of pattern expect the subject to
5522 contain a local part and a domain, and therefore, for them, an empty
5523 address still always fails if the pattern is not itself empty.
5525 30. Exim went into a mad DNS loop when attempting to do a callout where the
5526 host was specified on an smtp transport, and looking it up yielded more
5527 than one IP address.
5529 31. Re-factored the code for checking spool and log partition space into a
5530 function that finds that data and another that does the check. The former
5531 is then used to implement four new variables: $spool_space, $log_space,
5532 $spool_inodes, and $log_inodes.
5534 32. The RFC2047 encoding function was originally intended for short strings
5535 such as real names; it was not keeping to the 75-character limit for
5536 encoded words that the RFC imposes. It now respects the limit, and
5537 generates multiple encoded words if necessary. To be on the safe side, I
5538 have increased the buffer size for the ${rfc2047: expansion operator from
5541 33. It is now permitted to omit both strings after an "if" condition; if the
5542 condition is true, the result is "true". As before, when the second string
5543 is omitted, a false condition yields an empty string. This makes it less
5544 cumbersome to write custom ACL and router conditions.
5546 34. Failure to deliver a bounce message always caused it to be frozen, even if
5547 there was an errors_to setting on the router. The errors_to setting is now
5550 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5551 canonical form (fully expanded) before being placed in
5552 $sender_host_address.
5554 36. The table in the code that translates DNS record types into text (T_A to
5555 "A" for instance) was missing entries for NS and CNAME. It is just possible
5556 that this could have caused confusion if both these types were looked up
5557 for the same domain, because the text type is used as part of Exim's
5558 per-process caching. But the chance of anyone hitting this buglet seems
5561 37. The dnsdb lookup has been extended in a number of ways.
5563 (1) There is a new type, "zns", which walks up the domain tree until it
5564 finds some nameserver records. It should be used with care.
5566 (2) There is a new type, "mxh", which is like "mx" except that it returns
5567 just the host names, not the priorities.
5569 (3) It is now possible to give a list of domains (or IP addresses) to be
5570 looked up. The behaviour when one of the lookups defers can be
5571 controlled by a keyword.
5573 (4) It is now possible to specify the separator character for use when
5574 multiple records are returned.
5576 38. The dnslists ACL condition has been extended: it is now possible to supply
5577 a list of IP addresses and/or domains to be looked up in a particular DNS
5580 39. Added log_selector=+queue_time_overall.
5582 40. When running the queue in the test harness, wait just a tad after forking a
5583 delivery process, to get repeatability of debugging output.
5585 41. Include certificate and key file names in error message when GnuTLS fails
5586 to set them up, because the GnuTLS error message doesn't include the name
5587 of the failing file when there is a problem reading it.
5589 42. Allow both -bf and -bF in the same test run.
5591 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
5593 44. The "Exiscan patch" is now merged into the mainline Exim source.
5595 45. Sometimes the final signoff response after QUIT could fail to get
5596 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5597 < 0 before doing a fflush(). This bug looks as though it goes back to the
5598 introduction of TLS in release 3.20, but "sometimes" must have been rare
5599 because the tests only now provoked it.
5601 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
5602 (this can affect the format of dates).
5604 47. exim_tidydb, when checking for the continued existence of a message for
5605 which it has found a message-specific retry record, was not finding
5606 messages that were in split spool directories. Consequently, it was
5607 deleting retry records that should have stayed in existence.
5609 48. Steve fixed some bugs in eximstats.
5611 49. The SPA authentication driver was not abandoning authentication and moving
5612 on to the next authenticator when an expansion was forced to fail,
5613 contradicting the general specification for all authenticators. Instead it
5614 was generating a temporary error. It now behaves as specified.
5616 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
5617 (the order specifies the preference for clients). The order is now AES256,
5618 AES128, 3DES, ARCFOUR128.
5620 51. Small patch to Sieve code - explicitly set From: when generating an
5623 52. Exim crashed if a remote delivery caused a very long error message to be
5624 recorded - for instance if somebody sent an entire SpamAssassin report back
5625 as a large number of 550 error lines. This bug was coincidentally fixed by
5626 increasing the size of one of Exim's internal buffers (big_buffer) that
5627 happened as part of the Exiscan merge. However, to be on the safe side, I
5628 have made the code more robust (and fixed the comments that describe what
5631 53. Now that there can be additional text after "Completed" in log lines (if
5632 the queue_time_overall log selector is set), a one-byte patch to exigrep
5633 was needed to allow it to recognize "Completed" as not the last thing in
5636 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
5637 patch that reportedly fixes this has been added. I am not expert enough to
5638 create a test for it. This is what the patch creator wrote:
5640 "I found a little strange behaviour of ldap code when working with
5641 Windows 2003 AD Domain, where users was placed in more than one
5642 Organization Units. When I tried to give exim partial DN, the exit code
5643 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
5644 But simultaneously result of request was absolutely normal ldap result,
5645 so I produce this patch..."
5647 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
5648 REFERENCE, so I have modified the code to exclude the patch when that macro
5651 55. Some experimental protocols are using DNS PTR records for new purposes. The
5652 keys for these records are domain names, not reversed IP addresses. The
5653 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5654 leaves it alone. Component reversal etc. now happens only for IP addresses.
5657 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
5659 57. Double the size of the debug message buffer (to 2048) so that more of very
5660 long debug lines gets shown.
5662 58. The exicyclog utility now does better if the number of log files to keep
5663 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
5665 59. Two changes related to the smtp_active_hostname option:
5667 (1) $smtp_active_hostname is now available as a variable.
5668 (2) The default for smtp_banner uses $smtp_active_hostname instead
5669 of $primary_hostname.
5671 60. The host_aton() function is supposed to be passed a string that is known
5672 to be a valid IP address. However, in the case of IPv6 addresses, it was
5673 not checking this. This is a hostage to fortune. Exim now panics and dies
5674 if the condition is not met. A case was found where this could be provoked
5675 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5676 components; fortuitously, this particular loophole had already been fixed
5677 by change 4.50/55 above.
5679 If there are any other similar loopholes, the new check in host_aton()
5680 itself should stop them being exploited. The report I received stated that
5681 data on the command line could provoke the exploit when Exim was running as
5682 exim, but did not say which command line option was involved. All I could
5683 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5684 running as the user.
5687 61. There was a buffer overflow vulnerability in the SPA authentication code
5688 (which came originally from the Samba project). I have added a test to the
5689 spa_base64_to_bits() function which I hope fixes it.
5692 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
5693 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
5695 63. The daemon start-up calls getloadavg() while still root for those OS that
5696 need the first call to be done as root, but it missed one case: when
5697 deliver_queue_load_max is set with deliver_drop_privilege. This is
5698 necessary for the benefit of the queue runner, because there is no re-exec
5699 when deliver_drop_privilege is set.
5701 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
5702 This has been fixed.
5704 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5705 was in use, was not putting the data itself into the right store pool;
5706 consequently, it could be overwritten for a subsequent message in the same
5707 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5710 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
5712 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
5713 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
5714 the function were treating the return as a boolean value, which happened to
5715 work because 0=false and not-0=true, but is not correct code.
5717 68. The host_aton() function was not handling scoped IPv6 addresses (those
5718 with, for example, "%eth0" on the end) correctly.
5720 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
5721 negated items (that is, ~something) in unsigned ints. Some compilers
5722 apparently mutter when there is no cast.
5724 70. If an address verification called from an ACL failed, and did not produce a
5725 user-specific message (i.e. there was only a "system" message), nothing was
5726 put in $acl_verify_message. In this situation, it now puts the system
5729 71. Change 4.23/11 added synchronization checking at the start of an SMTP
5730 session; change 4.31/43 added the unwanted input to the log line - except
5731 that it did not do this in the start of session case. It now does.
5733 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
5734 This is wrong and can cause the other end to generate a synchronization
5735 error if it is another Exim or anything else that does the synchronization
5736 check. A QUIT command is no longer sent after a timeout.
5738 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
5739 during host lookups.
5741 74. The defer_ok option of callout verification was not working if it was used
5742 when verifying addresses in header lines, that is, for this case:
5744 verify = header_sender/callout=defer_ok
5746 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
5747 those file descriptors could be used for SMTP connections. If anything
5748 wrote to stderr (the example that came up was "warn" in embedded Perl), it
5749 could be sent to the SMTP client, causing chaos. The daemon now opens
5750 stdin, stdout, and stderr to /dev/null when it puts itself into the
5753 76. Arrange for output from Perl's "warn" command to be written to Exim's main
5754 log by default. The user can override this with suitable Perl magic.
5756 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
5757 the log message when discard triggers, was not working for the DATA ACL or
5758 for the non-SMTP ACL.
5760 78. Error message wording change in sieve.c.
5762 79. If smtp_accept_max_per_host was set, the number of connections could be
5763 restricted to fewer than expected, because the daemon was trying to set up
5764 a new connection before checking whether the processes handling previous
5765 connections had finished. The check for completed processes is now done
5766 earlier. On busy systems, this bug wouldn't be noticed because something
5767 else would have woken the daemon, and it would have reaped the completed
5770 80. If a message was submitted locally by a user whose login name contained one
5771 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
5772 It caused a spool format error. I have fixed the spool reading code. A
5773 related problem was that the "from" clause in the Received: line became
5774 illegal because of the space(s). It is now covered by ${quote_local_part.
5776 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
5779 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
5781 83. Patch from Sieve maintainer to fix unterminated string problem in
5782 "vacation" handling.
5784 84. Some minor changes to the Linux configuration files to help with other
5785 OS variants using glibc.
5787 85. One more patch for Sieve to update vacation handling to latest spec.
5790 ----------------------------------------------------
5791 See the note above about the 4.44 and 4.50 releases.
5792 ----------------------------------------------------
5798 1. Change 4.43/35 introduced a bug that caused file counts to be
5799 incorrectly computed when quota_filecount was set in an appendfile
5802 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5803 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5806 3. Give more explanation in the error message when the command for a transport
5807 filter fails to execute.
5809 4. There are several places where Exim runs a non-Exim command in a
5810 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5811 was being done only for the command run by the queryprogram router. It is
5812 now done for all such subprocesses. The other cases are: ${run, transport
5813 filters, and the commands run by the lmtp and pipe transports.
5815 5. Some older OS have a limit of 256 on the maximum number of file
5816 descriptors. Exim was using setrlimit() to set 1000 as a large value
5817 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5818 systems. I've change it so that if it can't get 1000, it tries for 256.
5820 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5821 was an oversight, and furthermore, ever since the addition of extra
5822 controls (e.g. 4.43/32), the checks on when to allow different forms of
5823 "control" were broken. There should now be diagnostics for all cases when a
5824 control that does not make sense is encountered.
5826 7. $recipients is now available in the predata ACL (oversight).
5828 8. Tidy the search cache before the fork to do a delivery from a message
5829 received from the command line. Otherwise the child will trigger a lookup
5830 failure and thereby defer the delivery if it tries to use (for example) a
5831 cached ldap connection that the parent has called unbind on.
5833 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5834 of $address_data from the recipient verification was clobbered by the
5835 sender verification.
5837 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5838 was its contents. (It was OK if the option was not defined at all.)
5840 11. A "Completed" log line is now written for messages that are removed from
5841 the spool by the -Mrm option.
5843 12. $host_address is now set to the target address during the checking of
5844 ignore_target_hosts.
5846 13. When checking ignore_target_hosts for an ipliteral router, no host name was
5847 being passed; this would have caused $sender_host_name to have been used if
5848 matching the list had actually called for a host name (not very likely,
5849 since this list is usually IP addresses). A host name is now passed as
5852 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5853 code that specifies a non-restarting handler (typically sigaction() in
5854 modern systems) in an attempt to fix a rare and obscure crash bug.
5856 15. Narrowed the window for a race in the daemon that could cause it to ignore
5857 SIGCHLD signals. This is not a major problem, because they are used only to
5858 wake it up if nothing else does.
5860 16. A malformed maildirsize file could cause Exim to calculate negative values
5861 for the mailbox size or file count. Odd effects could occur as a result.
5862 The maildirsize information is now recalculated if the size or filecount
5865 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5866 support for a long time. Removed HAVE_SYS_VFS_H.
5868 18. Updated exipick to current release from John Jetmore.
5870 19. Allow an empty sender to be matched against a lookup in an address list.
5871 Previously the only cases considered were a regular expression, or an
5874 20. Exim went into a mad DNS lookup loop when doing a callout where the
5875 host was specified on the transport, if the DNS lookup yielded more than
5878 21. The RFC2047 encoding function was originally intended for short strings
5879 such as real names; it was not keeping to the 75-character limit for
5880 encoded words that the RFC imposes. It now respects the limit, and
5881 generates multiple encoded words if necessary. To be on the safe side, I
5882 have increased the buffer size for the ${rfc2047: expansion operator from
5885 22. Failure to deliver a bounce message always caused it to be frozen, even if
5886 there was an errors_to setting on the router. The errors_to setting is now
5889 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5890 canonical form (fully expanded) before being placed in
5891 $sender_host_address.
5893 24. Updated eximstats to version 1.33
5895 25. Include certificate and key file names in error message when GnuTLS fails
5896 to set them up, because the GnuTLS error message doesn't include the name
5897 of the failing file when there is a problem reading it.
5899 26. Expand error message when OpenSSL has problems setting up cert/key files.
5902 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5903 (this can affect the format of dates).
5905 28. exim_tidydb, when checking for the continued existence of a message for
5906 which it has found a message-specific retry record, was not finding
5907 messages that were in split spool directories. Consequently, it was
5908 deleting retry records that should have stayed in existence.
5910 29. eximstats updated to version 1.35
5911 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5912 1.35 - bugfix such that pie charts by volume are generated correctly
5914 30. The SPA authentication driver was not abandoning authentication and moving
5915 on to the next authenticator when an expansion was forced to fail,
5916 contradicting the general specification for all authenticators. Instead it
5917 was generating a temporary error. It now behaves as specified.
5919 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5920 (the order specifies the preference for clients). The order is now AES256,
5921 AES128, 3DES, ARCFOUR128.
5923 31. Small patch to Sieve code - explicitly set From: when generating an
5926 32. Exim crashed if a remote delivery caused a very long error message to be
5927 recorded - for instance if somebody sent an entire SpamAssassin report back
5928 as a large number of 550 error lines. This bug was coincidentally fixed by
5929 increasing the size of one of Exim's internal buffers (big_buffer) that
5930 happened as part of the Exiscan merge. However, to be on the safe side, I
5931 have made the code more robust (and fixed the comments that describe what
5934 33. Some experimental protocols are using DNS PTR records for new purposes. The
5935 keys for these records are domain names, not reversed IP addresses. The
5936 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5937 leaves it alone. Component reversal etc. now happens only for IP addresses.
5940 34. The host_aton() function is supposed to be passed a string that is known
5941 to be a valid IP address. However, in the case of IPv6 addresses, it was
5942 not checking this. This is a hostage to fortune. Exim now panics and dies
5943 if the condition is not met. A case was found where this could be provoked
5944 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5945 components; fortuitously, this particular loophole had already been fixed
5946 by change 4.50/55 or 4.44/33 above.
5948 If there are any other similar loopholes, the new check in host_aton()
5949 itself should stop them being exploited. The report I received stated that
5950 data on the command line could provoke the exploit when Exim was running as
5951 exim, but did not say which command line option was involved. All I could
5952 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5953 running as the user.
5956 35. There was a buffer overflow vulnerability in the SPA authentication code
5957 (which came originally from the Samba project). I have added a test to the
5958 spa_base64_to_bits() function which I hope fixes it.
5961 36. The daemon start-up calls getloadavg() while still root for those OS that
5962 need the first call to be done as root, but it missed one case: when
5963 deliver_queue_load_max is set with deliver_drop_privilege. This is
5964 necessary for the benefit of the queue runner, because there is no re-exec
5965 when deliver_drop_privilege is set.
5967 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5968 was in use, was not putting the data itself into the right store pool;
5969 consequently, it could be overwritten for a subsequent message in the same
5970 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5973 38. Sometimes the final signoff response after QUIT could fail to get
5974 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5975 < 0 before doing a fflush(). This bug looks as though it goes back to the
5976 introduction of TLS in release 3.20, but "sometimes" must have been rare
5977 because the tests only now provoked it.
5983 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5984 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5985 Now it returns an integer. A number of calls were still expecting a T/F
5986 return. Fortuitously, in all cases, the tests worked in OK situations,
5987 which is the norm. However, things would have gone wrong on any write
5988 failures on the smtp file descriptor. This function is used when sending
5989 messages over SMTP and also when doing verify callouts.
5991 2. When Exim is called to do synchronous delivery of a locally submitted
5992 message (the -odf or -odi options), it no longer closes stderr before doing
5995 3. Implemented the mua_wrapper option.
5997 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5999 5. Implemented the functions header_remove(), header_testname(),
6000 header_add_at_position(), and receive_remove_recipient(), and exported them
6003 6. If an ACL "warn" statement specified the addition of headers, Exim already
6004 inserted X-ACL-Warn: at the start if there was no header name. However, it
6005 was not making this test for the second and subsequent header lines if
6006 there were newlines in the string. This meant that an invalid header could
6007 be inserted if Exim was badly configured.
6009 7. Allow an ACL "warn" statement to add header lines at the start or after all
6010 the Received: headers, as well as at the end.
6012 8. Added the rcpt_4xx retry error code.
6014 9. Added postmaster_mailfrom=xxx to callout verification option.
6016 10. Added mailfrom=xxxx to the callout verification option, for verify=
6019 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
6020 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
6022 12. Inserted some casts to stop certain compilers warning when using pointer
6023 differences as field lengths or precisions in printf-type calls (mostly
6024 affecting debugging statements).
6026 13. Added optional readline() support for -be (dynamically loaded).
6028 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
6029 same clock tick as a message's arrival, so that its received time was the
6030 same as the "first fail" time on the retry record, and that message
6031 remained on the queue past the ultimate address timeout, every queue runner
6032 would try a delivery (because it was past the ultimate address timeout) but
6033 after another failure, the ultimate address timeout, which should have then
6034 bounced the address, did not kick in. This was a "< instead of <=" error;
6035 in most cases the first failure would have been in the next clock tick
6036 after the received time, and all would be well.
6038 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
6039 being recognized when the domain list was tested by the match_domain
6040 condition in an expansion string.
6042 16. Added the ${str2b64: operator.
6044 17. Exim was always calling setrlimit() to set a large limit for the number of
6045 processes, without checking whether the existing limit was already
6046 adequate. (It did check for the limit on file descriptors.) Furthermore,
6047 errors from getrlimit() and setrlimit() were being ignored. Now they are
6048 logged to the main and panic logs, but Exim does carry on, to try to do its
6049 job under whatever limits there are.
6051 18. Imported PCRE 5.0.
6053 19. Trivial typo in log message " temporarily refused connection" (the leading
6056 20. If the log selector return_path_on_delivery was set and an address was
6057 redirected to /dev/null, the delivery process crashed because it assumed
6058 that a return path would always be set for a "successful" delivery. In this
6059 case, the whole delivery is bypassed as an optimization, and therefore no
6062 21. Internal re-arrangement: the function for sending a challenge and reading
6063 a response while authentication was assuming a zero-terminated challenge
6064 string. It's now changed to take a pointer and a length, to allow for
6065 binary data in such strings.
6067 22. Added the cyrus_sasl authenticator (code supplied by MBM).
6069 23. Exim was not respecting finduser_retries when seeking the login of the
6070 uid under which it was called; it was always trying 10 times. (The default
6071 setting of finduser_retries is zero.) Also, it was sleeping after the final
6072 failure, which is pointless.
6074 24. Implemented tls_on_connect_ports.
6076 25. Implemented acl_smtp_predata.
6078 26. If the domain in control=submission is set empty, Exim assumes that the
6079 authenticated id is a complete email address when it generates From: or
6080 Sender: header lines.
6082 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
6083 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
6084 chown and chgrp in /bin and hostname in /usr/bin.
6086 28. Exim was keeping the "process log" file open after each use, just as it
6087 does for the main log. This opens the possibility of it remaining open for
6088 long periods when the USR1 signal hits a daemon. Occasional processlog
6089 errors were reported, that could have been caused by this. Anyway, it seems
6090 much more sensible not to leave this file open at all, so that is what now
6093 29. The long-running daemon process does not normally write to the log once it
6094 has entered its main loop, and it closes the log before doing so. This is
6095 so that log files can straightforwardly be renamed and moved. However,
6096 there are a couple of unusual error situations where the daemon does write
6097 log entries, and I had neglected to close the log afterwards.
6099 30. The text of an SMTP error response that was received during a remote
6100 delivery was being truncated at 512 bytes. This is too short for some of
6101 the long messages that one sometimes sees. I've increased the limit to
6104 31. It is now possible to make retry rules that apply only when a message has a
6105 specific sender, in particular, an empty sender.
6107 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
6108 it possible to be selective about when SMTP synchronization is enforced.
6110 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
6112 32. Implemented hosts_connection_nolog.
6114 33. Added an ACL for QUIT.
6116 34. Setting "delay_warning=" to disable warnings was not working; it gave a
6119 35. Added mailbox_size and mailbox_filecount to appendfile.
6121 36. Added control = no_multiline_responses to ACLs.
6123 37. There was a bug in the logic of the code that waits for the clock to tick
6124 in the case where the clock went backwards by a substantial amount such
6125 that the microsecond fraction of "now" was more than the microsecond
6126 fraction of "then" (but the whole seconds number was less).
6128 38. Added support for the libradius Radius client library this is found on
6129 FreeBSD (previously only the radiusclient library was supported).
6135 1. When certain lookups returned multiple values in the form name=value, the
6136 quoting of the values was not always being done properly. Specifically:
6137 (a) If the value started with a double quote, but contained no whitespace,
6139 (b) If the value contained whitespace other than a space character (i.e.
6140 tabs or newlines or carriage returns) it was not quoted.
6141 This fix has been applied to the mysql and pgsql lookups by writing a
6142 separate quoting function and calling it from the lookup code. The fix
6143 should probably also be applied to nisplus, ibase and oracle lookups, but
6144 since I cannot test any of those, I have not disturbed their existing code.
6146 2. A hit in the callout cache for a specific address caused a log line with no
6147 reason for rejecting RCPT. Now it says "Previous (cached) callout
6148 verification failure".
6150 3. There was an off-by-one bug in the queryprogram router. An over-long
6151 return line was truncated at 256 instead of 255 characters, thereby
6152 overflowing its buffer with the terminating zero. As well as fixing this, I
6153 have increased the buffer size to 1024 (and made a note to document this).
6155 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
6156 when Exim is waiting for an SMTP response from a remote server, Exim
6157 restarts its select() call on the socket, thereby resetting its timeout.
6158 This is not a problem when such interrupts are rare. Somebody set up a cron
6159 job to run exiwhat every 2 minutes, which is less than the normal select()
6160 timeout (5 or 10 minutes). This meant that the select() timeout never
6161 kicked in because it was always reset. I have fixed this by comparing the
6162 time when an interrupt arrives with the time at the start of the first call
6163 to select(). If more time than the timeout has elapsed, the interrupt is
6164 treated as a timeout.
6166 5. Some internal re-factoring in preparation for the addition of Sieve
6167 extensions (by MH). In particular, the "personal" test is moved to a
6168 separate function, and given an option for scanning Cc: and Bcc: (which is
6169 not set for Exim filters).
6171 6. When Exim created an email address using the login of the caller as the
6172 local part (e.g. when creating a From: or Sender: header line), it was not
6173 quoting the local part when it contained special characters such as @.
6175 7. Installed new OpenBSD configuration files.
6177 8. Reworded some messages for syntax errors in "and" and "or" conditions to
6178 try to make them clearer.
6180 9. Callout options, other than the timeout value, were being ignored when
6181 verifying sender addresses in header lines. For example, when using
6183 verify = header_sender/callout=no_cache
6185 the cache was (incorrectly) being used.
6187 10. Added a missing instance of ${EXE} to the exim_install script; this affects
6188 only the Cygwin environment.
6190 11. When return_path_on_delivery was set as a log selector, if different remote
6191 addresses in the same message used different return paths and parallel
6192 remote delivery occurred, the wrong values would sometimes be logged.
6193 (Whenever a remote delivery process finished, the return path value from
6194 the most recently started remote delivery process was logged.)
6196 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
6197 lines when AUTH and/or TLS are in use. This is the "received protocol"
6198 field. Exim used to use "asmtp" for authenticated SMTP, without any
6199 indication (in the protocol name) for TLS use. Now it follows the RFC and
6200 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
6201 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
6202 names appear in log lines as well as in Received: header lines.
6204 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
6205 extensions, and comparison tests, and to fix some bugs.
6207 14. Changes to the "personal" filter test:
6209 (1) The test was buggy in that it was just doing the equivalent of
6210 "contains" tests on header lines. For example, if a user's address was
6211 anne@some.where, the "personal" test would incorrectly be true for
6213 To: susanne@some.where
6215 This test is now done by extracting each address from the header in turn,
6216 and checking the entire address. Other tests that are part of "personal"
6217 are now done using regular expressions (for example, to check local parts
6218 of addresses in From: header lines).
6220 (2) The list of non-personal local parts in From: addresses has been
6221 extended to include "listserv", "majordomo", "*-request", and "owner-*",
6222 taken from the Sieve specification recommendations.
6224 (3) If the message contains any header line starting with "List-" it is
6225 treated as non-personal.
6227 (4) The test for "circular" in the Subject: header line has been removed
6228 because it now seems ill-conceived.
6230 15. Minor typos in src/EDITME comments corrected.
6232 16. Installed latest exipick from John Jetmore.
6234 17. If headers_add on a router specified a text string that was too long for
6235 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
6236 of string_sprintf() is now avoided.
6238 18. $message_body_size was not set (it was always zero) when running the DATA
6239 ACL and the local_scan() function.
6241 19. For the "mail" command in an Exim filter, no default was being set for
6242 the once_repeat time, causing a random time value to be used if "once" was
6243 specified. (If the value happened to be <= 0, no repeat happened.) The
6244 default is now 0s, meaning "never repeat". The "vacation" command was OK
6245 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
6246 (I found it when inspecting the code).
6248 20. There is now an overall timeout for performing a callout verification. It
6249 defaults to 4 times the callout timeout, which applies to individual SMTP
6250 commands during the callout. The overall timeout applies when there is more
6251 than one host that can be tried. The timeout is checked before trying the
6252 next host. This prevents very long delays if there are a large number of
6253 hosts and all are timing out (e.g. when the network connections are timing
6254 out). The value of the overall timeout can be changed by specifying an
6255 additional sub-option for "callout", called "maxwait". For example:
6257 verify = sender/callout=5s,maxwait=20s
6259 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
6260 to the end before writing, but this should make it even safer).
6262 22. Exim was forgetting that it had advertised PIPELINING for the second and
6263 subsequent messages on an SMTP connection. It was also not resetting its
6264 memory on STARTTLS and an internal HELO.
6266 23. When Exim logs an SMTP synchronization error within a session, it now
6267 records whether PIPELINING has been advertised or not.
6269 24. Added 3 instances of "(long int)" casts to time_t variables that were being
6270 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
6271 rather than long int.
6273 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
6275 26. Added the never_mail option to autoreply.
6281 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
6282 crash if the getsockname() call failed; this can happen if a connection is
6283 closed very soon after it is established. The problem was simply in the
6284 order in which certain operations were done, causing Exim to try to write
6285 to the SMTP stream before it had set up the file descriptor. The bug has
6286 been fixed by making things happen in the correct order.
6292 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
6293 before the connection was closed, thus losing the rejection response.
6295 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
6296 some early Solaris releases, but causes trouble in current releases where
6297 socklen_t is defined.
6299 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
6302 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
6305 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
6306 that releasing the top of it at the end releases what was used for sub-
6307 expansions (unless the block got too big). However, discard this block if
6308 the first thing is a variable or header, so that we can use its block when
6309 it is dynamic (useful for very large $message_headers, for example).
6311 6. Lookups now cache *every* query, not just the most recent. A new, separate
6312 store pool is used for this. It can be recovered when all lookup caches are
6313 flushed. Lookups now release memory at the end of their result strings.
6314 This has involved some general refactoring of the lookup sources.
6316 7. Some code has been added to the store_xxx() functions to reduce the amount
6317 of flapping under certain conditions.
6319 8. log_incoming_interface used to affect only the <= reception log lines. Now
6320 it causes the local interface and port to be added to several more SMTP log
6321 lines, for example "SMTP connection from", and rejection lines.
6323 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
6325 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
6327 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
6328 could be overwritten at the end of the current message (or the start of a
6329 new message if it was set in a HELO ACL). The value is now preserved for
6330 the duration of the SMTP connection.
6332 12. If a transport had a headers_rewrite setting, and a matching header line
6333 contained an unqualified address, that address was qualified, even if it
6334 did not match any rewriting rules. The underlying bug was that the values
6335 of the flags that permit the existence of unqualified sender and recipient
6336 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
6337 non-local messages, and by -bnq for local messages) were not being
6338 preserved with the message after it was received.
6340 13. When Exim was logging an SMTP synchronization error, it could sometimes log
6341 "next input=" as part of the text comprising the host identity instead of
6342 the correct text. The code was using the same buffer for two different
6343 strings. However, depending on which order the printing function evaluated
6344 its arguments, the bug did not always show up. Under Linux, for example, my
6345 test suite worked just fine.
6347 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
6348 doesn't work with some older versions of Perl. It has been changed to "my",
6349 which in any case is probably the better facility to use.
6351 15. A really picky compiler found some instances of statements for creating
6352 error messages that either had too many or two few arguments for the format
6355 16. The size of the buffer for calls to the DNS resolver has been increased
6356 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
6357 for addresses that have a lot of PTR records. This alleviates a problem; it
6358 does not fully solve it.
6360 17. A dnsdb lookup for PTR records that receives more data than will fit in the
6361 buffer now truncates the list and logs the incident, which is the same
6362 action as happens when Exim is looking up a host name and its aliases.
6363 Previously in this situation something unpredictable would happen;
6364 sometimes it was "internal error: store_reset failed".
6366 18. If a server dropped the connection unexpectedly when an Exim client was
6367 using GnuTLS and trying to read a response, the client delivery process
6368 crashed while trying to generate an error log message.
6370 19. If a "warn" verb in an ACL added multiple headers to a message in a single
6371 string, for example:
6373 warn message = H1: something\nH2: something
6375 the text was added as a single header line from Exim's point of view
6376 though it ended up OK in the delivered message. However, searching for the
6377 second and subsequent header lines using $h_h2: did not work. This has been
6378 fixed. Similarly, if a system filter added multiple headers in this way,
6379 the routers could not see them.
6381 20. Expanded the error message when iplsearch is called with an invalid key to
6382 suggest using net-iplsearch in a host list.
6384 21. When running tests using -bh, any delays imposed by "delay" modifiers in
6385 ACLs are no longer actually imposed (and a message to that effect is
6388 22. If a "gecos" field in a passwd entry contained escaped characters, in
6389 particular, if it contained a \" sequence, Exim got it wrong when building
6390 a From: or a Sender: header from that name. A second bug also caused
6391 incorrect handling when an unquoted " was present following a character
6392 that needed quoting.
6394 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
6395 was not being matched caselessly.
6397 24. Arranged for all hyphens in the exim.8 source to be escaped with
6400 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
6401 properly. Recipient callout cache records were still being keyed to include
6402 the sender, even when use_sender was set false. This led to far more
6403 callouts that were necessary. The sender is no longer included in the key
6404 when use_sender is false.
6406 26. Added "control = submission" modifier to ACLs.
6408 27. Added the ${base62d: operator to decode base 62 numbers.
6410 28. dnsdb lookups can now access SRV records.
6412 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
6413 the configuration file.
6415 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
6416 (-v) mode. This makes the output for a verbose queue run more intelligible.
6418 31. Added a use_postmaster feature to recipient callouts.
6420 32. Added the $body_zerocount variable, containing the number of binary zero
6421 bytes in the message body.
6423 33. The time of last modification of the "new" subdirectory is now used as the
6424 "mailbox time last read" when there is a quota error for a maildir
6427 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
6429 35. Added +ignore_unknown as a special item in host lists.
6431 36. Code for decoding IPv6 addresses in host lists is now included, even if
6432 IPv6 support is not being compiled. This fixes a bug in which an IPv6
6433 address was recognized as an IP address, but was then not correctly decoded
6434 into binary, causing unexpected and incorrect effects when compared with
6441 1. Very minor rewording of debugging text in manualroute to say "list of
6442 hosts" instead of "hostlist".
6444 2. If verify=header_syntax was set, and a header line with an unqualified
6445 address (no domain) and a large number of spaces between the end of the
6446 name and the colon was received, the reception process suffered a buffer
6447 overflow, and (when I tested it) crashed. This was caused by some obsolete
6448 code that should have been removed. The fix is to remove it!
6450 3. When running in the test harness, delay a bit after writing a bounce
6451 message to get a bit more predictability in the log output.
6453 4. Added a call to search_tidyup() just before forking a reception process. In
6454 theory, someone could use a lookup in the expansion of smtp_accept_max_
6455 per_host which, without the tidyup, could leave open a database connection.
6457 5. Added the variables $recipient_data and $sender_data which get set from a
6458 lookup success in an ACL "recipients" or "senders" condition, or a router
6459 "senders" option, similar to $domain_data and $local_part_data.
6461 6. Moved the writing of debug_print from before to after the "senders" test
6464 7. Change 4.31/66 (moving the time when the Received: is generated) caused
6465 problems for message scanning, either using a data ACL, or using
6466 local_scan() because the Received: header was not generated till after they
6467 were called (in order to set the time as the time of reception completion).
6468 I have revised the way this works. The header is now generated after the
6469 body is received, but before the ACL or local_scan() are called. After they
6470 are run, the timestamp in the header is updated.
6476 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
6477 before starting a queue runner without re-exec. This happened only when
6478 deliver_drop_privilege was set or when the Exim user was set to root. The
6479 effect of the bug was that timeouts during subsequent deliveries caused
6480 crashes instead of being properly handled. The handler is now left at its
6481 default (and expected) setting.
6483 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
6484 message, again when deliver_drop_privilege is set or Exim is run as root.
6485 The bug described in (1) was not present in this case, but the tidying up
6486 of the other signals was missing. I have made the two cases consistent.
6488 3. The ignore_target_hosts setting on a manualroute router was being ignored
6489 for hosts that were looked up using the /MX notation.
6491 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
6494 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
6495 operated on the sender address. After changing the $sender_address to <>
6496 for the sender address verify, Exim was re-instated it as the original
6497 (before rewriting) address, but remembering that it had rewritten it, so it
6498 wasn't rewriting it again. This bug also had the effect of breaking the
6499 sender address verification caching when the sender address was rewritten.
6501 6. The ignore_target_hosts option was being ignored by the ipliteral router.
6502 This has been changed so that if the ip literal address matches
6503 ignore_target_hosts, the router declines.
6505 7. Added expansion conditions match_domain, match_address, and match_local_
6506 part (NOT match_host).
6508 8. The placeholder for the Received: header didn't have a length field set.
6510 9. Added code to Exim itself and to exim_lock to test for a specific race
6511 condition that could lead to file corruption when using MBX delivery. The
6512 issue is with the lockfile that is created in /tmp. If this file is removed
6513 after a process has opened it but before that process has acquired a lock,
6514 there is the potential for a second process to recreate the file and also
6515 acquire a lock. This could lead to two Exim processes writing to the file
6516 at the same time. The added code performs the same test as UW imapd; it
6517 checks after acquiring the lock that its file descriptor still refers to
6518 the same named file.
6520 10. The buffer for building added header lines was of fixed size, 8192 bytes.
6521 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
6524 11. Added the smtp_active_hostname option. If used, this will typically be made
6525 to depend on the incoming interface address. Because $interface_address is
6526 not set up until the daemon has forked a reception process, error responses
6527 that can happen earlier (such as "too many connections") no longer contain
6530 12. If an expansion in a condition on a "warn" statement fails because a lookup
6531 defers, the "warn" statement is abandoned, and the next ACL statement is
6532 processed. Previously this caused the whole ACL to be aborted.
6534 13. Added the iplsearch lookup type.
6536 14. Added ident_timeout as a log selector.
6538 15. Added tls_certificate_verified as a log selector.
6540 16. Added a global option tls_require_ciphers (compare the smtp transport
6541 option of the same name). This controls incoming TLS connections.
6543 17. I finally figured out how to make tls_require_ciphers do a similar thing
6544 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
6545 before starting the TLS session.
6547 18. Tabs are now shown as \t in -bP output.
6549 19. If the log selector return_path_on_delivery was set, Exim crashed when
6550 bouncing a message because it had too many Received: header lines.
6552 20. If two routers both had headers_remove settings, and the first one included
6553 a superfluous trailing colon, the final name in the first list and the
6554 first name in the second list were incorrectly joined into one item (with a
6555 colon in the middle).
6561 1. Added -C and -D options to the exinext utility, mainly to make it easier
6562 to include in the automated testing, but these could be helpful when
6563 multiple configurations are in use.
6565 2. The exinext utility was not formatting the output nicely when there was
6566 an alternate port involved in the retry record key, nor when there was a
6567 message id as well (for retries that were specific to a specific message
6568 and a specific host). It was also confused by IPv6 addresses, because of
6569 the additional colons they contain. I have fixed the IPv4 problem, and
6570 patched it up to do a reasonable job for IPv6.
6572 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
6573 delivery, the log line now contains "pipelined" if PIPELINING was used.
6575 4. An SMTP transport process used to panic and die if the bind() call to set
6576 an explicit outgoing interface failed. This has been changed; it is now
6577 treated in the same way as a connect() failure.
6579 5. A reference to $sender_host_name in the part of a conditional expansion
6580 that was being skipped was still causing a DNS lookup. This no longer
6583 6. The def: expansion condition was not recognizing references to header lines
6584 that used bh_ and bheader_.
6586 7. Added the _cache feature to named lists.
6588 8. The code for checking quota_filecount in the appendfile transport was
6589 allowing one more file than it should have been.
6591 9. For compatibility with Sendmail, the command line option
6599 and sets the incoming protocol and host name (for trusted callers). The
6600 host name and its colon can be omitted when only the protocol is to be set.
6601 Note the Exim already has two private options, -pd and -ps, that refer to
6602 embedded Perl. It is therefore impossible to set a protocol value of "d" or
6603 "s", but I don't think that's a major issue.
6605 10. A number of refactoring changes to the code, none of which should affect
6608 (a) The number of logging options was getting close to filling up the
6609 32-bit word that was used as a bit map. I have split them into two classes:
6610 those that are passed in the argument to log_write(), and those that are
6611 only ever tested independently outside of that function. These are now in
6612 separate 32-bit words, so there is plenty of room for expansion again.
6613 There is no change in the user interface or the logging behaviour.
6615 (b) When building, for example, log lines, the code previously used a
6616 macro that called string_cat() twice, in order to add two strings. This is
6617 not really sufficiently general. Furthermore, there was one instance where
6618 it was actually wrong because one of the argument was used twice, and in
6619 one call a function was used. (As it happened, calling the function twice
6620 did not affect the overall behaviour.) The macro has been replaced by a
6621 function that can join an arbitrary number of extra strings onto a growing
6624 (c) The code for expansion conditions now uses a table and a binary chop
6625 instead of a serial search (which was left over from when there were very
6626 few conditions). Also, it now recognizes conditions like "pam" even when
6627 the relevant support is not compiled in: a suitably worded error message is
6628 given if an attempt is made to use such a condition.
6630 11. Added ${time_interval:xxxxx}.
6632 12. A bug was causing one of the ddress fields not to be passed back correctly
6633 from remote delivery subprocesses. The field in question was not being
6634 subsequently used, so this caused to problems in practice.
6636 13. Added new log selectors queue_time and deliver_time.
6638 14. Might have fixed a bug in maildirsizefile handling that threw up
6639 "unexpected character" debug warnings, and recalculated the data
6640 unnecessarily. In any case, I expanded the warning message to give more
6643 15. Added the message "Restricted characters in address" to the statements in
6644 the default ACL that block characters like @ and % in local parts.
6646 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
6647 Three changes have been made:
6649 (a) There was a serious bug; a negative response to MAIL caused the whole
6650 recipient domain to be cached as invalid, thereby blocking all messages
6651 to all local parts at the same domain, from all senders. This bug has
6652 been fixed. The domain is no longer cached after a negative response to
6653 MAIL if the sender used is not empty.
6655 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
6658 (c) A new callout option, "use_sender" has been added for people who want
6659 the modified behaviour.
6665 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
6668 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
6669 indeed breaks things for older releases.
6671 3. Added additional logging to the case where there is a problem reading data
6672 from a filter that is running in a subprocess using a pipe, in order to
6673 try to track down a specific problem.
6675 4. Testing facility fudge: when running in the test harness and attempting
6676 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
6677 getting "No route to host". Convert this to a timeout.
6679 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
6682 6. Some OS don't have socklen_t but use size_t instead. This affects the
6683 fifth argument of getsockopt() amongst other things. This is now
6684 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
6685 can be set for individual OS. I have set it for SunOS5, OSF1, and
6686 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
6687 some earlier ones do not.
6689 7. Change 4.30/15 was not doing the test caselessly.
6691 8. The standard form for an IPv6 address literal was being rejected by address
6692 parsing in, for example, MAIL and RCPT commands. An example of this kind of
6693 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6694 this, as well as the form without the "IPv6" on the front (but only when
6695 address literals are enabled, of course).
6697 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
6699 10. Exim crashed if a message with an empty sender address specified by -f
6700 encountered a router with an errors_to setting. This could be provoked only
6701 by a command such as
6705 where an empty string was supplied; "<>" did not hit this bug.
6707 11. Installed PCRE release 4.5.
6709 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
6710 remained set. It is now erased.
6712 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
6713 times from message ids (which are base 36 rather than the normal 62).
6715 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
6716 were being counted as actual protocol errors, and logged if the log
6717 selector +smtp_protocol_error was set. One cannot be perfect in this test,
6718 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
6719 and DATA following a set of rejected RCPTs do not count as protocol errors.
6720 In other words, Exim assumes they were pipelined, though this may not
6721 actually be the case. Of course, in all cases the client gets an
6722 appropriate error code.
6724 15. If a lookup fails in an ACL condition, a message about the failure may
6725 be available; it is used if testing the ACL cannot continue, because most
6726 such messages specify what the cause of the deferral is. However, some
6727 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
6728 that caused an old message to be retained and used if a later statement
6729 caused a defer, replacing the real cause of the deferral.
6731 16. If an IP address had so many PTR records that the DNS lookup buffer
6732 was not large enough to hold them, Exim could crash while trying to process
6733 the truncated data. It now detects and logs this case.
6735 17. Further to 4.21/58, another change has been made: if (and only if) the
6736 first line of a message (the first header line) ends with CRLF, a bare LF
6737 in a subsequent header line has a space inserted after it, so as not to
6738 terminate the header.
6740 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
6741 unnecessarily, used atoi() instead of strtol(), and didn't check the
6742 termination when getting file sizes from file names by regex.
6744 19. Completely re-implemented the support for maildirsize files, in the light
6745 of a number of problems with the previous contributed implementation
6746 (4.30/29). In particular:
6748 . If the quota is zero, the maildirsize file is maintained, but no quota is
6751 . If the maildir directory does not exist, it is created before any attempt
6752 to write a maildirsize file.
6754 . The quota value in the file is just a cache; if the quota is changed in
6755 the transport, the new value overrides.
6757 . A regular expression is available for excluding directories from the
6760 20. The autoreply transport checks the characters in options that define the
6761 message's headers; it allows continued headers, but it was checking with
6762 isspace() after an embedded newline instead of explicitly looking for a
6765 21. If all the "regular" hosts to which an address was routed had passed their
6766 expiry times, and had not reached their retry times, the address was
6767 bounced, even if fallback hosts were defined. Now Exim should go on to try
6770 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
6771 equivalent code in the SMTP transport. Some hosts send humungous responses
6772 to HELO/EHLO, more than 1024 it seems.
6774 23. Refactoring: code in filter.c used (void *) for "any old type" but this
6775 gives compiler warnings in some environments. I've now done it "properly",
6778 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
6779 (because of problems with the built-in one) was declared to return uschar *
6780 instead of char *, causing compiler failure.
6782 25. Fixed a file descriptor leak when processing alias/forward files.
6784 26. Fixed a minor format string issue in dbfn.c.
6786 27. Typo in exim.c: ("dmbnz" for "dbmnz").
6788 28. If a filter file refered to $h_xxx or $message_headers, and the headers
6789 contained RFC 2047 "words", Exim's memory could, under certain conditions,
6792 29. When a sender address is verified, it is cached, to save repeating the test
6793 when there is more than one recipient in a message. However, when the
6794 verification involves a callout, it is possible for different callout
6795 options to be set for different recipients. It is too complicated to keep
6796 track of this in the cache, so now Exim always runs a verification when a
6797 callout is required, relying on the callout cache for the optimization.
6798 The overhead is duplication of the address routing, but this should not be
6801 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
6802 to be verified with callout=postmaster, and the main callout worked but the
6803 postmaster check failed, the verification correctly failed. However, if a
6804 subsequent RCPT command asked for sender verification *without* the
6805 postmaster check, incorrect caching caused this verification also to fail,
6808 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
6809 it was not caching the DNS options (qualify_single, search_parents) that
6810 were used when the lookup failed. A subsequent lookup with different
6811 options therefore always gave the same answer, though there were cases
6812 where it should not have. (Example: a "domains = !$mx_any" option on a
6813 dnslookup router: the "domains" option is always processed without any
6814 widening, but the router might have qualify_single set.) Now Exim uses the
6815 cached value only when the same options are set.
6817 32. Added John Jetmore's "exipick" utility to the distribution.
6819 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
6820 than a timeout (e.g. a certificate is required, and is not provided), an
6821 Exim server now closes the connection immediately. Previously it waited for
6822 the client to close - but if the client is SSL, it seems that they each
6823 wait for each other, leading to a delay before one of them times out.
6825 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
6826 maintained 0.8.x compatibility because I don't think many are using it, and
6827 it is clearly obsolete.
6829 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
6832 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
6833 client certificate was expired. A simple patch fixes this, though I don't
6834 understand the full logic of why the verify callback is called multiple
6837 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
6838 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
6839 which causes problems with some clients (such as the Certicom SSL Plus
6840 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
6841 disables the coutermeasure allowing Eudora to connect."
6843 38. Exim was not checking that a write() to a log file succeeded. This could
6844 lead to Bad Things if a log got too big, in particular if it hit a file
6845 size limit. Exim now panics and dies if it cannot write to a log file, just
6846 as it does if it cannot open a log file.
6848 39. Modified OS/Makefile-Linux so that it now contains
6850 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
6852 The two -D definitions ensure that Exim is compiled with large file
6853 support, which makes it possible to handle log files that are bigger than
6856 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
6857 instance) a domain was checked against a named list that involved a lookup,
6858 causing $domain_data to be set, then another domain was checked against the
6859 same list, then the first domain was re-checked, the value of $domain_data
6860 after the final check could be wrong. In particular, if the second check
6861 failed, it could be set empty. This bug probably also applied to
6864 41. The strip_trailing_dot option was not being applied to the address given
6865 with the -f command-line option.
6867 42. The code for reading a message's header from the spool was incrementing
6868 $received_count, but never initializing it. This meant that the value was
6869 incorrect (doubled) while delivering a message in the same process in which
6870 it was received. In the most common configuration of Exim, this never
6871 happens - a fresh exec is done - but it can happen when
6872 deliver_drop_privilege is set.
6874 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6875 it now includes up to 150 characters of the unexpected data in the log
6878 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6879 and building data strings. The size of both of these buffers was 10 000
6880 bytes - far larger than anybody would *ever* want, thought I. Needless to
6881 say, somebody hit the limit. I have increased the maximum line length to
6882 20 000 and the maximum data length of concatenated lines to 100 000. I have
6883 also fixed two bugs, because there was no checking on these buffers. Tsk,
6884 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6885 buffer is too small.
6887 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6888 lsearch lookups. Now it does.
6890 46. When parsing a route_list item in a manualroute router, a fixed-length
6891 buffer was used for the list of hosts. I made this 1024 bytes long,
6892 thinking that nobody would ever have a list of hosts that long. Wrong.
6893 Somebody had a whole pile of complicated expansion conditions, and the
6894 string was silently truncated, leading to an expansion error. It turns out
6895 that it is easier to change to an unlimited length (owing to other changes
6896 that have happened since this code was originally written) than to build
6897 structure for giving a limitation error. The length of the item that
6898 expands into the list of hosts is now unlimited.
6900 47. The lsearch lookup could not handle data where the length of text line was
6901 more than 4095 characters. Such lines were truncated, leading to shortened
6902 data being returned. It should now handle lines of any length.
6904 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6905 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6908 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6909 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6910 no longer happens in comment lines. A list of replacements is now placed
6911 at the head of all of the source files, except those whose only change is
6912 to replace PERL_COMMAND in the very first #! line.
6914 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6915 messages on the queue, with a bottom-up merge sort, using code contributed
6916 by Michael Haardt. This should make operations like -bp somewhat faster on
6917 large queues. It won't affect queue runners, except when queue_run_in_order
6920 51. Installed eximstats 1.31 in the distribution.
6922 52. Added support for SRV lookups to the dnslookup router.
6924 53. If an ACL referred to $message_body or $message_body_end, the value was not
6925 reset for any messages that followed in the same SMTP session.
6927 54. The store-handling optimization for building very long strings was not
6928 differentiating between the different store pools. I don't think this
6929 actually made any difference in practice, but I've tidied it.
6931 55. While running the routers to verify a sender address, $sender_address
6932 was still set to the sender address. This is wrong, because when routing to
6933 send a bounce to the sender, it would be empty. Therefore, I have changed
6934 it so that, while verifying a sender address, $sender_address is set to <>.
6935 (There is no change to what happens when verifying a recipient address.)
6937 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6938 target A or AAAA records (if not already returned) without resetting the
6939 qualify_single or search_parents options of the DNS resolver. These are
6940 inappropriate in this case because the targets of MX and SRV records must
6941 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6942 target that, when qualified, matched something in the local domain. These
6943 two options are now turned off when doing these lookups.
6945 57. It seems that at least some releases of Reiserfs (which does not have the
6946 concept of a fixed number of inodes) returns zero and not -1 for the
6947 number of available inodes. This interacted badly with check_spool_inodes,
6948 which assumed that -1 was the "no such thing" setting. What I have done is
6949 to check that the total number of inodes is greater than zero before doing
6950 the test of how many are available.
6952 58. When a "warn" ACL statement has a log_message modifier, the message is
6953 remembered, and not repeated. This is to avoid a lot of repetition when a
6954 message has many recipients that cause the same warning to be written.
6955 However, Exim was preserving the list of already written lines for an
6956 entire SMTP session, which doesn't seem right. The memory is now reset if a
6957 new message is started.
6959 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6960 debugging output unless log_rewrite was also set.
6962 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6963 of (int)(handle) when we know that handle contains (void *)(-1).
6965 61. The Exim daemon panic-logs an error return when it closes the incoming
6966 connection. However "connection reset by peer" seems to be common, and
6967 isn't really an error worthy of noting specially, so that particular error
6970 62. When Exim is trying to find all the local interfaces, it used to panic and
6971 die if the ioctl to get the interface flags failed. However, it seems that
6972 on at least one OS (Solaris 9) it is possible to have an interface that is
6973 included in the list of interfaces, but for which you get a failure error
6974 for this call. This happens when the interface is not "plumbed" into a
6975 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6976 failure of the "get flags" call assumes that the interface is down.
6978 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6979 makes life easier for people who are doing arithmetic on fields extracted
6980 from dates, where you often get leading zeros that should not be
6981 interpreted as octal.
6983 64. Added qualify_domain to the redirect router, to override the global
6986 65. If a pathologically long header line contained very many addresses (the
6987 report of this problem mentioned 10 000) and each of them was rewritten,
6988 Exim could use up a very large amount of memory. (It kept on making new
6989 copies of the header line as it rewrote, and never released the old ones.)
6990 At the expense of a bit more processing, the header rewriting function has
6991 been changed so that it no longer eats memory in this way.
6993 66. The generation of the Received: header has been moved from the time that a
6994 message starts to be received, to the time that it finishes. The timestamp
6995 in the Received: header should now be very close to that of the <= log
6996 line. There are two side-effects of this change:
6998 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6999 logged header lines no longer include the local Received: line, because
7000 it has not yet been created. The same applies to a copy of the message
7001 that is returned to a non-SMTP sender when a message is rejected.
7003 (b) When a filter file is tested using -bf, no additional Received: header
7004 is added to the test message. After some thought, I decided that this
7007 This change does not affect the value of $received_for. It is still set
7008 after address rewriting, but before local_scan() is called.
7010 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
7012 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
7013 gave an unhelpful panic error message, and a defer error. I have managed to
7014 change this behaviour so that it now rejects any supplied certificate,
7015 which seems right, as the list of acceptable certificates is empty.
7017 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
7018 gave an unhelpful defer error. I have not managed to make this reject any
7019 supplied certificates, but the error message it gives is "no certificate
7020 supplied", which is not helpful.
7022 70. exigrep's output now also includes lines that are not associated with any
7023 message, but which match the given pattern. Implemented by a patch from
7024 Martin Sluka, which also tidied up the Perl a bit.
7026 71. Recipient callout verification, like sender verification, was using <> in
7027 the MAIL FROM command. This isn't really the right thing, since the actual
7028 sender may affect whether the remote host accepts the recipient or not. I
7029 have changed it to use the actual sender in the callout; this means that
7030 the cache record is now keyed on a recipient/sender pair, not just the
7031 recipient address. There doesn't seem to be a real danger of callout loops,
7032 since a callout by the remote host to check the sender would use <>.
7033 [SEE ABOVE: changed after hitting problems.]
7035 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
7036 temporary errors. However, in the case of such a code being given after
7037 the end of a data transmission (i.e. after ".") Exim was failing to write
7038 a retry record for the message. (Yes, there was some broken host that was
7039 actually sending 8xx at this point.)
7041 73. An unknown lookup type in a host list could cause Exim to panic-die when
7042 the list was checked. (An example that provoked this was putting <; in the
7043 middle of a list instead of at the start.) If this happened during a DATA
7044 ACL check, a -D file could be left lying around. This kind of configuration
7045 error no longer causes Exim to die; instead it causes a defer error. The
7046 incident is still logged to the main and panic logs.
7048 74. Buglet left over from Exim 3 conversion. The message "too many messages
7049 in one connection" was written to the rejectlog but not the mainlog, except
7050 when address rewriting (yes!) was being logged.
7052 75. Added write_rejectlog option.
7054 76. When a system filter was run not as root (that is, when system_filter_user
7055 was set), the values of the $n variables were not being returned to the
7056 main process; thus, they were not subsequently available in the $sn
7059 77. Added +return_path_on_delivery log selector.
7061 78. A connection timeout was being treated differently from recipients deferred
7062 when testing hosts_max_try with a message that was older than the host's
7063 retry timeout. (The host should not be counted, thus allowing all hosts to
7064 be tried at least once before bouncing.) This may have been the cause of an
7065 occasionally reported bug whereby a message would remain on the queue
7066 longer than the retry timeout, but would be bounced if a delivery was
7067 forced. I say "may" because I never totally pinned down the problem;
7068 setting up timeout/retry tests is difficult. See also the next item.
7070 79. The ultimate address timeout was not being applied to errors that involved
7071 a combination of host plus message (for example, a timeout on a MAIL
7072 command). When an address resolved to a number of possible hosts, and they
7073 were not all tried for each delivery (e.g. because of hosts_max_try), a
7074 message could remain on the queue longer than the retry timeout.
7076 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
7079 81. Fixed an obscure SMTP outgoing bug which required at least the following
7080 conditions: (a) there was another message waiting for the same server;
7081 (b) the server returned 5xx to all RCPT commands in the first message so
7082 that the message was not completed; (c) the server dropped the connection
7083 or gave a negative response to the RSET that Exim sends to abort the
7084 transaction. The observed case was a dropped connection after DATA that had
7085 been sent in pipelining mode. That is, the server had advertised PIPELINING
7086 but was not implementing it correctly. The effect of the bug was incorrect
7087 behaviour, such as trying another host, and this could lead to a crash.
7093 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
7094 and daemon.c were passed as pointers to ints; they should have been
7095 pointers to socklen_t variables (which are typically unsigned ints).
7097 2. Some signed/unsigned type warnings in the os.c file for Linux have been
7100 3. Fixed a really odd bug that affected only the testing scheme; patching a
7101 certain fixed string in the binary changed the value of another string that
7102 happened to be identical to the end of the original first string.
7104 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
7105 name", it returns that address as the IP address. On some operating
7106 systems (e.g. Solaris), it also passes back the IP address string as the
7107 "host name". However, on others (e.g. Linux), it passes back an empty
7108 string. Exim wasn't checking for this, and was changing the host name to an
7109 empty string, assuming it had been canonicalized.
7111 5. Although rare, it is permitted to have more than one PTR record for a given
7112 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
7113 all the names associated with an address, because they do in Solaris.
7114 However, it seems that they do not in Linux for data that comes from the
7115 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
7116 I found this out when I moved to a new Linux workstation and tried to run
7117 the Exim test suite.
7119 To get round this problem I have changed the code so that it now does its
7120 own call to the DNS to look up PTR records when searching for a host name.
7121 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
7122 addresses that are only in /etc/hosts are still found.
7124 This behaviour is, however, controlled by an option called host_lookup_
7125 order, which defaults to "bydns:byaddr". If people want to use the other
7126 order, or indeed, just use one or the other means of lookup, they can
7127 specify it in this variable.
7129 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
7130 some operating systems, this comes back from gethostbyaddr() as an empty
7131 string, and this is what Exim used to test for. However, it seems that in
7132 other systems, "." is yielded. Exim now tests for this case too.
7134 7. The values of check_spool_space and check_log_space are now held internally
7135 as a number of kilobytes instead of an absolute number of bytes. If a
7136 numbers is specified without 'K' or 'M', it is rounded up to the nearest
7137 kilobyte. This means that much larger values can be stored.
7139 8. Exim monitor: an attempt to get the action menu when not actually pointing
7140 at a message produces an empty menu entitled "No message selected". This
7141 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
7142 no entries in it ("Shell widget menu has zero width and/or height"). So I
7143 have added a single, blank menu entry in this case.
7145 9. Added ${quote_local_part.
7147 10. MIME decoding is now applied to the contents of Subject: header lines when
7150 11. Now that a reference to $sender_host_address automatically causes a reverse
7151 lookup to occur if necessary (4.13/18), there is no need to arrange for a
7152 host lookup before query-style lookups in lists that might use this
7153 variable. This has therefore been abolished, and the "net-" prefix is no
7154 longer necessary for query-style lookups.
7156 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
7157 have been a typo for LFLAGS, so it has been changed.
7159 13. The install script calls Exim with "-C /dev/null" in order to find the
7160 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
7161 to be output. However, since Exim outputs its version number before the
7162 error, it didn't break the script. It just looked ugly. I fixed this by
7163 always allowing "-C /dev/null" if the caller is root.
7165 14. Ignore overlarge ACL variable number when reading spool file - insurance
7166 against a later release with more variables having written the file.
7168 15. The standard form for an IPv6 address literal was being rejected by EHLO.
7169 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
7170 this, as well as the form without the "IPv6" on the front.
7172 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
7173 OS/Makefile-Darwin file.
7175 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
7176 only to LDAP libraries that do not have LDAP_OPT_DEREF.
7178 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
7179 variable. However, some OS use ints for inodes. I've added cast to long int
7180 to get rid of the compiler warning.
7182 19. I had forgotten to lock out "/../" in configuration file names when
7183 ALT_CONFIG_PREFIX was set.
7185 20. Routers used for verification do not need to specify transports. However,
7186 if such a router generated a host list, and callout was configured, Exim
7187 crashed, because it could not find a port number from the (non-existent)
7188 transport. It now assumes port 25 in this circumstance.
7190 21. Added the -t option to exigrep.
7192 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
7193 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
7194 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
7195 them is set, LOOKUP_LSEARCH is forced.
7197 23. "exim -bV" now outputs a list of lookups that are included in the binary.
7199 24. Added sender and host information to the "rejected by local_scan()" log
7200 line; previously there was no indication of these.
7202 25. Added .include_if_exists.
7204 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
7205 Linux. It turns out that not all file systems support this. Apparently some
7206 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
7207 do it.) To cope with this, the error EINVAL, which means that sync-ing is
7208 not supported on the file descriptor, is now ignored when Exim is trying to
7209 sync a directory. This applies only to Linux.
7211 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
7213 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
7214 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
7217 29. Added support for maildirsize files from supplied patch (modified a bit).
7219 30. The use of :fail: followed by an empty string could lead Exim to respond to
7220 sender verification failures with (e.g.):
7222 550 Verification failed for <xxx>
7223 550 Sender verify failed
7225 where the first response line was missing the '-' that indicates it is not
7226 the final line of the response.
7228 31. The loop for finding the name of the user that called Exim had a hardwired
7229 limit of 10; it now uses the value of finduser_retries, which is used for
7230 all other user lookups.
7232 32. Added $received_count variable, available in data and not_smtp ACLs, and at
7235 33. Exim was neglecting to zero errno before one call of strtol() when
7236 expanding a string and expecting an integer value. On some systems this
7237 resulted in spurious "integer overflow" errors. Also, it was casting the
7238 result into an int without checking.
7240 34. Testing for a connection timeout using "timeout_connect" in the retry rules
7241 did not work. The code looks as if it has *never* worked, though it appears
7242 to have been documented since at least release 1.62. I have made it work.
7244 35. The "timeout_DNS" error in retry rules, also documented since at least
7245 1.62, also never worked. As it isn't clear exactly what this means, and
7246 clearly it isn't a major issue, I have abolished the feature by treating it
7247 as "timeout", and writing a warning to the main and panic logs.
7249 36. The display of retry rules for -brt wasn't always showing the error code
7252 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
7253 timeout_connect_A, timeout_connect_MX.
7255 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
7256 to the empty sender.
7258 39. The daemon was not analysing the content of -oX till after it had closed
7259 stderr and disconnected from the controlling terminal. This meant that any
7260 syntax errors were only noted on the panic log, and the return code from
7261 the command was 0. By re-arranging the code a little, I've made the
7262 decoding happen first, so such errors now appear on stderr, and the return
7263 code is 1. However, the actual setting up of the sockets still happens in
7264 the disconnected process, so errors there are still only recorded on the
7267 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
7268 connections (as happens on some IP stacks) was logged at start up time as
7269 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
7270 it from "IPv6 and IPv4", which means that two separate sockets are being
7273 41. The debug output for gethostbyname2() or getipnodebyname() failures now
7274 says whether AF_INET or AF_INET6 was passed as an argument.
7276 42. Exiwhat output was messed up when time zones were included in log
7279 43. Exiwhat now gives more information about the daemon's listening ports,
7280 and whether -tls-on-connect was used.
7282 44. The "port" option of the smtp transport is now expanded.
7284 45. A "message" modifier in a "warn" statement in a non-message ACL was being
7285 silently ignored. Now an error message is written to the main and panic
7288 46. There's a new ACL modifier called "logwrite" which writes to a log file
7289 as soon as it is encountered.
7291 47. Added $local_user_uid and $local_user_gid at routing time.
7293 48. Exim crashed when trying to verify a sender address that was being
7296 49. Exim was recognizing only a space character after ".include". It now also
7297 recognizes a tab character.
7299 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
7300 extracting the relevant information from the specification. The man page no
7301 longer contains scrambled data for the -d option, and I've added a section
7302 at the front about calling Exim under different names.
7304 51. Added "extra_headers" argument to the "mail" command in filter files.
7306 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
7309 53. Installed eximstats 1.29.
7311 54. Added transport_filter_timeout as a generic transport option.
7313 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
7314 Cc: header lines. This was required by RFC 822, but it not required by RFC
7317 56. Exim used to add From:, Date:, and Message-Id: header lines to any
7318 incoming messages that did not have them. Now it does so only if the
7319 message originates locally, that is, if there is no associated remote host
7320 address. When Resent- header lines are present, this applies to the Resent-
7321 lines rather than the non-Resent- lines.
7323 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
7324 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
7326 58. Messages for configuration errors now include the name of the main
7327 configuration file - useful now that there may be more than one file in a
7328 list (.included file names were always shown).
7330 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
7331 for those rare installations that do not start the daemon as root or run it
7332 setuid root. I've cut out the call to initgroups() if the daemon is not
7335 60. The Exim user and group can now be bound into the binary as text strings
7336 that are looked up at the start of Exim's processing.
7338 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
7340 62. Added $mailstore_basename variable.
7342 63. Installed patch to sieve.c from Michael Haardt.
7344 64. When Exim failed to open the panic log after failing to open the main log,
7345 the original message it was trying to log was written to stderr and debug
7346 output, but if they were not available (the usual case in production), it
7347 was lost. Now it is written to syslog before the two lines that record the
7348 failures to open the logs.
7350 65. Users' Exim filters run in subprocesses under the user's uid. It is
7351 possible for a "deliver" command or an alias in a "personal" command to
7352 provoke an address rewrite. If logging of address rewriting is configured,
7353 this fails because the process is not running as root or exim. There may be
7354 a better way of dealing with this, but for the moment (because 4.30 needs
7355 to be released), I have disabled address rewrite logging when running a
7356 filter in a non-root, non-exim process.
7362 1. The buildconfig auxiliary program wasn't quoting the value set for
7363 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
7364 not defined. This bug was masked in 4.22 by the effect that was fixed in
7367 2. Some messages that were rejected after a message id was allocated were
7368 shown as "incomplete" by exigrep. It no longer does this for messages that
7369 are rejected by local_scan() or the DATA or non-SMTP ACLs.
7371 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
7372 have allow_domain_literals set, the ID did not get logged in the <= line.
7373 Domain literals are now always recognized in Message-ID: header lines.
7375 4. The first argument for a ${extract expansion item is the key name or field
7376 number. Leading and trailing spaces in this item were not being ignored,
7377 causing some misleading effects.
7379 5. When deliver_drop_privilege was set, single queue runner processes started
7380 manually (i.e. by the command "exim -q") or by the daemon (which uses the
7381 same command in the process it spins off) were not dropping privilege.
7383 6. When the daemon running as "exim" started a queue runner, it always
7384 re-executed Exim in the spun-off process. This is a waste of effort when
7385 deliver_drop_privilege is set. The new process now just calls the
7386 queue-runner function directly.
7392 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
7395 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
7396 ignored. Though the use of -oP was forcing the writing of a pid file, it
7397 was always written to the default place.
7399 3. If the message "no IP address found for host xxxx" is generated during
7400 incoming verification, it is now followed by identification of the incoming
7401 connection (so you can more easily find what provoked it).
7403 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
7405 5. Added some features to "harden" Exim a bit more against certain attacks:
7407 (a) There is now a build-time option called FIXED_NEVER_USERS that can
7408 be put in Local/Makefile. This is like the never_users runtime option,
7409 but it cannot be overridden. The default setting is "root".
7411 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
7412 prefix string with which any file named in a -C command line option
7415 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
7416 is retained for -C and -D only if the caller of Exim is root. Without
7417 it, the exim user may also use -C and -D and retain privilege.
7419 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
7420 command line option is disabled.
7422 6. Macro names set by the -D option must start with an upper case letter, just
7423 like macro names defined in the configuration file.
7425 7. Added "dereference=" facility to LDAP.
7427 8. Two instances of the typo "uknown" in the source files are fixed.
7429 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
7430 the Configure-Makefile script screwed up while processing it.
7432 10. Incorporated PCRE 4.4.
7434 11. The SMTP synchronization check was not operating right at the start of an
7435 SMTP session. For example, it could not catch a HELO sent before the client
7436 waited for the greeting. There is now a check for outstanding input at the
7437 point when the greeting is written. Because of the duplex, asynchronous
7438 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
7439 way, but not yet received, when the check is performed.
7441 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
7442 on TCP/IP sockets, because this apparently causes some broken clients to
7445 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
7446 unchanged) from the Cygwin maintainer.
7448 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
7449 was supported instead of testing for mbx. Effectively a typo.
7451 15. The spa authenticator server code was not checking that the input it
7452 received was valid base64.
7454 16. The debug output line for the "set" modifier in ACLs was not showing the
7455 name of the variable that was being set.
7457 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
7459 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
7460 lookup on its own. Something else was needed to trigger the lookup. For
7461 example, a match in host_lookup or the need for a host name in a host list.
7462 Now, if $sender_host_name is referenced and the host name has not yet been
7463 looked up, a lookup is performed. If the lookup fails, the variable remains
7464 empty, and $host_lookup_failed is set to "1".
7466 19. Added "eqi" as a case-independent comparison operator.
7468 20. The saslauthd authentication condition could segfault if neither service
7469 nor realm was specified.
7471 21. If an overflowing value such as "2048M" was set for message_size_limit, the
7472 error message that was logged was misleading, and incoming SMTP
7473 connections were dropped. The message is now more accurate, and temporary
7474 errors are given to SMTP connections.
7476 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
7477 (except RSET) with a 421 error, until QUIT is received. However, it was
7478 failing to send a response to QUIT.
7480 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
7481 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
7482 ACL is now run after the helo_try_verify_hosts code.
7484 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
7485 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
7486 case-independent, so other case variants are also recognized). Apparently
7487 some systems use these upper case variants.
7489 25. If more than two messages were waiting for the same host, and a transport
7490 filter was specified for the transport, Exim sent two messages over the
7491 same TCP/IP connection, and then failed with "socket operation on non-
7492 socket" when it tried to send the third.
7494 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
7496 27. The extern definition of crypt16() in expand.c was not being excluded when
7497 the OS had its own crypt16() function.
7499 28. Added bounce_return_body as a new option, and bounce_return_size_limit
7500 as a preferred synonym for return_size_limit, both as an option and as an
7503 29. Added LIBS=-liconv to OS/Makefile-OSF1.
7505 30. Changed the default configuration ACL to relax the local part checking rule
7506 for addresses that are not in any local domains. For these addresses,
7507 slashes and pipe symbols are allowed within local parts, but the sequence
7508 /../ is explicitly forbidden.
7510 31. SPA server authentication was not clearing the challenge buffer before
7513 32. log_message in a "warn" ACL statement was writing to the reject log as
7514 well as to the main log, which contradicts the documentation and doesn't
7515 seem right (because no rejection is happening). So I have stopped it.
7517 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
7518 However, I am unable to do any testing of this.
7520 34. Fixed an infelicity in the appendfile transport. When checking directories
7521 for a mailbox, to see if any needed to be created, it was accidentally
7522 using path names with one or more superfluous leading slashes; tracing
7523 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
7525 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
7526 contents are added to the log line that is written for every discarded
7527 recipient. (Previously a log_message setting was ignored.)
7529 36. The ${quote: operator now quotes the string if it is empty.
7531 37. The install script runs exim in order to find its version number. If for
7532 some reason other than non-existence or emptiness, which it checks, it
7533 could not run './exim', it was installing it with an empty version number,
7534 i.e. as "exim-". This error state is now caught, and the installation is
7537 38. An argument was missing from the function that creates an error message
7538 when Exim fails to connect to the socket for saslauthd authentication.
7539 This could cause Exim to crash, or give a corrupted message.
7541 39. Added isip, isip4, and isip6 to ${if conditions.
7543 40. The ACL variables $acl_xx are now saved with the message, and can be
7544 accessed later in routers, transports, and filters.
7546 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
7547 strings in the file are not string-expanded.
7549 42. If a MAIL command specified a SIZE value that was too large to fit into an
7550 int variable, the check against message_size_limit failed. Such values are
7551 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
7552 day this will have to be increased, but I don't think I want to be around
7553 when emails are that large.
7560 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
7561 iconv() is not standard in FreeBSD.
7563 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
7564 IPv6 enabled. The observed symptom was a segmentation fault on return from
7565 the function os_common_find_running_interfaces() in src/os.c.
7567 3. In the check_special_case() function in daemon.c I had used "errno" as an
7568 argument name, which causes warnings on some systems. This was basically a
7569 typo, since it was named "eno" in the comments!
7571 4. The code that waits for the clock to tick (at a resolution of some fraction
7572 of a second) so as to ensure message-id uniqueness was always waiting for
7573 at least one whole tick, when it could have waited for less. [This is
7574 almost certainly not relevant at current processor speeds, where it is
7575 unlikely to ever wait at all. But we try to future-proof.]
7577 5. The function that sleeps for a time interval that includes fractions of a
7578 second contained a race. It did not block SIGALRM between setting the
7579 timer, and suspending (a couple of lines later). If the interval was short
7580 and the sigsuspend() was delayed until after it had expired, the suspension
7581 never ended. On busy systems this could lead to processes getting stuck for
7584 6. Some uncommon configurations may cause a lookup to happen in a queue runner
7585 process, before it forks any delivery processes. The open lookup caching
7586 mechanism meant that the open file or database connection was passed into
7587 the delivery process. The problem was that delivery processes always tidy
7588 up cached lookup data. This could cause a problem for the next delivery
7589 process started by the queue runner, because the external queue runner
7590 process does not know about the closure. So the next delivery process
7591 still has data in the lookup cache. In the case of a file lookup, there was
7592 no problem because closing a file descriptor in a subprocess doesn't affect
7593 the parent. However, if the lookup was caching a connection to a database,
7594 the connection was closed, and the second delivery process was likely to
7595 see errors such as "PGSQL: query failed: server closed the connection
7596 unexpectedly". The problem has been fixed by closing all cached lookups
7597 in a queue runner before running a delivery process.
7599 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
7600 seem to have the "const" qualifier which it has on other OS. I've
7603 8. Change 4.21/2 was too strict. It is only if there are two authenticators
7604 *of the same type* (client or server) with the same public name that an
7605 error should be diagnosed.
7607 9. When Exim looked up a host name for an IP address, but failed to find the
7608 original IP address when looking up the host name (a safety check), it
7609 output the message "<ip address> does not match any IP for NULL", which was
7610 confusing, to say the least. The bug was that the host name should have
7611 appeared instead of "NULL".
7613 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
7614 user that is built into the binary, and the -C or -D options is used, root
7615 privilege is dropped before the configuration file is read. In addition,
7616 logging is switched to stderr instead of the normal log files. If the
7617 configuration then re-defines the Exim user, the unprivileged environment
7618 is probably not what is expected, so Exim logs a panic warning message (but
7621 However, if deliver_drop_privilege is set, the unprivileged state may well
7622 be exactly what is intended, so the warning has been cut out in that case,
7623 and Exim is allowed to try to write to its normal log files.
7629 1. smtp_return_error_details was not giving details for temporary sender
7630 or receiver verification errors.
7632 2. Diagnose a configuration error if two authenticators have the same public
7635 3. Exim used not to create the message log file for a message until the first
7636 delivery attempt. This could be confusing when incoming messages were held
7637 for policy or load reasons. The message log file is now created at the time
7638 the message is received, and an initial "Received" line is written to it.
7640 4. The automatically generated man page for command line options had a minor
7641 bug that caused no ill effects; however, a more serious problem was that
7642 the procedure for building the man page automatically didn't always
7643 operate. Consequently, release 4.20 contains an out-of-date version. This
7644 shouldn't happen again.
7646 5. When building Exim with embedded Perl support, the script that builds the
7647 Makefile was calling 'perl' to find its compile-time parameters, ignoring
7648 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
7650 6. The freeze_tell option was not being used for messages that were frozen on
7651 arrival, either by an ACL or by local_scan().
7653 7. Added the smtp_incomplete_transaction log selector.
7655 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
7656 was accepting AUTH without a new EHLO.
7658 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
7659 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
7662 10. Logging of TCP/IP connections (when configured) now happens in the main
7663 daemon process instead of the child process, so that the TCP/IP connection
7664 count is more accurate (but it can never be perfect).
7666 11. The use of "drop" in a nested ACL was not being handled correctly in the
7667 outer ACL. Now, if condition failure induced by the nested "drop" causes
7668 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
7669 or "require"), the connection is dropped.
7671 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
7672 that yield "discard" can now be used with an "accept" or a "discard" verb,
7673 but an error is generated for any others (because I can't see a useful way
7674 to define what should happen).
7676 13. When an ACL is read dynamically from a file (or anywhere else), the lines
7677 are now processed in the same way as lines in the Exim configuration file.
7678 In particular, continuation lines are supported.
7680 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
7682 15. Added -ti meaning -t -i.
7684 16. Check for letters, digits, hyphens, and dots in the names of dnslist
7685 domains, and warn by logging if others are found.
7687 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
7688 returned from GIFCONF when Exim is trying to find the list of interfaces on
7689 a host. The code in os.c has been modified to copy each ifreq to an aligned
7690 structure in all cases.
7692 Also, in some cases, the returned ifreq's were being copied to a 'struct
7693 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
7694 means the last couple of bytes of an IPv6 address could be chopped if the
7695 ifreq contained only a normal sockaddr (14 bytes storage).
7697 18. Named domain lists were not supported in the hosts_treat_as_local option.
7698 An entry such as +xxxx was not recognized, and was treated as a literal
7701 19. Ensure that header lines added by a DATA ACL are included in the reject log
7702 if the ACL subsequently rejects the message.
7704 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
7705 MD5 (which is deprecated).
7707 21. When testing a filter file using -bf, Exim was writing a message when it
7708 took the sender from a "From " line in the message, but it was not doing so
7709 when it took $return_path from a Return-Path: header line. It now does.
7711 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
7712 with a valid header line field name (a series of printing characters
7713 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
7715 23. Changed "disc" in the source to "disk" to conform to the documentation and
7716 the book and for uniformity.
7718 24. Ignore Sendmail's -Ooption=value command line item.
7720 25. When execve() failed while trying to run a command in a pipe transport,
7721 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
7722 could be confused with a return value of 69 from the command itself. This
7723 has been changed to 127, the value the shell returns if it is asked to run
7724 a non-existent command. The wording for the related log line suggests a
7725 non-existent command as the problem.
7727 26. If received_header_text expands to an empty string, do not add a Received:
7728 header line to the message. (Well, it adds a token one on the spool, but
7729 marks it "old" so that it doesn't get used or transmitted.)
7731 27. Installed eximstats 1.28 (addition of -nt option).
7733 28. There was no check for failure on the call to getsockname() in the daemon
7734 code. This can fail if there is a shortage of resources on the system, with
7735 ENOMEM, for example. A temporary error is now given on failure.
7737 29. Contrary to the C standard, it seems that in some environments, the
7738 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
7739 program. Exim now does this explicitly; it affects the formatting of
7740 timestamps using strftime().
7742 30. If exiqsumm was given junk data, it threw up some uninitialized variable
7743 complaints. I've now initialized all the variables, to avoid this.
7745 32. Header lines added by a system filter were not being "seen" during
7746 transport-time rewrites.
7748 33. The info_callback() function passed to OpenSSL is set up with type void
7749 (*)(SSL *, int, int), as described somewhere. However, when calling the
7750 function (actually a macro) that sets it up, the type void(*)() is
7751 expected. I've put in a cast to prevent warnings from picky compilers.
7753 34. If a DNS black list lookup found a CNAME record, but there were no A
7754 records associated with the domain it pointed at, Exim crashed.
7756 35. If a DNS black list lookup returned more than one A record, Exim ignored
7757 all but the first. It now scans all returned addresses if a particular IP
7758 value is being sought. In this situation, the contents of the
7759 $dnslist_value variable are a list of all the addresses, separated by a
7762 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
7763 to accept a host name and all its aliases if the forward lookup for any of
7764 them yielded the IP address of the incoming connection. Now it accepts only
7765 those names whose forward lookup yields the correct IP address. Any other
7766 names are discarded. This closes a loophole whereby a rogue DNS
7767 administrator could create reverse DNS records to break through a
7768 wildcarded host restriction in an ACL.
7770 37. If a user filter or a system filter that ran in a subprocess used any of
7771 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
7772 the wrong values were passed to the pipe command ($thisaddress had the
7773 value of $0, $0 had the value of $1, etc). This bug was introduced by
7774 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
7776 38. Improved the line breaking for long SMTP error messages from ACLs.
7777 Previously, if there was no break point between 40 and 75 characters, Exim
7778 left the rest of the message alone. Two changes have been made: (a) I've
7779 reduced the minimum length to 35 characters; (b) if it can't find a break
7780 point between 35 and 75 characters, it looks ahead and uses the first one
7781 that it finds. This may give the occasional overlong line, but at least the
7782 remaining text gets split now.
7784 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
7785 file descriptors might be low, and that setting 1000 would always raise it.
7786 It turns out that in some environments, the limit is already over 1000 and
7787 that lowering it causes trouble. So now Exim takes care not to decrease it.
7789 40. When delivering a message, the value of $return_path is set to $sender_
7790 address at the start of routing (routers may change the value). By an
7791 oversight, this default was not being set up when an address was tested by
7792 -bt or -bv, which affected the outcome if any router or filter referred to
7795 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
7796 log only when "message" or "log_message" are set. However, if one of the
7797 conditions was an address verification, or a call to a nested ACL, the
7798 messages generated by the underlying test were being passed through. This
7799 no longer happens. The underlying message is available in $acl_verify_
7800 message for both "message" and "log_message" expansions, so it can be
7801 passed through if needed.
7803 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
7804 new expansion $bh_ to give the encoded byte string without charset
7805 translation. Translation happens only if iconv() is available; HAVE_ICONV
7806 indicates this at build time. HEADERS_CHARSET gives the charset to
7807 translate to; headers_charset can change it in the configuration, and
7808 "headers charset" can change it in an individual filter file.
7810 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
7811 that creates RFC 2047 encoded "words" labels them as that charset instead
7812 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
7813 expansion operator; (ii) when Exim creates a From: line for a local
7814 message; (iii) when a header line is rewritten to include a "phrase" part.
7816 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
7817 buggy, causing it to skip the first lines of messages whose message ID
7818 ended in 'D'. This would not have bitten before Exim release 4.14, because
7819 message IDs were unlikely to end in 'D' before then. The effect was to have
7820 incorrect size information for certain domains.
7822 45. #include "config.h" was missing at the start of the crypt16.c module. This
7823 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
7826 46. If there was a timeout during a "random" callout check, Exim treated it as
7827 a failure of the random address, and carried on sending RSET and the real
7828 address. If the delay was just some slowness somewhere, the response to the
7829 original RCPT would be taken as a response to RSET and so on, causing
7830 mayhem of various kinds.
7832 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
7833 when I implemented it. It didn't allow for the fact that some option values
7834 may legitimately be negative (e.g. size_addition), and it didn't even do
7835 the right test for positive values.
7837 48. Domain names in DNS records are case-independent. Exim always looks them up
7838 in lower case. Some resolvers return domain names in exactly the case they
7839 appear in the zone file, that is, they may contain uppercase letters. Not
7840 all resolvers do this - some return always lower case. Exim was treating a
7841 change of case by a resolver as a change of domain, similar to a widening
7842 of a domain abbreviation. This triggered its re-routing code and so it was
7843 trying to route what was effectively the same domain again. This normally
7844 caused routing to fail (because the router wouldn't handle the domain
7845 twice). Now Exim checks for this case specially, and just changes the
7846 casing of the domain that it ultimately uses when it transmits the message
7849 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
7852 50. If a filter generated a file delivery with a non-absolute name (possible if
7853 no home directory exists for the router), the forbid_file option was not
7856 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
7857 the existing equality matching.
7859 52. Exim was using ints instead of ino_t variables in some places where it was
7860 dealing with inode numbers.
7862 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
7863 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
7864 called TMPDIR, and if it finds it is different, it changes its value.
7866 54. The smtp_printf() function is now made available to local_scan() so
7867 additional output lines can be written before returning. There is also an
7868 smtp_fflush() function to enable the detection of a dropped connection.
7869 The variables smtp_input and smtp_batched_input are exported to
7872 55. Changed the default runtime configuration: the message "Unknown user"
7873 has been removed from the ACL, and instead placed on the localuser router,
7874 using the cannot_route_message feature. This means that any verification
7875 failures that generate their own messages won't get overridden. Similarly,
7876 the "Unrouteable address" message that was in the ACL for unverifiable
7877 relay addresses has also been removed.
7879 56. Added hosts_avoid_esmtp to the smtp transport.
7881 57. The exicyclog script was not checking for the esoteric option
7882 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7883 will work only if exicyclog is run under the appropriate euid.
7885 58. Following a discussion on the list, the rules by which Exim recognises line
7886 endings on incoming messages have been changed. The -dropcr and drop_cr
7887 options are now no-ops, retained only for backwards compatibility. The
7888 following line terminators are recognized: LF CRLF CR. However, special
7889 processing applies to CR:
7891 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7892 nor a local message in the state where . is a terminator.
7894 (ii) If a bare CR is encountered in a header line, an extra space is added
7895 after the line terminator so as not to end the header. The reasoning
7896 behind this is that bare CRs in header lines are most likely either
7897 to be mistakes, or people trying to play silly games.
7899 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7900 was being incorrectly given as 18 bytes larger than it should have been.
7901 This is a VOB (very old bug).
7903 60. This may never have affected anything current, but just in case it has:
7904 When the local host is found other than at the start of a list of hosts,
7905 the local host, those with the same MX, and any that follow, are discarded.
7906 When the list in question was part of a longer list of hosts, the following
7907 hosts (not currently being processed) were also being discarded. This no
7908 longer happens. I'm not sure if this situation could ever has previously
7911 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7914 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7915 header. This is something that is recommended in a new Internet Draft, and
7916 is something that is documented as being done by Sendmail. There are two
7917 possible values. For messages generated by the autoreply transport, Exim
7920 Auto-Submitted: auto-replied
7922 whereas for all other generated messages (e.g. bounces) it adds
7924 Auto-Submitted: auto-generated
7926 63. The "personal" condition in filters now includes a test for the
7927 Auto-Submitted: header. If it contains the string "auto-" the message it
7928 not considered personal.
7930 64. Added rcpt_include_affixes as a generic transport option.
7932 65. Added queue_only_override (default true).
7934 66. Added the syslog_duplication option.
7936 67. If what should have been the first header line of a message consisted of
7937 a space followed by a colon, Exim was mis-interpreting it as a header line.
7938 It isn't of course - it is syntactically invalid and should therefore be
7939 treated as the start of the message body. The misbehaviour could have
7940 caused a number of strange effects, including loss of data in subsequent
7941 header lines, and spool format errors.
7943 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7944 client host had authenticated. This control can now be exercised by an ACL
7945 for more flexibility.
7947 69. By default, callouts do not happen when testing with -bh. There is now a
7948 variant, -bhc, which does actually run the callout code, including
7949 consulting and updating the callout cache.
7951 70. Added support for saslauthd authentication, courtesy of Alexander
7954 71. If statvfs() failed on the spool or log directories while checking their
7955 size for availability, Exim confusingly gave the error "space shortage".
7956 Furthermore, in debugging mode it crashed with a floating point exception.
7957 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7958 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7959 serious problem, Exim now writes to the main and panic logs when this
7960 happens, with details of the failure. It then refuses to accept the
7961 incoming message, giving the message "spool directory problem" or "log
7962 directory problem" with a 421 code for SMTP messages.
7964 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7965 0, 1, and 2 exist, because some OS complain for execs without them (see
7966 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7967 descriptors. However, the code omitted to check that the open succeeded,
7968 causing mysterious errors if for some reason the permissions on /dev/null
7969 got screwed. Now Exim writes a message to the main and panic logs, and
7970 bombs out if it can't open /dev/null.
7972 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7973 interact so that it is all more flexible. It is supposed to remain
7974 backwards compatible. Also added extra_local_interfaces.
7976 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7977 to bomb out with an assertion failure - to the client this appears as a
7978 connection drop. This problem occurs in the part of the code that was taken
7979 from the Samba project. Fortunately, the assertion is in a very simple
7980 function, so I have fixed this by reproducing the function inline in the
7981 one place where it is called, and arranging for authentication to fail
7982 instead of killing the process with assert().
7984 75. The SPA client code was not working when the server requested OEM rather
7985 than Unicode encoding.
7987 76. Added code to make require_files with a specific uid setting more usable in
7988 the case where statting the file as root fails - usually a non-root-mounted
7989 NFS file system. When this happens and the failure is EACCES, Exim now
7990 forks a subprocess and does the per-uid checking as the relevant uid.
7992 77. Added process_log_path.
7994 78. If log_file_path was not explicitly set, a setting of check_log_space or
7995 check_log_inodes was ignored.
7997 79. If a space check for the spool or log partitions fails, the incident is now
7998 logged. Of course, in the latter case the data may get lost...
8000 80. Added the %p formatting code to string_format() so that it can be used to
8001 print addresses in debug_print(). Adjusted all the address printing in the
8002 debugging in store.c to use %p rather than %d.
8004 81. There was a concern that a line of code in smtp_in.c could overflow a
8005 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
8006 initially expressed, the concern was not well-founded, because trailing
8007 spaces are removed early. However, if the trailing spaces were followed by
8008 a NULL, they did not get removed, so the overflow was possible. Two fixes
8011 (a) I re-wrote the offending code in a cleaner fashion.
8012 (b) If an incoming SMTP command contains a NULL character, it is rejected
8015 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
8016 runs initgroups(), so that if the Exim user is in any additional groups,
8017 they will be used during message reception.
8023 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob