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.
239 JH/01 Avoid costly startup code when not strictly needed. This reduces time
240 for some exim process initialisations. It does mean that the logging
241 of TLS configuration problems is only done for the daemon startup.
243 JH/02 Early-pipelining support code is now included unless disabled in Makefile.
245 JH/03 DKIM verification defaults no long accept sha1 hashes, to conform to
246 RFC 8301. They can still be enabled, using the dkim_verify_hashes main
249 JH/04 Support CHUNKING from an smtp transport using a transport_filter, when
250 DKIM signing is being done. Previously a transport_filter would always
251 disable CHUNKING, falling back to traditional DATA.
253 JH/05 Regard command-line recipients as tainted.
255 JH/06 Bug 340: Remove the daemon pid file on exit, when due to SIGTERM.
257 JH/07 Bug 2489: Fix crash in the "pam" expansion condition. It seems that the
258 PAM library frees one of the arguments given to it, despite the
259 documentation. Therefore a plain malloc must be used.
261 JH/08 Bug 2491: Use tainted buffers for the transport smtp context. Previously
262 on-stack buffers were used, resulting in a taint trap when DSN information
263 copied from a received message was written into the buffer.
265 JH/09 Bug 2493: Harden ARC verify against Outlook, whick has been seen to mix
266 the ordering of its ARC headers. This caused a crash.
268 JH/10 Bug 2492: Use tainted memory for retry record when needed. Previously when
269 a new record was being constructed with information from the peer, a trap
272 JH/11 Bug 2494: Unset the default for dmarc_tld_file. Previously a naiive
273 installation would get error messages from DMARC verify, when it hit the
274 nonexistent file indicated by the default. Distros wanting DMARC enabled
275 should both provide the file and set the option.
276 Also enforce no DMARC verification for command-line sourced messages.
278 JH/12 Fix an uninitialised flag in early-pipelining. Previously connections
279 could, depending on the platform, hang at the STARTTLS response.
281 JH/13 Bug 2498: Reset a counter used for ARC verify before handling another
282 message on a connection. Previously if one message had ARC headers and
283 the following one did not, a crash could result when adding an
284 Authentication-Results: header.
286 JH/14 Bug 2500: Rewind some of the common-coding in string handling between the
287 Exim main code and Exim-related utities. The introduction of taint
288 tracking also did many adjustments to string handling. Since then, eximon
289 frequently terminated with an assert failure.
291 JH/15 When PIPELINING, synch after every hundred or so RCPT commands sent and
292 check for 452 responses. This slightly helps the inefficieny of doing
293 a large alias-expansion into a recipient-limited target. The max_rcpt
294 transport option still applies (and at the current default, will override
295 the new feature). The check is done for either cause of synch, and forces
296 a fast-retry of all 452'd recipients using a new MAIL FROM on the same
297 connection. The new facility is not tunable at this time.
299 JH/16 Fix the variables set by the gsasl authenticator. Previously a pointer to
300 library live data was being used, so the results became garbage. Make
301 copies while it is still usable.
303 JH/17 Logging: when the deliver_time selector ise set, include the DT= field
304 on delivery deferred (==) and failed (**) lines (if a delivery was
305 attemtped). Previously it was only on completion (=>) lines.
307 JH/18 Authentication: the gsasl driver not provides the $authN variables in time
308 for the expansion of the server_scram_iter and server_scram_salt options.
310 WB/01 SPF: DNS lookups for the obsolete SPF RR type done by the libspf2 library
311 are now specifically given a NO_DATA response without hitting the system
312 resolver. The library goes on to do the now-standard TXT lookup.
313 Use of dnsdb lookups is not affected.
315 JH/19 Bug 2507: Modules: on handling a dynamic-module (lookups) open failure,
316 only retrieve the errormessage once. Previously two calls to dlerror()
317 were used, and the second one (for mainlog/paniclog) retrieved null
320 JH/20 Taint checking: disallow use of tainted data for
321 - the appendfile transport file and directory options
322 - the pipe transport command
323 - the autoreply transport file, log and once options
324 - file names used by the redirect router (including filter files)
326 - paths used by single-key lookups
327 Previously this was permitted.
329 JH/21 Bug 2501: Fix init call in the heimdal authenticator. Previously it
330 adjusted the size of a major service buffer; this failed because the
331 buffer was in use at the time. Change to a compile-time increase in the
332 buffer size, when this authenticator is compiled into exim.
334 JH/22 Taint-checking: move to safe-mode taint checking on all platforms. The
335 previous fast-mode was untenable in the face of glibs using mmap to
336 support larger malloc requests.
338 PP/01 Update the openssl_options possible values through OpenSSL 1.1.1c.
339 New values supported, if defined on system where compiled:
340 allow_no_dhe_kex, cryptopro_tlsext_bug, enable_middlebox_compat,
341 no_anti_replay, no_encrypt_then_mac, prioritize_chacha, tlsext_padding
343 JH/23 Performance improvement in the initial phase of a two-pass queue run. By
344 running a limited number of proceses in parallel, a benefit is gained. The
345 amount varies with the platform hardware and load. The use of the option
346 queue_run_in_order means we cannot do this, as ordering becomes
349 JH/24 Bug 2524: fix the cyrus_sasl auth driver gssapi usage. A previous fix
350 had introduced a string-copy (for ensuring NUL-termination) which was not
351 appropriate for that case, which can include embedded NUL bytes in the
352 block of data. Investigation showed the copy to actually be needless, the
353 data being length-specified.
355 JH/25 Fix use of concurrent TLS connections under GnuTLS. When a callout was
356 done during a receiving connection, and both used TLS, global info was
357 used rather than per-connection info for tracking the state of data
358 queued for transmission. This could result in a connection hang.
360 JH/26 Fix use of the SIZE parameter on MAIL commands, on continued connections.
361 Previously, when delivering serveral messages down a single connection
362 only the first would provide a SIZE. This was due to the size information
363 not being properly tracked.
365 JH/27 Bug 2530: When operating in a timezone with sub-minute offset, such as
366 TAI (at 37 seconds currently), pretend to be in UTC for time-related
367 expansion and logging. Previously, spurious values such as a future
368 minute could be seen.
370 JH/28 Bug 2533: Fix expansion of ${tr } item. When called in some situations
371 it could crash from a null-deref. This could also affect the
372 ${addresses: } operator and ${readsock } item.
374 JH/29 Bug 2537: Fix $mime_part_count. When a single connection had a non-mime
375 message following a mime one, the variable was not reset.
377 JH/30 When an pipelined-connect fails at the first response, assume incorrect
378 cached capability (perhaps the peer reneged?) and immediately retry in
381 JH/31 Fix spurious detection of timeout while writing to transport filter.
383 JH/32 Bug 2541: Fix segfault on bad cmdline -f (sender) argument. Previously
384 an attempt to copy the string was made before checking it.
386 JH/33 Fix the dsearch lookup to return an untainted result. Previously the
387 taint of the lookup key was maintained; we now regard the presence in the
388 filesystem as sufficient validation.
390 JH/34 Fix the readsocket expansion to not segfault when an empty "options"
391 argument is supplied.
393 JH/35 The dsearch lookup now requires that the directory is an absolute path.
394 Previously this was not checked, and nonempty relative paths made an
395 access under Exim's current working directory.
397 JH/36 Bug 2554: Fix msg:defer event for the hosts_max_try_hardlimit case.
398 Previously no event was raised.
400 JH/37 Bug 2552: Fix the check on spool space during reception to use the SIZE
401 parameter supplied by the sender MAIL FROM command. Previously it was
402 ignored, and only the check_spool_space option value for the required
405 JH/38 Fix $dkim_key_length. This should, after a DKIM verification, present
406 the size of the signing public-key. Previously it was instead giving
407 the size of the signature hash.
409 JH/39 DKIM verification: the RFC 8301 restriction on sizes of RSA keys is now
410 the default. See the (new) dkim_verify_min_keysizes option.
412 JH/40 Fix a memory-handling bug: when a connection carried multiple messages
413 and an ACL use a lookup for checking either the local_part or domain,
414 stale data could be accessed. Ensure that variable references are
415 dropped between messages.
417 JH/41 Bug 2571: Fix SPA authenticator. Running as a server, an offset supplied
418 by the client was not checked as pointing within response data before
419 being used. A malicious client could thus cause an out-of-bounds read and
420 possibly gain authentication. Fix by adding the check.
422 JH/42 Internationalisation: change the default for downconversion in the smtp
423 transport to be "if needed". Previously it was "as previously set" for
424 the message, which usually meant "if needed" for message-submission but
425 "no" for everything else. However, MTAs have been seen using SMTPUTF8
426 even when the envelope addresses did not need it, resulting in forwarding
427 failures to non-supporting MTAs. A downconvert in such cases will be
428 a no-op on the addresses, merely dropping the use of SMTPUTF8 by the
429 transport. The change does mean that addresses needing conversion will
430 be converted when previously a delivery failure would occur.
432 JH/43 Fix possible long line in DSN. Previously when a very long SMTP error
433 response was received it would be used unchecked in a fail-DSN, violating
434 standards on line-length limits. Truncate if needed.
436 HS/01 Remove parameters of the link to www.open-spf.org. The linked form
437 doesn't work. (Additionally add a new main config option to configure the
444 JH/01 OpenSSL: With debug enabled output keying information sufficient, server
445 side, to decode a TLS 1.3 packet capture.
447 JH/02 OpenSSL: Suppress the sending of (stateful) TLS1.3 session tickets.
448 Previously the default library behaviour applied, sending two, each in
451 JH/03 Debug output for ACL now gives the config file name and line number for
454 JH/04 The default received_header_text now uses the RFC 8314 tls cipher clause.
456 JH/05 DKIM: ensure that dkim_domain elements are lowercased before use.
458 JH/06 Fix buggy handling of autoreply bounce_return_size_limit, and a possible
459 buffer overrun for (non-chunking) other transports.
461 JH/07 GnuTLS: Our use of late (post-handshake) certificate verification, under
462 TLS1.3, means that a server rejecting a client certificate is not visible
463 to the client until the first read of encrypted data (typically the
464 response to EHLO). Add detection for that case and treat it as a failed
465 TLS connection attempt, so that the normal retry-in-clear can work (if
466 suitably configured).
468 JB/01 Bug 2375: fix expansions of 822 addresses having comments in local-part
469 and/or domain. Found and fixed by Jason Betts.
471 JH/08 Add hardening against SRV & TLSA lookups the hit CNAMEs (a nonvalid
472 configuration). If a CNAME target was not a wellformed name pattern, a
475 JH/09 Logging: Fix initial listening-on line for multiple ports for an IP when
476 the OS reports them interleaved with other addresses.
478 JH/10 OpenSSL: Fix aggregation of messages. Previously, when PIPELINING was
479 used both for input and for a verify callout, both encrypted, SMTP
480 responses being sent by the server could be lost. This resulted in
481 dropped connections and sometimes bounces generated by a peer sending
484 JH/11 Harden plaintext authenticator against a badly misconfigured client-send
485 string. Previously it was possible to cause undefined behaviour in a
486 library routine (usually a crash). Found by "zerons".
488 JH/12 Bug 2384: fix "-bP smtp_receive_timeout". Previously it returned no
491 JH/13 Bug 2386: Fix builds with Dane under LibreSSL 2.9.0 onward. Some old
492 API was removed, so update to use the newer ones.
494 JH/14 Bug 1891: Close the log file if receiving a non-smtp message, without
495 any timeout set, is taking a long time. Previously we would hang on to a
496 rotated logfile "forever" if the input was arriving with long gaps
497 (a previous attempt to fix addressed lack, for a long time, of initial
500 HS/01 Bug 2390: Use message_id for tempfile creation to avoid races in a
501 shared (NFS) environment. The length of the tempfile name is now
502 4 + 16 ("hdr.$message_exim_id") which might break on file
503 systems which restrict the file name length to lower values.
506 HS/02 Bug 2390: Use message_id for tempfile creation to avoid races in a
507 shared (NFS) environment.
509 HS/03 Bug 2392: exigrep does case sensitive *option* processing (as it
510 did for all versions <4.90). Notably -M, -m, --invert, -I may be
513 JH/15 Use unsigned when creating bitmasks in macros, to avoid build errors
514 on some platforms for bit 31.
516 JH/16 GnuTLS: rework ciphersuite strings under recent library versions. Thanks
517 to changes apparently associated with TLS1.3 handling some of the APIs
518 previously used were either nonfunctional or inappropriate. Strings
519 like TLS1.3:ECDHE_SECP256R1__RSA_PSS_RSAE_SHA256__AES_256_GCM__AEAD:256
520 and TLS1.2:ECDHE_SECP256R1__RSA_SHA256__AES_128_CBC__SHA256:128 replace
521 the previous TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256 .
522 This affects log line X= elements, the $tls_{in,out}_cipher variables,
523 and the use of specific cipher names in the encrypted= ACL condition.
525 JH/17 OpenSSL: the default openssl_options now disables ssl_v3.
527 JH/18 GnuTLS: fix $tls_out_ocsp under hosts_request_ocsp. Previously the
528 verification result was not updated unless hosts_require_ocsp applied.
530 JH/19 Bug 2398: fix listing of a named-queue. Previously, even with the option
531 queue_list_requires_admin set to false, non-admin users were denied the
534 JH/20 Bug 2389: fix server advertising of usable certificates, under GnuTLS in
535 directory-of-certs mode. Previously they were advertised despite the
538 JH/21 The smtp transport option "hosts_noproxy_tls" is now unset by default.
539 A single TCP connection by a client will now hold a TLS connection open
540 for multiple message deliveries, by default. Previously the default was to
543 JH/22 The smtp transport option "hosts_try_dane" now enables all hosts by
544 default. If built with the facility, DANE will be used. The facility
545 SUPPORT_DANE is now enabled in the prototype build Makefile "EDITME".
547 JH/23 The build default is now for TLS to be included; the SUPPORT_TLS define
548 is replaced with DISABLE_TLS. Either USE_GNUTLS or (the new) USE_OPENSSL
549 must be defined and you must still, unless you define DISABLE_TLS, manage
550 the the include-dir and library-file requirements that go with that
551 choice. Non-TLS builds are still supported.
553 JH/24 Fix duplicated logging of peer name/address, on a transport connection-
556 JH/25 The smtp transport option "hosts_try_fastopen" now enables all hosts by
557 default. If the platform supports and has the facility enabled, it will
558 be requested on all coneections.
560 JH/26 The PIPE_CONNECT facility is promoted from experimental status and is now
561 controlled by the build-time option SUPPORT_PIPE_CONNECT.
563 PP/01 Unbreak heimdal_gssapi, broken in 4.92.
565 JH/27 Bug 2404: Use the main-section configuration option "dsn_from" for
566 success-DSN messages. Previously the From: header was always the default
567 one for these; the option was ignored.
569 JH/28 Fix the timeout on smtp response to apply to the whole response.
570 Previously it was reset for every read, so a teergrubing peer sending
571 single bytes within the time limit could extend the connection for a
572 long time. Credit to Qualsys Security Advisory Team for the discovery.
574 JH/29 Fix DSN Final-Recipient: field. Previously it was the post-routing
575 delivery address, which leaked information of the results of local
576 forwarding. Change to the original envelope recipient address, per
579 JH/30 Bug 2411: Fix DSN generation when RFC 3461 failure notification is
580 requested. Previously not bounce was generated and a log entry of
581 error ignored was made.
583 JH/31 Avoid re-expansion in ${sort } expansion. (CVE-2019-13917)
585 JH/32 Introduce a general tainting mechanism for values read from the input
586 channel, and values derived from them. Refuse to expand any tainted
587 values, to catch one form of exploit.
589 JH/33 Bug 2413: Fix dkim_strict option. Previously the expansion result
590 was unused and the unexpanded text used for the test. Found and
591 fixed by Ruben Jenster.
593 JH/34 Fix crash after TLS shutdown. When the TCP/SMTP channel was left open,
594 an attempt to use a TLS library read routine dereffed a nul pointer,
597 JH/35 Bug 2409: filter out-of-spec chars from callout response before using
598 them in our smtp response.
600 JH/36 Have the general router option retry_use_local_part default to true when
601 any of the restrictive preconditions are set (to anything). Previously it
602 was only for check_local user. The change removes one item of manual
603 configuration which is required for proper retries when a remote router
604 handles a subset of addresses for a domain.
606 JH/37 Appendfile: when evaluating quota use (non-quota_size_regex) take the file
607 link count into consideration.
609 HS/04 Fix handling of very log lines in -H files. If a -<key> <value> line
610 caused the extension of big_buffer, the following lines were ignored.
612 JH/38 Bug 1395: Teach the DNS negative-cache about TTL value from the SOA in
613 accordance with RFC 2308. Previously there was no expiry, so a longlived
614 receive process (eg. due to ACL delays) versus a short SOA value could
617 HS/05 Handle trailing backslash gracefully. (CVE-2019-15846)
619 JH/39 Promote DMARC support to mainline.
621 JH/40 Bug 2452: Add a References: header to DSNs.
623 JH/41 With GnuTLS 3.6.0 (and later) do not attempt to manage Diffie-Hellman
624 parameters. The relevant library call is documented as "Deprecated: This
625 function is unnecessary and discouraged on GnuTLS 3.6.0 or later. Since
626 3.6.0, DH parameters are negotiated following RFC7919."
628 HS/06 Change the default of dnssec_request_domains to "*"
630 JH/42 Bug 2545: Fix CHUNKING for all RCPT commands rejected. Previously we
631 carried on and emitted a BDAT command, even when PIPELINING was not
634 JH/43 Bug 2465: Fix taint-handling in dsearch lookup. Previously a nontainted
635 buffer was used for the filename, resulting in a trap when tainted
636 arguments (eg. $domain) were used.
638 JH/44 With OpenSSL 1.1.1 (onwards) disable renegotiation for TLS1.2 and below;
639 recommended to avoid a possible server-load attack. The feature can be
640 re-enabled via the openssl_options main cofiguration option.
642 JH/45 local_scan API: documented the current smtp_printf() call. This changed
643 for version 4.90 - adding a "more data" boolean to the arguments.
644 Bumped the ABI version number also, this having been missed previously;
645 release versions 4.90 to 4.92.3 inclusive were effectively broken in
646 respect of usage of smtp_printf() by either local_scan code or libraries
647 accessed via the ${dlfunc } expansion item. Both will need coding
648 adjustment for any calls to smtp_printf() to match the new function
649 signature; a FALSE value for the new argument is always safe.
651 JH/46 FreeBSD: fix use of the sendfile() syscall. The shim was not updating
652 the file-offset (which the Linux syscall does, and exim expects); this
653 resulted in an indefinite loop.
655 JH/47 ARC: fix crash in signing, triggered when a configuration error failed
656 to do ARC verification. The Authentication-Results: header line added
657 by the configuration then had no ARC item.
663 JH/01 Remove code calling the customisable local_scan function, unless a new
664 definition "HAVE_LOCAL_SCAN=yes" is present in the Local/Makefile.
666 JH/02 Bug 1007: Avoid doing logging from signal-handlers, as that can result in
667 non-signal-safe functions being used.
669 JH/03 Bug 2269: When presented with a received message having a stupidly large
670 number of DKIM-Signature headers, disable DKIM verification to avoid
671 a resource-consumption attack. The limit is set at twenty.
673 JH/04 Add variables $arc_domains, $arc_oldest_pass for ARC verify. Fix the
674 report of oldest_pass in ${authres } in consequence, and separate out
675 some descriptions of reasons for verification fail.
677 JH/05 Bug 2273: Cutthrough delivery left a window where the received messsage
678 files in the spool were present and unlocked. A queue-runner could spot
679 them, resulting in a duplicate delivery. Fix that by doing the unlock
680 after the unlink. Investigation by Tim Stewart. Take the opportunity to
681 add more error-checking on spoolfile handling while that code is being
684 PP/01 Refuse to open a spool data file (*-D) if it's a symlink.
685 No known attacks, no CVE, this is defensive hardening.
687 JH/06 Bug 2275: The MIME ACL unlocked the received message files early, and
688 a queue-runner could start a delivery while other operations were ongoing.
689 Cutthrough delivery was a common victim, resulting in duplicate delivery.
690 Found and investigated by Tim Stewart. Fix by using the open message data
691 file handle rather than opening another, and not locally closing it (which
692 releases a lock) for that case, while creating the temporary .eml format
693 file for the MIME ACL. Also applies to "regex" and "spam" ACL conditions.
695 JH/07 Bug 177: Make a random-recipient callout success visible in ACL, by setting
696 $sender_verify_failure/$recipient_verify_failure to "random".
698 JH/08 When generating a selfsigned cert, use serial number 1 since zero is not
701 JH/09 Bug 2274: Fix logging of cmdline args when starting in an unlinked cwd.
702 Previously this would segfault.
704 JH/10 Fix ARC signing for case when DKIM signing failed. Previously this would
707 JH/11 Bug 2264: Exim now only follows CNAME chains one step by default. We'd
708 like zero, since the resolver should be doing this for us, But we need one
709 as a CNAME but no MX presence gets the CNAME returned; we need to check
710 that doesn't point to an MX to declare it "no MX returned" rather than
711 "error, loop". A new main option is added so the older capability of
712 following some limited number of chain links is maintained.
714 JH/12 Add client-ip info to non-pass iprev ${authres } lines.
716 JH/13 For receent Openssl versions (1.1 onward) use modern generic protocol
717 methods. These should support TLS 1.3; they arrived with TLS 1.3 and the
718 now-deprecated earlier definitions used only specified the range up to TLS
719 1.2 (in the older-version library docs).
721 JH/14 Bug 2284: Fix DKIM signing for body lines starting with a pair of dots.
723 JH/15 Rework TLS client-side context management. Stop using a global, and
724 explicitly pass a context around. This enables future use of TLS for
725 connections to service-daemons (eg. malware scanning) while a client smtp
726 connection is using TLS; with cutthrough connections this is quite likely.
728 JH/16 Fix ARC verification to do AS checks in reverse order.
730 JH/17 Support a "tls" option on the ${readsocket } expansion item.
732 JH/18 Bug 2287: Fix the protocol name (eg utf8esmtp) for multiple messages
733 using the SMTPUTF8 option on their MAIL FROM commands, in one connection.
734 Previously the "utf8" would be re-prepended for every additional message.
736 JH/19 Reject MAIL FROM commands with SMTPUTF8 when the facility was not advertised.
737 Previously thery were accepted, resulting in issues when attempting to
738 forward messages to a non-supporting MTA.
740 PP/02 Let -n work with printing macros too, not just options.
742 JH/20 Bug 2296: Fix cutthrough for >1 address redirection. Previously only
743 one parent address was copied, and bogus data was used at delivery-logging
744 time. Either a crash (after delivery) or bogus log data could result.
745 Discovery and analysis by Tim Stewart.
747 PP/03 Make ${utf8clean:} expansion operator detect incomplete final character.
748 Previously if the string ended mid-character, we did not insert the
749 promised '?' replacement.
751 PP/04 Documentation: current string operators work on bytes, not codepoints.
753 JH/21 Change as many as possible of the global flags into one-bit bitfields; these
754 should pack well giving a smaller memory footprint so better caching and
755 therefore performance. Group the declarations where this can't be done so
756 that the byte-sized flag variables are not interspersed among pointer
757 variables, giving a better chance of good packing by the compiler.
759 JH/22 Bug 1896: Fix the envelope from for DMARC forensic reports to be possibly
760 non-null, to avoid issues with sites running BATV. Previously reports were
761 sent with an empty envelope sender so looked like bounces.
763 JH/23 Bug 2318: Fix the noerror command within filters. It wasn't working.
764 The ignore_error flag wasn't being returned from the filter subprocess so
765 was not set for later routers. Investigation and fix by Matthias Kurz.
767 JH/24 Bug 2310: Raise a msg:fail:internal event for each undelivered recipient,
768 and a msg:complete for the whole, when a message is manually removed using
769 -Mrm. Developement by Matthias Kurz, hacked on by JH.
771 JH/25 Avoid fixed-size buffers for pathnames in DB access. This required using
772 a "Gnu special" function, asprintf() in the DB utility binary builds; I
773 hope that is portable enough.
775 JH/26 Bug 2311: Fix DANE-TA verification under GnuTLS. Previously it was also
776 requiring a known-CA anchor certificate; make it now rely entirely on the
777 TLSA as an anchor. Checking the name on the leaf cert against the name
778 on the A-record for the host is still done for TA (but not for EE mode).
780 JH/27 Fix logging of proxy address. Previously, a pointless "PRX=[]:0" would be
781 included in delivery lines for non-proxied connections, when compiled with
782 SUPPORT_SOCKS and running with proxy logging enabled.
784 JH/28 Bug 2314: Fire msg:fail:delivery event even when error is being ignored.
785 Developement by Matthias Kurz, tweaked by JH. While in that bit of code,
786 move the existing event to fire before the normal logging of message
787 failure so that custom logging is bracketed by normal logging.
789 JH/29 Bug 2322: A "fail" command in a non-system filter (file) now fires the
790 msg:fail:internal event. Developement by Matthias Kurz.
792 JH/30 Bug 2329: Increase buffer size used for dns lookup from 2k, which was
793 far too small for todays use of crypto signatures stored there. Go all
794 the way to the max DNS message size of 64kB, even though this might be
795 overmuch for IOT constrained device use.
797 JH/31 Fix a bad use of a copy function, which could be used to pointlessly
798 copy a string over itself. The library routine is documented as not
799 supporting overlapping copies, and on MacOS it actually raised a SIGABRT.
801 JH/32 For main options check_spool_space and check_inode_space, where the
802 platform supports 64b integers, support more than the previous 2^31 kB
803 (i.e. more than 2 TB). Accept E, P and T multipliers in addition to
804 the previous G, M, k.
806 JH/33 Bug 2338: Fix the cyrus-sasl authenticator to fill in the
807 $authenticated_fail_id variable on authentication failure. Previously
810 JH/34 Increase RSA keysize of autogen selfsign cert from 1024 to 2048. RHEL 8.0
811 OpenSSL didn't want to use such a weak key. Do for GnuTLS also, and for
812 more-modern GnuTLS move from GNUTLS_SEC_PARAM_LOW to
813 GNUTLS_SEC_PARAM_MEDIUM.
815 JH/35 OpenSSL: fail the handshake when SNI processing hits a problem, server
816 side. Previously we would continue as if no SNI had been received.
818 JH/36 Harden the handling of string-lists. When a list consisted of a sole
819 "<" character, which should be a list-separator specification, we walked
820 off past the nul-terimation.
822 JH/37 Bug 2341: Send "message delayed" warning MDNs (restricted to external
823 causes) even when the retry time is not yet met. Previously they were
824 not, meaning that when (say) an account was over-quota and temp-rejecting,
825 and multiple senders' messages were queued, only one sender would get
826 notified on each configured delay_warning cycle.
828 JH/38 Bug 2351: Log failures to extract envelope addresses from message headers.
830 JH/39 OpenSSL: clear the error stack after an SSL_accept(). With anon-auth
831 cipher-suites, an error can be left on the stack even for a succeeding
832 accept; this results in impossible error messages when a later operation
835 AM/01 Bug 2359: GnuTLS: repeat lowlevel read and write operations while they
836 return error codes indicating retry. Under TLS1.3 this becomes required.
838 JH/40 Fix the feature-cache refresh for EXPERIMENTAL_PIPE_CONNECT. Previously
839 it only wrote the new authenticators, resulting in a lack of tracking of
840 peer changes of ESMTP extensions until the next cache flush.
842 JH/41 Fix the loop reading a message header line to check for integer overflow,
843 and more-often against header_maxsize. Previously a crafted message could
844 induce a crash of the recive process; now the message is cleanly rejected.
846 JH/42 Bug 2366: Fix the behaviour of the dkim_verify_signers option. It had
847 been totally disabled for all of 4.91. Discovery and fix by "Mad Alex".
853 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
854 When redis_servers is set to a list of > 1 element, and the Redis servers
855 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
856 case of MOVED into a DEFER case instead, thus moving the query onto the
857 next server in the list. For a cluster of N elements, all N servers must
858 be defined in redis_servers.
860 GF/02 Catch and remove uninitialized value warning in exiqsumm
861 Check for existence of @ARGV before looking at $ARGV[0]
863 JH/01 Replace the store_release() internal interface with store_newblock(),
864 which internalises the check required to safely use the old one, plus
865 the allocate and data copy operations duplicated in both (!) of the
866 extant use locations.
868 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
869 modifier. This matches the restriction on the commandline.
871 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
872 Previously only the last row was returned.
874 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
875 we assumed that tags in the header were well-formed, and parsed the
876 element content after inspecting only the first char of the tag.
877 Assumptions at that stage could crash the receive process on malformed
880 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
881 While running the DKIM ACL we operate on the Permanent memory pool so that
882 variables created with "set" persist to the DATA ACL. Also (at any time)
883 DNS lookups that fail create cache records using the Permanent pool. But
884 expansions release any allocations made on the current pool - so a dnsdb
885 lookup expansion done in the DKIM ACL releases the memory used for the
886 DNS negative-cache, and bad things result. Solution is to switch to the
887 Main pool for expansions.
888 While we're in that code, add checks on the DNS cache during store_reset,
889 active in the testsuite.
890 Problem spotted, and debugging aided, by Wolfgang Breyha.
892 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
893 When none of the hosts presented to a transport match an already-open
894 connection, close it and proceed with the list. Previously we would
895 queue the message. Spotted by Lena with Yahoo, probably involving
898 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
899 Previously a spurious "250 OK id=" response was appended to the proper
902 JH/08 The "support for" informational output now, which built with Content
903 Scanning support, has a line for the malware scanner interfaces compiled
904 in. Interface can be individually included or not at build time.
906 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
907 by the template makefile "src/EDITME". The "STREAM" support for an older
908 ClamAV interface method is removed.
910 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
911 rows affected is given instead).
913 JH/11 The runtime Berkeley DB library version is now additionally output by
914 "exim -d -bV". Previously only the compile-time version was shown.
916 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
917 SMTP connection. Previously, when one had more recipients than the
918 first, an abortive onward connection was made. Move to full support for
919 multiple onward connections in sequence, handling cutthrough connection
920 for all multi-message initiating connections.
922 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
923 routers. Previously, a multi-recipient message would fail to match the
924 onward-connection opened for the first recipient, and cause its closure.
926 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
927 a timeout on read on a GnuTLS initiating connection, resulting in the
928 initiating connection being dropped. This mattered most when the callout
929 was marked defer_ok. Fix to keep the two timeout-detection methods
932 JH/15 Relax results from ACL control request to enable cutthrough, in
933 unsupported situations, from error to silently (except under debug)
934 ignoring. This covers use with PRDR, frozen messages, queue-only and
937 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
939 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
940 metadata, resulting in a crash in free().
942 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
943 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
944 Broken also in d185889f4, with init system revamp.
946 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
947 Previously we abruptly closed the connection after reading a malware-
948 found indication; now we go on to read the "scan ok" response line,
951 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
952 ACL. Previously, a crash would result.
954 JH/19 Speed up macro lookups during configuration file read, by skipping non-
955 macro text after a replacement (previously it was only once per line) and
956 by skipping builtin macros when searching for an uppercase lead character.
958 JH/20 DANE support moved from Experimental to mainline. The Makefile control
959 for the build is renamed.
961 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
962 was allocated for every new TLS startup, meaning one per message. Fix
963 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
965 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
966 reported the original. Fix to report (as far as possible) the ACL
967 result replacing the original.
969 JH/23 Fix memory leak during multi-message connections using STARTTLS under
970 OpenSSL. Certificate information is loaded for every new TLS startup,
971 and the resources needed to be freed.
973 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
975 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
978 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
979 DATA response info to the (existing) per-recipient response info for
980 the "C=" log element. It can have useful tracking info from the
981 destination system. Patch from Simon Arlott.
983 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
984 length value. Previously this would segfault.
986 HS/02 Support Avast multiline protoocol, this allows passing flags to
987 newer versions of the scanner.
989 JH/28 Ensure that variables possibly set during message acceptance are marked
990 dead before release of memory in the daemon loop. This stops complaints
991 about them when the debug_store option is enabled. Discovered specifically
992 for sender_rate_period, but applies to a whole set of variables.
993 Do the same for the queue-runner and queue-list loops, for variables set
994 from spool message files. Do the same for the SMTP per-message loop, for
995 certain variables indirectly set in ACL operations.
997 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
998 as a multi-recipient message from a mailinglist manager). The coding had
999 an arbitrary cutoff number of characters while checking for more input;
1000 enforced by writing a NUL into the buffer. This corrupted long / fast
1001 input. The problem was exposed more widely when more pipelineing of SMTP
1002 responses was introduced, and one Exim system was feeding another.
1003 The symptom is log complaints of SMTP syntax error (NUL chars) on the
1004 receiving system, and refused recipients seen by the sending system
1005 (propating to people being dropped from mailing lists).
1006 Discovered and pinpointed by David Carter.
1008 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
1009 replaced by the ${authresults } expansion.
1011 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
1013 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
1014 allows proper process termination in container environments.
1016 JH/32 Bug 2258: Fix spool_wireformat in combination with LMTP transport.
1017 Previously the "final dot" had a newline after it; ensure it is CR,LF.
1019 JH/33 SPF: remove support for the "spf" ACL condition outcome values "err_temp"
1020 and "err_perm", deprecated since 4.83 when the RFC-defined words
1021 "temperror" and "permerror" were introduced.
1023 JH/34 Re-introduce enforcement of no cutthrough delivery on transports having
1024 transport-filters or DKIM-signing. The restriction was lost in the
1025 consolidation of verify-callout and delivery SMTP handling.
1026 Extend the restriction to also cover ARC-signing.
1028 JH/35 Cutthrough: for a final-dot response timeout (and nonunderstood responses)
1029 in defer=pass mode supply a 450 to the initiator. Previously the message
1032 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
1033 tls_require_ciphers is used as before.
1035 HS/03 Malware Avast: Better match the Avast multiline protocol. Add
1036 "pass_unscanned". Only tmpfails from the scanner are written to
1037 the paniclog, as they may require admin intervention (permission
1038 denied, license issues). Other scanner errors (like decompression
1039 bombs) do not cause a paniclog entry.
1041 JH/36 Fix reinitialisation of DKIM logging variable between messages.
1042 Previously it was possible to log spurious information in receive log
1045 JH/37 Bug 2255: Revert the disable of the OpenSSL session caching. This
1046 triggered odd behaviour from Outlook Express clients.
1048 PP/03 Add util/renew-opendmarc-tlds.sh script for safe renewal of public
1051 JH/38 DKIM: accept Ed25519 pubkeys in SubjectPublicKeyInfo-wrapped form,
1052 since the IETF WG has not yet settled on that versus the original
1053 "bare" representation.
1055 JH/39 Fix syslog logging for syslog_timestamp=no and log_selector +millisec.
1056 Previously the millisecond value corrupted the output.
1057 Fix also for syslog_pid=no and log_selector +pid, for which the pid
1058 corrupted the output.
1064 JH/01 Rework error string handling in TLS interface so that the caller in
1065 more cases is responsible for logging. This permits library-sourced
1066 string to be attached to addresses during delivery, and collapses
1067 pairs of long lines into single ones.
1069 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
1070 during configuration. Wildcards are allowed and expanded.
1072 JH/02 Rework error string handling in DKIM to pass more info back to callers.
1073 This permits better logging.
1075 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
1076 do not close it down and have the child transport start it up again on
1077 the passed-on TCP connection. Instead, proxy the child (and any
1078 subsequent ones) for TLS via a unix-domain socket channel. Logging is
1079 affected: the continued delivery log lines do not have any DNSSEC, TLS
1080 Certificate or OCSP information. TLS cipher information is still logged.
1082 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
1083 identical IP addresses on different listening ports. Will also affect
1086 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
1087 add noisy ifdef guards to special-case this sillyness.
1088 Patch from Bernd Kuhls.
1090 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
1091 than 255 are no longer allowed.
1093 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
1094 Disable the session-cache too, which might reduce our load. Since we
1095 currrectly use a new context for every connection, both as server and
1096 client, there is no benefit for these.
1097 GnuTLS appears to not support tickets server-side by default (we don't
1098 call gnutls_session_ticket_enable_server()) but client side is enabled
1099 by default on recent versions (3.1.3 +) unless the PFS priority string
1102 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
1103 <https://reproducible-builds.org/specs/source-date-epoch/>.
1105 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
1106 the check for any unsuccessful recipients did not notice the limit, and
1107 erroneously found still-pending ones.
1109 JH/08 Pipeline CHUNKING command and data together, on kernels that support
1110 MSG_MORE. Only in-clear (not on TLS connections).
1112 JH/09 Avoid using a temporary file during transport using dkim. Unless a
1113 transport-filter is involved we can buffer the headers in memory for
1114 creating the signature, and read the spool data file once for the
1115 signature and again for transmission.
1117 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
1118 4.77 as the kernel support then wasn't solid, having issues in 64bit
1119 mode. Now, it's been long enough. Add support for FreeBSD also.
1121 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
1122 case where the routing stage had gathered several addresses to send to
1123 a host before calling the transport for the first, we previously failed
1124 to close down TLS in the old transport process before passing the TCP
1125 connection to the new process. The new one sent a STARTTLS command
1126 which naturally failed, giving a failed delivery and bloating the retry
1127 database. Investigation and fix prototype from Wolfgang Breyha.
1129 JH/12 Fix check on SMTP command input synchronisation. Previously there were
1130 false-negatives in the check that the sender had not preempted a response
1131 or prompt from Exim (running as a server), due to that code's lack of
1132 awareness of the SMTP input buffering.
1134 PP/04 Add commandline_checks_require_admin option.
1135 Exim drops privileges sanely, various checks such as -be aren't a
1136 security problem, as long as you trust local users with access to their
1137 own account. When invoked by services which pass untrusted data to
1138 Exim, this might be an issue. Set this option in main configuration
1139 AND make fixes to the calling application, such as using `--` to stop
1142 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
1143 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
1144 responses to those, into a single TLS record each way (this usually means
1145 a single packet). As a side issue, smtp_enforce_sync now works on TLS
1148 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
1149 affects you only if you're dancing at the edge of the param size limits.
1150 If you are, and this message makes sense to you, then: raise the
1151 configured limit or use OpenSSL 1.1. Nothing we can do for older
1154 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
1155 cmdline element to get the documented default one. Previously it was
1158 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
1159 are unsuable for non-message-specific reasons - eg. connection timeout,
1160 banner-time rejection.
1162 JH/16 Fix logging of delivery remote port, when specified by router, under
1165 PP/06 Repair manualroute's ability to take options in any order, even if one
1166 is the name of a transport.
1169 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
1171 JH/17 Change the list-building routines interface to use the expanding-string
1172 triplet model, for better allocation and copying behaviour.
1174 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
1175 Previously it was constructed the first time a possibly-matching string
1176 was met in the configuration file input during startup; now it is done
1179 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
1180 compatible one, to avoid the (poorly documented) possibility of a config
1181 file in the working directory redirecting the DB files, possibly correpting
1182 some existing file. CVE-2017-10140 assigned for BDB.
1184 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
1185 cache-hot. Previously, although the result was properly cached, the
1186 initial verify call returned a defer.
1188 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
1189 the main verify for receipient in uncached-mode.
1191 JH/22 Retire historical build files to an "unsupported" subdir. These are
1192 defined as "ones for which we have no current evidence of testing".
1194 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
1195 if present. Previously it was ignored.
1197 JH/24 Start using specified-initialisers in C structure init coding. This is
1198 a C99 feature (it's 2017, so now considered safe).
1200 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
1201 if was a fixed-sized field and bitmask ops via macros; it is now more
1204 PP/07 GitHub PR 56: Apply MariaDB build fix.
1205 Patch provided by Jaroslav Škarvada.
1207 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
1208 during Coverity cleanups [4.87 JH/47]
1209 Diagnosis and fix provided by Michael Fischer v. Mollard.
1211 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
1212 the right size to place the terminating semicolon on its own folded
1213 line, the header hash was calculated to an incorrect value thanks to
1214 the (relaxed) space the fold became.
1216 HS/02 Fix Bug 2130: large writes from the transport subprocess were chunked
1217 and confused the parent.
1219 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
1220 which could crash as a result. This could lead to undeliverable messages.
1222 JH/28 Logging: "next input sent too soon" now shows where input was truncated
1225 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
1226 matters on fast-turnover and PID-randomising systems, which were getting
1227 out-of-order delivery.
1229 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
1230 a possibly-overlapping copy. The symptom was that "Remote host closed
1231 connection in response to HELO" was logged instead of the actual 4xx
1234 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
1235 Previously only that bufferd was discarded, resulting in SYMTP command
1238 JH/32 DKIM: when a message has multiple signatures matching an identity given
1239 in dkim_verify_signers, run the dkim acl once for each. Previously only
1240 one run was done. Bug 2189.
1242 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
1243 "panic the current process" to "deliberately defer". The panic log is
1244 still written with the problem list name; the mail and reject logs now
1245 get a temp-reject line for the message that was being handled, saying
1246 something like "domains check lookup or other defer". The SMTP 451
1247 message is still "Temporary local problem".
1249 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
1250 A crafted sequence of BDAT commands could result in in-use memory beeing
1251 freed. CVE-2017-16943.
1253 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
1254 from SMTP input. Previously it was always done; now only done for DATA
1255 and not BDAT commands. CVE-2017-16944.
1257 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
1258 to the message (such as an overlong header line). Previously this was
1259 not done and we did not exit BDAT mode. Followon from the previous item
1260 though a different problem.
1266 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
1267 than -2003 did; needs libidn2 in addition to libidn.
1269 JH/02 The path option on a pipe transport is now expanded before use.
1271 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
1272 Patch provided by "Björn", documentation fix added too.
1274 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
1275 missing a wire-to-host endian conversion.
1277 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
1278 close after a BDAT command line could be taken as a following command,
1279 giving a synch failure. Fix by only checking for synch immediately
1280 before acknowledging the chunk.
1282 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
1283 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
1284 macro. Patches provided by Josh Soref.
1286 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
1287 Previously we did not; the RFC seems ambiguous and VRFY is not listed
1288 by IANA as a service extension. However, John Klensin suggests that we
1291 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
1292 the dkim code may be unix-mode line endings rather than smtp wire-format
1293 CRLF, so prepend a CR to any bare LF.
1295 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
1296 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
1298 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
1299 the first were themselves being wrongly included in the feed into dkim
1300 processing; with most chunk sizes in use this resulted in an incorrect
1301 body hash calculated value.
1303 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
1304 DKIM signature block, for verification. Although advised against by
1305 standards it is specifically not ruled illegal.
1307 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
1309 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
1310 missing a body hash (the bh= tag).
1312 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
1313 It seems that HAProxy sends the Proxy Protocol information in clear and
1314 only then does a TLS startup, so do the same.
1316 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
1317 TCP connections (such as for Spamd) unless the daemon successfully set
1318 Fast Open mode on its listening sockets. This fixes breakage seen on
1319 too-old kernels or those not configured for Fast Open, at the cost of
1320 requiring both directions being enabled for TFO, and TFO never being used
1321 by non-daemon-related Exim processes.
1323 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
1324 endings, at least on the first header line. Try to canonify any that get
1325 past that check, despite the cost.
1327 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
1328 now limited to an arbitrary five deep, while parsing addresses with the
1329 strip_excess_angle_brackets option enabled.
1331 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
1332 instead leave the unprompted TLS handshake in socket buffer for the
1333 TLS library to consume.
1335 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
1337 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
1339 JH/16 Drop variables when they go out of scope. Memory management drops a whole
1340 region in one operation, for speed, and this leaves assigned pointers
1341 dangling. Add checks run only under the testsuite which checks all
1342 variables at a store-reset and panics on a dangling pointer; add code
1343 explicitly nulling out all the variables discovered. Fixes one known
1344 bug: a transport crash, where a dangling pointer for $sending_ip_address
1345 originally assigned in a verify callout, is re-used.
1347 PP/06 Drop '.' from @INC in various Perl scripts.
1349 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
1351 PP/08 Reduce a number of compilation warnings under clang; building with
1352 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
1353 should be warning-free.
1355 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
1357 HS/01 Fix portability problems introduced by PP/08 for platforms where
1358 realloc(NULL) is not equivalent to malloc() [SunOS et al].
1360 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
1361 chunk. This allows us to accept broken chunked messages. We need a more
1362 general solution here.
1364 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
1365 already-broken messages in the queue.
1367 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
1369 JH/19 Fix reference counting bug in routing-generated-address tracking.
1375 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
1376 supports it and a size is available (ie. the sending peer gave us one).
1378 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
1379 years of being deprecated). The replacements are the ACLs
1380 acl_smtp_mime and acl_not_smtp_mime.
1382 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
1383 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
1384 or even an in-clear connection were permitted. Now, if the host lookup
1385 was dnssec and dane was requested then the host is only used if the
1386 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
1387 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
1388 if one fails this test.
1389 This means that a poorly-configured remote DNS will make it incommunicado;
1390 but it protects against a DNS-interception attack on it.
1392 JH/04 Bug 1810: make continued-use of an open smtp transport connection
1393 non-noisy when a race steals the message being considered.
1395 JH/05 If main configuration option tls_certificate is unset, generate a
1396 self-signed certificate for inbound TLS connections.
1398 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
1399 in rewrites and routers.
1401 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
1402 and logged a warning sing 4.83; now they are a configuration file error.
1404 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
1405 (lacking @domain). Apply the same qualification processing as RCPT.
1407 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
1409 JH/10 Support ${sha256:} applied to a string (as well as the previous
1412 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
1413 a cutthrough deliver is pending, as we always want to make a connection.
1414 This also avoids re-routing the message when later placing the cutthrough
1415 connection after a verify cache hit.
1416 Do not update it with the verify result either.
1418 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
1419 when routing results in more than one destination address.
1421 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
1422 signing (which inhibits the cutthrough capability). Previously only
1423 the presence of an option was tested; now an expansion evaluating as
1424 empty is permissible (obviously it should depend only on data available
1425 when the cutthrough connection is made).
1427 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
1428 the relevant preceding SMTP command did not note the pipelining mode.
1430 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
1431 Previously they were not counted.
1433 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
1434 as one having no matching records. Previously we deferred the message
1435 that needed the lookup.
1437 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
1438 distinguished as "(=".
1440 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
1441 for missing MX records. Previously it only worked for missing A records.
1443 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
1445 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
1446 after the data-go-ahead and data-ack. Patch from Jason Betts.
1448 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
1449 even for a "none" policy. Patch from Tony Meyer.
1451 JH/22 Fix continued use of a connection for further deliveries. If a port was
1452 specified by a router, it must also match for the delivery to be
1455 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
1456 When one of the recipients of a message was unsuitable for the connection
1457 (has no matching addresses), we lost track of needing to mark it
1458 deferred. As a result mail would be lost.
1460 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
1462 JH/25 Decoding ACL controls is now done using a binary search; the source code
1463 takes up less space and should be simpler to maintain. Merge the ACL
1464 condition decode tables also, with similar effect.
1466 JH/26 Fix problem with one_time used on a redirect router which returned the
1467 parent address unchanged. A retry would see the parent address marked as
1468 delivered, so not attempt the (identical) child. As a result mail would
1471 JH/27 Fix a possible security hole, wherein a process operating with the Exim
1472 UID can gain a root shell. Credit to http://www.halfdog.net/ for
1473 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
1476 JH/28 Enable {spool,log} filesystem space and inode checks as default.
1477 Main config options check_{log,spool}_{inodes,space} are now
1478 100 inodes, 10MB unless set otherwise in the configuration.
1480 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
1481 Previously it only applied to the main-section connection policy
1484 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
1486 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
1487 by me. Added RFC7919 DH primes as an alternative.
1489 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
1490 are not in the system include path.
1492 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
1493 GnuTLS, when a session startup failed (eg because the client disconnected)
1494 Exim did stdio operations after fclose. This was exposed by a recent
1495 change which nulled out the file handle after the fclose.
1497 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
1498 signed directly by the cert-signing cert, rather than an intermediate
1499 OCSP-signing cert. This is the model used by LetsEncrypt.
1501 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
1503 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
1504 an incoming connection.
1506 HS/02 Bug 1802: Do not half-close the connection after sending a request
1509 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
1510 fallback to "prime256v1".
1512 JH/34 SECURITY: Use proper copy of DATA command in error message.
1513 Could leak key material. Remotely exploitable. CVE-2016-9963.
1519 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
1520 and 3.4.4 - once the server is enabled to respond to an OCSP request
1521 it does even when not requested, resulting in a stapling non-aware
1522 client dropping the TLS connection.
1524 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
1525 support variable-length bit vectors. No functional change.
1527 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
1528 The I= interface field on outgoing lines is now after the H= remote
1529 host field, same as incoming lines. There is a separate
1530 outgoing_interface log selector which allows you to disable the
1533 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
1534 If not running log_selector +smtp_connection the mainlog would be held
1535 open indefinitely after a "too many connections" event, including to a
1536 deleted file after a log rotate. Leave the per net connection logging
1537 leaving it open for efficiency as that will be quickly detected by the
1538 check on the next write.
1540 HS/01 Bug 1671: Fix post transport crash.
1541 Processing the wait-<transport> messages could crash the delivery
1542 process if the message IDs didn't exist for some reason. When
1543 using 'split_spool_directory=yes' the construction of the spool
1544 file name failed already, exposing the same netto behaviour.
1546 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
1547 mime_regex ACL conditions.
1549 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
1550 to DSN fail messages (bounces): remote IP, remote greeting, remote response
1551 to HELO, local diagnostic string.
1553 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
1554 log line to debug. Even when configured with a tls authenticator many
1555 client connections are expected to not authenticate in this way, so
1556 an authenticate fail is not an error.
1558 HS/02 Add the Exim version string to the process info. This way exiwhat
1559 gives some more detail about the running daemon.
1561 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
1562 matter for fast-change records such as DNSBLs.
1564 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
1565 retry record, even if constant. There may be multiple transports with
1566 different interface settings and the retry behaviour needs to be kept
1569 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
1571 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
1573 JH/10 Bug 840: fix log_defer_output option of pipe transport
1575 JH/11 Bug 830: use same host for all RCPTS of a message, even under
1576 hosts_randomize. This matters a lot when combined with mua_wrapper.
1578 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
1579 ${quote_pgsql:<string>} operator.
1581 JH/13 Bug 1708: avoid misaligned access in cached lookup.
1583 JH/14 Change header file name for freeradius-client. Relevant if compiling
1584 with Radius support; from the Gentoo tree and checked under Fedora.
1586 JH/15 Bug 1712: Introduce $prdr_requested flag variable
1588 JH/16 Bug 1714: Permit an empty string as expansion result for transport
1589 option transport_filter, meaning no filtering.
1591 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
1593 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
1594 defaults to "*" (all hosts). The variable is now available when not built
1595 with TLS, default unset, mainly to enable keeping the testsuite sane.
1596 If a server certificate is not supplied (via tls_certificate) an error is
1597 logged, and clients will find TLS connections fail on startup. Presumably
1598 they will retry in-clear.
1599 Packagers of Exim are strongly encouraged to create a server certificate
1600 at installation time.
1602 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
1603 with the $config_file variable.
1605 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
1606 in transport context, after the attempt, and per-recipient. The latter type
1607 is per host attempted. The event data is the error message, and the errno
1608 information encodes the lookup type (A vs. MX) used for the (first) host,
1609 and the trailing two digits of the smtp 4xx response.
1611 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
1612 to write to mainlog (or rejectlog, paniclog) in the window between file
1613 creation and permissions/ownership being changed. Particularly affects
1614 installations where exicyclog is run as root, rather than exim user;
1615 result is that the running daemon panics and dies.
1617 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
1619 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
1620 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
1621 "pri" and "weight". Note that the previous implicit priority given by the
1622 list order is no longer honoured.
1624 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
1625 for DKIM processing.
1627 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
1628 by defining SUPPORT_SOCKS.
1630 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
1631 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
1632 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
1633 variables are renamed to proxy_{local,external}_{address,port}.
1635 JH/27 Move Internationalisation support from Experimental to mainline, enabled
1636 for a build by defining SUPPORT_I18N
1638 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
1639 of the query string, and make ${quote_redis:} do that quoting.
1641 JH/29 Move Events support from Experimental to mainline, enabled by default
1642 and removable for a build by defining DISABLE_EVENT.
1644 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
1646 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
1647 cached by the daemon.
1649 JH/32 Move Redis support from Experimental to mainline, enabled for a build
1650 by defining LOOKUP_REDIS. The libhiredis library is required.
1652 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
1653 keys are given for lookup.
1655 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
1656 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
1657 only supported when built with TLS support. The PolarSSL SHA routines
1658 are still used when the TLS library is too old for convenient support.
1660 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
1661 openssl_options), for security. OpenSSL forces this from version 1.1.0
1662 server-side so match that on older versions.
1664 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
1665 allocation for $value could be released as the expansion processing
1666 concluded, but leaving the global pointer active for it.
1668 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
1669 and to use the domains and local_parts ACL conditions.
1671 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
1672 incorrectly not doubled on cutthrough transmission, hence seen as a
1673 body-termination at the receiving system - resulting in truncated mails.
1674 Commonly the sender saw a TCP-level error, and retransmitted the message
1675 via the normal store-and-forward channel. This could result in duplicates
1676 received - but deduplicating mailstores were liable to retain only the
1677 initial truncated version.
1679 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
1681 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
1683 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
1684 we're in there, support oversigning also; bug 1309.
1686 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
1688 HS/04 Add support for keep_environment and add_environment options.
1690 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
1691 either intentional arithmetic overflow during PRNG, or testing config-
1694 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
1695 delivery resulted in actual delivery. Cancel cutthrough before DATA
1698 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
1699 rejects a recipient: pass the reject to the originator.
1701 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
1702 Many were false-positives and ignorable, but it's worth fixing the
1705 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
1706 for the new environment-manipulation done at startup. Move the routines
1707 from being local to tls.c to being global via the os.c file.
1709 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
1710 an extract embedded as result-arg for a map, the first arg for extract
1711 is unavailable so we cannot tell if this is a numbered or keyed
1712 extraction. Accept either.
1718 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
1721 JH/02 The smtp transport option "multi_domain" is now expanded.
1723 JH/03 The smtp transport now requests PRDR by default, if the server offers
1726 JH/04 Certificate name checking on server certificates, when exim is a client,
1727 is now done by default. The transport option tls_verify_cert_hostnames
1728 can be used to disable this per-host. The build option
1729 EXPERIMENTAL_CERTNAMES is withdrawn.
1731 JH/05 The value of the tls_verify_certificates smtp transport and main options
1732 default to the word "system" to access the system default CA bundle.
1733 For GnuTLS, only version 3.0.20 or later.
1735 JH/06 Verification of the server certificate for a TLS connection is now tried
1736 (but not required) by default. The verification status is now logged by
1737 default, for both outbound TLS and client-certificate supplying inbound
1740 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
1743 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
1744 Status Notification (bounce) messages are now MIME format per RFC 3464.
1745 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
1746 under the control of the dsn_advertise_hosts option, and routers may
1747 have a dsn_lasthop option.
1749 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
1750 default, modifiable by a malware= option. The list separator for
1751 the options can now be changed in the usual way. Bug 68.
1753 JH/10 The smtp_receive_timeout main option is now expanded before use.
1755 JH/11 The incoming_interface log option now also enables logging of the
1756 local interface on delivery outgoing connections.
1758 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
1759 if the interface and destination host and port all match.
1761 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
1764 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
1765 Patch from Andrew Lewis.
1767 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
1768 now supports optional time-restrictions, weighting, and priority
1769 modifiers per server. Patch originally by <rommer@active.by>.
1771 JH/16 The spamd_address main option now supports a mixed list of local
1772 and remote servers. Remote servers can be IPv6 addresses, and
1773 specify a port-range.
1775 JH/17 Bug 68: The spamd_address main option now supports an optional
1776 timeout value per server.
1778 JH/18 Bug 1581: Router and transport options headers_add/remove can
1779 now have the list separator specified.
1781 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
1784 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
1787 JH/21 Support for the A6 type of dns record is withdrawn.
1789 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
1790 rather than the verbs used.
1792 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
1793 from 255 to 1024 chars.
1795 JH/24 Verification callouts now attempt to use TLS by default.
1797 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
1798 are generic router options now. The defaults didn't change.
1800 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
1801 Original patch from Alexander Shikoff, worked over by JH.
1803 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
1804 files if ZCAT_COMMAND is not executable.
1806 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
1808 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
1810 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
1811 Normally benign, it bites when the pair was led to by a CNAME;
1812 modern usage is to not canonicalize the domain to a CNAME target
1813 (and we were inconsistent anyway for A-only vs AAAA+A).
1815 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
1817 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
1818 when evaluating $sender_host_dnssec.
1820 JH/31 Check the HELO verification lookup for DNSSEC, adding new
1821 $sender_helo_dnssec variable.
1823 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
1825 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
1827 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
1829 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
1830 documented as working, but never had. Support all but $spam_report.
1832 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
1833 added for tls authenticator.
1835 HS/03 Add perl_taintmode main config option
1841 TL/01 When running the test suite, the README says that variables such as
1842 no_msglog_check are global and can be placed anywhere in a specific
1843 test's script, however it was observed that placement needed to be near
1844 the beginning for it to behave that way. Changed the runtest perl
1845 script to read through the entire script once to detect and set these
1846 variables, reset to the beginning of the script, and then run through
1847 the script parsing/test process like normal.
1849 TL/02 The BSD's have an arc4random API. One of the functions to induce
1850 adding randomness was arc4random_stir(), but it has been removed in
1851 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
1852 function when detected.
1854 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
1855 cause callback expansion.
1857 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
1858 syntax errors in an expansion can be treated as a string instead of
1859 logging or causing an error, due to the internal use of bool_lax
1860 instead of bool when processing it.
1862 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
1863 server certificates when making smtp deliveries.
1865 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
1867 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
1869 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1871 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1872 Merged patch from Sebastian Wiedenroth.
1874 JH/05 Fix results-pipe from transport process. Several recipients, combined
1875 with certificate use, exposed issues where response data items split
1876 over buffer boundaries were not parsed properly. This eventually
1877 resulted in duplicates being sent. This issue only became common enough
1878 to notice due to the introduction of connection certificate information,
1879 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1881 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1882 size buffer was used, resulting in syntax errors when an expansion
1885 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1886 version 3.3.6 or later.
1888 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1889 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1890 both become "event_action", the variables become $event_name, $event_data
1891 and $event_defer_errno. There is a new variable $verify_mode, usable in
1892 routers, transports and related events. The tls:cert event is now also
1893 raised for inbound connections, if the main configuration event_action
1896 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1897 early OCSP support, but no stapling (appears to be less than 1.0.0).
1899 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1900 server certificate names available under the smtp transport option
1901 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1904 JH/10 Time-related extraction expansions from certificates now use the main
1905 option "timezone" setting for output formatting, and are consistent
1906 between OpenSSL and GnuTLS compilations. Bug 1541.
1908 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1909 encoded parameter in the incoming message. Bug 1558.
1911 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1912 include certificate info, eximon was claiming there were spoolfile
1915 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1917 JH/14 Log delivery-related information more consistently, using the sequence
1918 "H=<name> [<ip>]" wherever possible.
1920 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1921 are problematic for Debian distribution, omit them from the release
1924 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1926 JH/16 Fix string representation of time values on 64bit time_t architectures.
1929 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1930 output list separator was used.
1935 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1936 checkers that were complaining about end of non-void function with no
1939 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1940 This was a regression introduced in 4.83 by another bugfix.
1942 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1944 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1945 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1951 TF/01 Correctly close the server side of TLS when forking for delivery.
1953 When a message was received over SMTP with TLS, Exim failed to clear up
1954 the incoming connection properly after forking off the child process to
1955 deliver the message. In some situations the subsequent outgoing
1956 delivery connection happened to have the same fd number as the incoming
1957 connection previously had. Exim would try to use TLS and fail, logging
1958 a "Bad file descriptor" error.
1960 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1961 utilities have not been installed.
1963 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1964 temporary space as the ACL may create new global variables.
1966 TL/01 LDAP support uses per connection or global context settings, depending
1967 upon the detected version of the libraries at build time.
1969 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1970 to extract and use the src ip:port in logging and expansions as if it
1971 were a direct connection from the outside internet. PPv2 support was
1972 updated based on HAProxy spec change in May 2014.
1974 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1976 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1977 Properly escape header and check for NULL return.
1979 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1982 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1984 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1985 characters in header names, implemented as a verify condition.
1986 Contributed by Michael Fischer v. Mollard.
1988 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1989 results permerror and temperror. Previous values are deprecated but
1990 still accepted. In a future release, err_perm and err_temp will be
1991 completely removed, which will be a backward incompatibility if the
1992 ACL tests for either of these two old results. Patch contributed by
1993 user bes-internal on the mailing list.
1995 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1997 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1998 selectors, in both main and reject logs.
2000 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
2003 JH/07 Add malware type "sock" for talking to simple daemon.
2005 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
2007 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
2008 routers/transports under cutthrough routing.
2010 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
2011 numbers. Touch up "bool" conditional to keep the same definition.
2013 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
2015 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
2017 JH/12 Expand items in router/transport headers_add or headers_remove lists
2018 individually rather than the list as a whole. Bug 1452.
2020 Required for reasonable handling of multiple headers_ options when
2021 they may be empty; requires that headers_remove items with embedded
2022 colons must have them doubled (or the list-separator changed).
2024 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
2025 view the policy declared in the DMARC record. Currently, $dmarc_status
2026 is a combined value of both the record presence and the result of the
2029 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
2031 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
2032 dnslookup router and the smtp transport (applying to the forward
2035 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
2036 of ldap servers used for a specific lookup. Patch provided by Heiko
2039 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
2040 New variable $lookup_dnssec_authenticated for observability.
2042 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
2043 Patch submitted by Lars Timman.
2045 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
2047 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
2048 Requires trusted mode and valid format message id, aborts otherwise.
2049 Patch contributed by Heiko Schlichting.
2051 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
2052 certextract with support for various fields. Bug 1358.
2054 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
2055 is requested by default, modifiable by smtp transport option
2058 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
2059 operate on certificate variables to give certificate fingerprints
2060 Also new ${sha256:cert_variable}.
2062 JH/23 The PRDR feature is moved from being Experimental into the mainline.
2064 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
2065 Christian Aistleitner.
2067 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
2069 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
2070 file. Patch from Wolfgang Breyha.
2072 JH/25 Expand the coverage of the delivery $host and $host_address to
2073 client authenticators run in verify callout. Bug 1476.
2075 JH/26 Port service names are now accepted for tls_on_connect_ports, to
2076 align with daemon_smtp_ports. Bug 72.
2078 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
2079 support and error reporting did not work properly.
2081 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
2082 and is readable. Patch from Andrew Colin Kissa.
2084 TL/14 Enhance documentation of ${run expansion and how it parses the
2085 commandline after expansion, particularly in the case when an
2086 unquoted variable expansion results in an empty value.
2088 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
2090 PP/02 Fix internal collision of T_APL on systems which support RFC3123
2091 by renaming away from it. Addresses GH issue 15, reported by
2094 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
2096 TL/15 SECURITY: prevent double expansion in math comparison functions
2097 (can expand unsanitized data). Not remotely exploitable.
2104 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
2106 PP/02 Make -n do something, by making it not do something.
2107 When combined with -bP, the name of an option is not output.
2109 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
2112 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
2113 $sender_host_name and config options to manage this, and basic check
2116 PP/05 DSCP support for outbound connections and control modifier for inbound.
2118 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
2119 (Only plugin which currently uses this is kerberos4, which nobody should
2120 be using, but we should make it available and other future plugins might
2121 conceivably use it, even though it would break NAT; stuff *should* be
2122 using channel bindings instead).
2124 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
2125 name; added for Sendmail compatibility; requires admin caller.
2126 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
2127 just ignore it); requires trusted caller.
2128 Also parse but ignore: -Ac -Am -X<logfile>
2131 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
2133 TL/02 Add +smtp_confirmation as a default logging option.
2135 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
2136 Patch by Magnus Holmgren from 2007-02-20.
2138 TL/04 Bugzilla 1281 - Spec typo.
2139 Bugzilla 1283 - Spec typo.
2140 Bugzilla 1290 - Spec grammar fixes.
2142 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
2144 TL/06 Add Experimental DMARC support using libopendmarc libraries.
2146 TL/07 Fix an out of order global option causing a segfault. Reported to dev
2147 mailing list by by Dmitry Isaikin.
2149 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
2151 JH/02 Support "G" suffix to numbers in ${if comparisons.
2153 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
2155 NM/01 Bugzilla 1197 - Spec typo
2156 Bugzilla 1196 - Spec examples corrections
2158 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
2160 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
2161 gnutls_enable_pkcs11, but renamed to more accurately indicate its
2164 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
2165 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
2167 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
2168 "acl {{name}{arg}...}", and optional args on acl condition
2171 JH/05 Permit multiple router/transport headers_add/remove lines.
2173 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
2175 JH/07 Avoid using a waiting database for a single-message-only transport.
2176 Performance patch from Paul Fisher. Bugzilla 1262.
2178 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
2181 JH/09 Add $headers_added variable, with content from use of ACL modifier
2182 add_header (but not yet added to the message). Bugzilla 199.
2184 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
2185 Pulled from Bugzilla 817 by Wolfgang Breyha.
2187 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
2189 (nb: this is the same fix as in Exim 4.80.1)
2191 JH/11 Add A= logging on delivery lines, and a client_set_id option on
2194 JH/12 Add optional authenticated_sender logging to A= and a log_selector
2197 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
2199 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
2200 advertise SMTP AUTH mechanism to us, instead of a generic
2201 protocol violation error. Also, make Exim more robust to bad
2202 data from the Dovecot auth socket.
2204 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
2206 When a queue runner is handling a message, Exim first routes the
2207 recipient addresses, during which it prunes them based on the retry
2208 hints database. After that it attempts to deliver the message to
2209 any remaining recipients. It then updates the hints database using
2212 So if a recipient address works intermittently, it can get repeatedly
2213 deferred at routing time. The retry hints record remains fresh so the
2214 address never reaches the final cutoff time.
2216 This is a fairly common occurrence when a user is bumping up against
2217 their storage quota. Exim had some logic in its local delivery code
2218 to deal with this. However it did not apply to per-recipient defers
2219 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
2221 This change adds a proper retry rule check during routing so that the
2222 final cutoff time is checked against the message's age. We only do
2223 this check if there is an address retry record and there is not a
2224 domain retry record; this implies that previous attempts to handle
2225 the address had the retry_use_local_parts option turned on. We use
2226 this as an approximation for the destination being like a local
2227 delivery, as in LMTP.
2229 I suspect this new check makes the old local delivery cutoff check
2230 redundant, but I have not verified this so I left the code in place.
2232 TF/02 Correct gecos expansion when From: is a prefix of the username.
2234 Test 0254 submits a message to Exim with the header
2238 When I ran the test suite under the user fanf2, Exim expanded
2239 the header to contain my full name, whereas it should have added
2240 a Resent-Sender: header. It erroneously treats any prefix of the
2241 username as equal to the username.
2243 This change corrects that bug.
2245 GF/01 DCC debug and logging tidyup
2246 Error conditions log to paniclog rather than rejectlog.
2247 Debug lines prefixed by "DCC: " to remove any ambiguity.
2249 TF/03 Avoid unnecessary rebuilds of lookup-related code.
2251 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
2252 Bug spotted by Jeremy Harris; was flawed since initial commit.
2253 Would have resulted in OCSP responses post-SNI triggering an Exim
2254 NULL dereference and crash.
2256 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
2258 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
2259 Bug detection, analysis and fix by Samuel Thibault.
2260 Bugzilla 1331, Debian bug #698092.
2262 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
2264 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
2265 Server implementation by Todd Lyons, client by JH.
2266 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
2267 config variable "prdr_enable" controls whether the server
2268 advertises the facility. If the client requests PRDR a new
2269 acl_data_smtp_prdr ACL is called once for each recipient, after
2270 the body content is received and before the acl_smtp_data ACL.
2271 The client is controlled by both of: a hosts_try_prdr option
2272 on the smtp transport, and the server advertisement.
2273 Default client logging of deliveries and rejections involving
2274 PRDR are flagged with the string "PRDR".
2276 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
2277 fclose(). Diagnosis by Todd Lyons.
2279 PP/17 Update configure.default to handle IPv6 localhost better.
2280 Patch by Alain Williams (plus minor tweaks).
2283 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
2284 This is now consistent with GnuTLS, and is now documented: the
2285 previous undocumented portable approach to treating the option as
2286 unset was to force an expansion failure. That still works, and
2287 an empty string is now equivalent.
2289 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
2290 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
2291 not performing validation itself.
2293 PP/20 Added force_command boolean option to pipe transport.
2294 Patch from Nick Koston, of cPanel Inc.
2296 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
2299 TF/04 Added udpsend ACL modifier and hexquote expansion operator
2301 PP/21 Fix eximon continuous updating with timestamped log-files.
2302 Broken in a format-string cleanup in 4.80, missed when I repaired the
2303 other false fix of the same issue.
2304 Report and fix from Heiko Schlichting.
2307 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
2308 Report from Prashanth Katuri.
2310 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
2311 It's SecureTransport, so affects any MacOS clients which use the
2312 system-integrated TLS libraries, including email clients.
2314 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
2315 using a MIME ACL for non-SMTP local injection.
2316 Report and assistance in diagnosis by Warren Baker.
2318 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
2320 JH/16 Fix comparisons for 64b. Bugzilla 1385.
2322 TL/09 Add expansion variable $authenticated_fail_id to keep track of
2323 last id that failed so it may be referenced in subsequent ACL's.
2325 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
2328 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
2329 ldap library initialization, allowing self-signed CA's to be
2330 used. Also properly sets require_cert option later in code by
2331 using NULL (global ldap config) instead of ldap handle (per
2332 session). Bug diagnosis and testing by alxgomz.
2334 TL/12 Enhanced documentation in the ratelimit.pl script provided in
2335 the src/util/ subdirectory.
2337 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
2338 renamed to Transport Post Delivery Action by Jeremy Harris, as
2341 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
2342 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
2343 redis_servers = needs to be configured which will be used by the redis
2344 lookup. Patch from Warren Baker, of The Packet Hub.
2346 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
2348 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
2349 hostname or reverse DNS when processing a host list. Used suggestions
2350 from multiple comments on this bug.
2352 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
2354 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
2355 Missed a few lines, added it to make the runtest require no keyboard
2358 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
2359 contains upper case chars. Make router use caseful_local_part.
2361 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
2362 support when GnuTLS has been built with p11-kit.
2368 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
2370 This, or similar/improved, will also be change PP/11 of 4.82.
2376 PP/01 Handle short writes when writing local log-files.
2377 In practice, only affects FreeBSD (8 onwards).
2378 Bugzilla 1053, with thanks to Dmitry Isaikin.
2380 NM/01 Bugzilla 949 - Documentation tweak
2382 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
2385 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
2387 PP/02 Implemented gsasl authenticator.
2389 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
2391 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
2392 `pkg-config foo` for cflags/libs.
2394 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
2395 with rest of GSASL and with heimdal_gssapi.
2397 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
2398 `pkg-config foo` for cflags/libs for the TLS implementation.
2400 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
2401 properties get this fed in as external SSF. A number of robustness
2402 and debugging improvements to the cyrus_sasl authenticator.
2404 PP/08 cyrus_sasl server now expands the server_realm option.
2406 PP/09 Bugzilla 1214 - Log authentication information in reject log.
2407 Patch by Jeremy Harris.
2409 PP/10 Added dbmjz lookup type.
2411 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
2413 PP/12 MAIL args handles TAB as well as SP, for better interop with
2414 non-compliant senders.
2415 Analysis and variant patch by Todd Lyons.
2417 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
2418 Bug report from Lars Müller <lars@samba.org> (via SUSE),
2419 Patch from Dirk Mueller <dmueller@suse.com>
2421 PP/13 tls_peerdn now print-escaped for spool files.
2422 Observed some $tls_peerdn in wild which contained \n, which resulted
2423 in spool file corruption.
2425 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
2426 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
2427 or write after TLS renegotiation, which otherwise led to messages
2430 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
2431 as a tracking header (ie: a signed header comes before the signature).
2432 Patch from Wolfgang Breyha.
2434 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
2435 comma-sep list; embedded commas doubled.
2437 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
2439 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
2441 Report and patch from Dmitry Banschikov.
2443 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
2444 Removed SSL_clear() after SSL_new() which led to protocol negotiation
2445 failures. We appear to now support TLS1.1+ with Exim.
2447 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
2448 lets Exim select keys and certificates based upon TLS SNI from client.
2449 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
2450 before an outbound SMTP session. New log_selector, +tls_sni.
2452 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
2453 NULL dereference. Report and patch from Alun Jones.
2455 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
2456 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
2457 Not seeing resolver debug output on NetBSD, but suspect this is a
2458 resolver implementation change.
2460 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
2461 Left warnings. Added "eximon gdb" invocation mode.
2463 PP/21 Defaulting "accept_8bitmime" to true, not false.
2465 PP/22 Added -bw for inetd wait mode support.
2467 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
2468 locate the relevant includes and libraries. Made this the default.
2470 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
2471 Bugzilla 1246, report and most of solution from Tomasz Kusy.
2473 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
2474 This may cause build issues on older platforms.
2476 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
2477 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
2478 gnutls_require_mac & gnutls_require_protocols (no longer supported).
2479 Added SNI support via GnuTLS too.
2480 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
2482 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
2484 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
2485 Applied second patch from Janne, implementing suggestion to default
2486 multiple-strings-in-record handling to match SPF spec.
2488 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
2490 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
2491 read-only, out of scope).
2492 Patch from Wolfgang Breyha, report from Stuart Northfield.
2494 PP/29 Fix three issues highlighted by clang analyser static analysis.
2495 Only crash-plausible issue would require the Cambridge-specific
2496 iplookup router and a misconfiguration.
2497 Report from Marcin Mirosław.
2499 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
2501 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
2502 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
2503 As part of this, removing so much warning spew let me fix some minor
2504 real issues in debug logging.
2506 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
2507 assignment on my part. Fixed.
2509 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
2510 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
2511 Janne Snabb (who went above and beyond: thank you).
2513 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
2514 string otherwise requires a connection and a bunch more work and it's
2515 relatively easy to get wrong. Should also expose TLS library linkage
2518 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
2519 64-bit ${eval} (JH/03).
2521 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
2522 GNU libc to support some of the 64-bit stuff, should not lead to
2523 conflicts. Defined before os.h is pulled in, so if a given platform
2524 needs to override this, it can.
2526 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
2527 protection layer was required, which is not implemented.
2528 Bugzilla 1254, patch from Wolfgang Breyha.
2530 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
2531 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
2532 tls_dhparam take prime identifiers. Also unbreak combination of
2533 OpenSSL+DH_params+TLSSNI.
2535 PP/39 Disable SSLv2 by default in OpenSSL support.
2541 PP/01 Solaris build fix for Oracle's LDAP libraries.
2542 Bugzilla 1109, patch from Stephen Usher.
2544 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
2546 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
2549 TF/02 Fix a couple more cases where we did not log the error message
2550 when unlink() failed. See also change 4.74-TF/03.
2552 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
2553 lock up or crash if it happened to be inside a call to libc when it
2554 got a SIGUSR1 from exiwhat.
2556 The SIGUSR1 handler appends the current process status to the process
2557 log which is later printed by exiwhat. It used to use the general
2558 purpose logging code to do this, but several functions it calls are
2559 not safe for signals.
2561 The new output code in the SIGUSR1 handler is specific to the process
2562 log, and simple enough that it's easy to inspect for signal safety.
2563 Removing some special cases also simplifies the general logging code.
2564 Removing the spurious timestamps from the process log simplifies
2567 TF/04 Improved ratelimit ACL condition.
2569 The /noupdate option has been deprecated in favour of /readonly which
2570 has clearer semantics. The /leaky, /strict, and /readonly update modes
2571 are mutually exclusive. The update mode is no longer included in the
2572 database key; it just determines when the database is updated. (This
2573 means that when you upgrade Exim will forget old rate measurements.)
2575 Exim now checks that the per_* options are used with an update mode that
2576 makes sense for the current ACL. For example, when Exim is processing a
2577 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
2578 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
2579 must specify per_mail/readonly. If you omit the update mode it defaults to
2580 /leaky where that makes sense (as before) or /readonly where required.
2582 The /noupdate option is now undocumented but still supported for
2583 backwards compatibility. It is equivalent to /readonly except that in
2584 ACLs where /readonly is required you may specify /leaky/noupdate or
2585 /strict/noupdate which are treated the same as /readonly.
2587 A useful new feature is the /count= option. This is a generalization
2588 of the per_byte option, so that you can measure the throughput of other
2589 aggregate values. For example, the per_byte option is now equivalent
2590 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
2592 The per_rcpt option has been generalized using the /count= mechanism
2593 (though it's more complicated than the per_byte equivalence). When it is
2594 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
2595 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
2596 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
2597 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
2598 Note that using per_rcpt with a non-readonly update mode in more than
2599 one ACL will cause the recipients to be double-counted. (The per_mail
2600 and per_byte options don't have this problem.)
2602 The handling of very low rates has changed slightly. If the computed rate
2603 is less than the event's count (usually one) then this event is the first
2604 after a long gap. In this case the rate is set to the same as this event's
2605 count, so that the first message of a spam run is counted properly.
2607 The major new feature is a mechanism for counting the rate of unique
2608 events. The new per_addr option counts the number of different
2609 recipients that someone has sent messages to in the last time period. It
2610 behaves like per_rcpt if all the recipient addresses are different, but
2611 duplicate recipient addresses do not increase the measured rate. Like
2612 the /count= option this is a general mechanism, so the per_addr option
2613 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
2614 example, measure the rate that a client uses different sender addresses
2615 with the options per_mail/unique=$sender_address. There are further
2616 details in the main documentation.
2618 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
2620 TF/06 Removed a few PCRE remnants.
2622 TF/07 Automatically extract Exim's version number from tags in the git
2623 repository when doing development or release builds.
2625 PP/02 Raise smtp_cmd_buffer_size to 16kB.
2626 Bugzilla 879. Patch from Paul Fisher.
2628 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
2629 Heavily based on revision 40f9a89a from Simon Arlott's tree.
2632 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
2634 PP/05 Variable $av_failed, true if the AV scanner deferred.
2635 Bugzilla 1078. Patch from John Horne.
2637 PP/06 Stop make process more reliably on build failure.
2638 Bugzilla 1087. Patch from Heiko Schlittermann.
2640 PP/07 Make maildir_use_size_file an _expandable_ boolean.
2641 Bugzilla 1089. Patch from Heiko Schlittermann.
2643 PP/08 Handle ${run} returning more data than OS pipe buffer size.
2644 Bugzilla 1131. Patch from Holger Weiß.
2646 PP/09 Handle IPv6 addresses with SPF.
2647 Bugzilla 860. Patch from Wolfgang Breyha.
2649 PP/10 GnuTLS: support TLS 1.2 & 1.1.
2651 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
2654 PP/11 match_* no longer expand right-hand-side by default.
2655 New compile-time build option, EXPAND_LISTMATCH_RHS.
2656 New expansion conditions, "inlist", "inlisti".
2658 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
2660 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
2662 PP/14 fix log_write() format string regression from TF/03.
2663 Bugzilla 1152. Patch from Dmitry Isaikin.
2669 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
2671 PP/02 Harmonised TLS library version reporting; only show if debugging.
2672 Layout now matches that introduced for other libraries in 4.74 PP/03.
2674 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
2676 PP/04 New "dns_use_edns0" global option.
2678 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
2681 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
2682 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
2684 TK/01 Updated PolarSSL code to 0.14.2.
2685 Bugzilla 1097. Patch from Andreas Metzler.
2687 PP/07 Catch divide-by-zero in ${eval:...}.
2688 Fixes bugzilla 1102.
2690 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
2693 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
2694 format-string attack -- SECURITY: remote arbitrary code execution.
2696 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
2697 time unintentionally subject to list matching rules, letting the header
2698 cause arbitrary Exim lookups (of items which can occur in lists, *not*
2699 arbitrary string expansion). This allowed for information disclosure.
2701 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
2702 INT_MIN/-1 -- value coerced to INT_MAX.
2708 NM/01 Workaround for PCRE version dependency in version reporting
2711 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
2712 This fixes portability to compilers other than gcc, notably
2713 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
2715 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
2716 makefiles for portability to HP-UX and POSIX correctness.
2718 PP/01 Permit LOOKUP_foo enabling on the make command-line.
2719 Also via indented variable definition in the Makefile.
2720 (Debugging by Oliver Heesakkers).
2722 PP/02 Restore caching of spamd results with expanded spamd_address.
2723 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
2725 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
2726 Improves build reliability. Fix from: Frank Elsner
2728 NM/02 Fix wide character breakage in the rfc2047 coding
2729 Fixes bug 1064. Patch from Andrey N. Oktyabrski
2731 NM/03 Allow underscore in dnslist lookups
2732 Fixes bug 1026. Patch from Graeme Fowler
2734 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
2735 Code patches from Adam Ciarcinski of NetBSD.
2737 NM/04 Fixed exiqgrep to cope with mailq missing size issue
2740 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
2741 is logged, to avoid truncation. Patch from John Horne.
2743 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
2744 Patch from Jakob Hirsch.
2746 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
2747 SQL string expansion failure details.
2748 Patch from Andrey Oktyabrski.
2750 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
2751 Patch from Simon Arlott.
2753 PP/09 New lookups functionality failed to compile on old gcc which rejects
2754 extern declarations in function scope.
2755 Patch from Oliver Fleischmann
2757 PP/10 Use sig_atomic_t for flags set from signal handlers.
2758 Check getgroups() return and improve debugging.
2759 Fixed developed for diagnosis in bug 927 (which turned out to be
2762 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
2763 Patch from Mark Zealey.
2765 PP/12 Bugzilla 1056: Improved spamd server selection.
2766 Patch from Mark Zealey.
2768 PP/13 Bugzilla 1086: Deal with maildir quota file races.
2769 Based on patch from Heiko Schlittermann.
2771 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
2772 Patch from Uwe Doering, sign-off by Michael Haardt.
2774 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
2775 variable declaration deep within a block. Bug and patch from
2778 PP/15 lookups-Makefile IRIX compatibility coercion.
2780 PP/16 Make DISABLE_DKIM build knob functional.
2782 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
2783 Patch by Simon Arlott
2785 TF/03 Fix valgrind.h portability to C89 compilers that do not support
2786 variable argument macros. Our copy now differs from upstream.
2792 TF/01 Failure to get a lock on a hints database can have serious
2793 consequences so log it to the panic log.
2795 TF/02 Log LMTP confirmation messages in the same way as SMTP,
2796 controlled using the smtp_confirmation log selector.
2798 TF/03 Include the error message when we fail to unlink a spool file.
2800 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
2801 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
2802 for maintaining out-of-tree patches for some time.
2804 PP/01 Bugzilla 139: Documentation and portability issues.
2805 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
2806 Handle per-OS dynamic-module compilation flags.
2808 PP/02 Let /dev/null have normal permissions.
2809 The 4.73 fixes were a little too stringent and complained about the
2810 permissions on /dev/null. Exempt it from some checks.
2811 Reported by Andreas M. Kirchwitz.
2813 PP/03 Report version information for many libraries, including
2814 Exim version information for dynamically loaded libraries. Created
2815 version.h, now support a version extension string for distributors
2816 who patch heavily. Dynamic module ABI change.
2818 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
2819 privilege escalation vulnerability whereby the Exim run-time user
2820 can cause root to append content of the attacker's choosing to
2823 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
2826 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
2827 If dropping privileges for untrusted macros, we disabled normal logging
2828 on the basis that it would fail; for the Exim run-time user, this is not
2829 the case, and it resulted in successful deliveries going unlogged.
2830 Fixed. Reported by Andreas Metzler.
2836 PP/01 Date: & Message-Id: revert to normally being appended to a message,
2837 only prepend for the Resent-* case. Fixes regression introduced in
2838 Exim 4.70 by NM/22 for Bugzilla 607.
2840 PP/02 Include check_rfc2047_length in configure.default because we're seeing
2841 increasing numbers of administrators be bitten by this.
2843 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
2845 PP/03 Bugzilla 994: added openssl_options main configuration option.
2847 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
2849 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
2851 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
2853 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
2854 without a peer certificate, leading to a segfault because of an
2855 assumption that peers always have certificates. Be a little more
2856 paranoid. Problem reported by Martin Tscholak.
2858 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
2859 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
2860 NB: ClamAV planning to remove STREAM in "middle of 2010".
2861 CL also introduces -bmalware, various -d+acl logging additions and
2862 more caution in buffer sizes.
2864 PP/09 Implemented reverse_ip expansion operator.
2866 PP/10 Bugzilla 937: provide a "debug" ACL control.
2868 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2870 PP/12 Bugzilla 973: Implement --version.
2872 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2874 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2876 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2878 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2879 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2880 ignore trailing whitespace.
2882 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2884 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2887 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2888 Notification from Dr Andrew Aitchison.
2890 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2891 ExtendedDetectionInfo response format.
2892 Notification from John Horne.
2894 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2897 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2898 XSL and documented dependency on system catalogs, with examples of how
2901 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2904 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2905 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2906 configuration file which is writeable by the Exim user or group.
2908 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2909 of configuration files to cover files specified with the -C option if
2910 they are going to be used with root privileges, not just the default
2913 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2914 option (effectively making it always true).
2916 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2917 files to be used while preserving root privileges.
2919 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2920 that rogue child processes cannot use them.
2922 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2923 run-time user, instead of root.
2925 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2926 Exim run-time user without dropping privileges.
2928 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2929 result string, instead of calling string_vformat() twice with the same
2932 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2933 for other users. Others should always drop root privileges if they use
2934 -C on the command line, even for a whitelisted configure file.
2936 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2938 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2944 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2945 $data_path, and $header_path variables; fixed documentation bugs and
2948 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2949 exipick to access non-standard spools, including the "frozen" queue
2952 NM/01 Bugzilla 965: Support mysql stored procedures.
2953 Patch from Alain Williams
2955 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2957 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2958 Patch from Andreas Metzler
2960 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2961 Patch from Kirill Miazine
2963 NM/05 Bugzilla 671: Added umask to procmail example.
2965 JJ/03 installed exipick 20100323.0, fixing doc bug
2967 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2968 directory. Notification and patch from Dan Rosenberg.
2970 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2972 TK/02 Improve log output when DKIM signing operation fails.
2974 MH/01 Treat the transport option dkim_domain as a colon separated
2975 list, not as a single string, and sign the message with each element,
2976 omitting multiple occurences of the same signer.
2978 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2979 Bugzilla 985, 986. Patch by Simon Arlott
2981 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2982 Patch by Simon Arlott
2984 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2985 MBX locking. Notification from Dan Rosenberg.
2991 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2993 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2995 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2997 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2999 NM/04 Bugzilla 915: Flag broken perl installation during build.
3005 TK/01 Added patch by Johannes Berg that expands the main option
3006 "spamd_address" if it starts with a dollar sign.
3008 TK/02 Write list of recipients to X-Envelope-Sender header when building
3009 the mbox-format spool file for content scanning (suggested by Jakob
3012 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
3013 (http://www.dcc-servers.net/) support via dccifd. Activated by
3014 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
3016 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
3017 by Mark Daniel Reidel <mr@df.eu>.
3019 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
3020 When building exim an external PCRE library is now needed -
3021 PCRE is a system library on the majority of modern systems.
3022 See entry on PCRE_LIBS in EDITME file.
3024 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
3025 conversation. Added nologin parameter to request.
3026 Patch contributed by Kirill Miazine.
3028 TF/01 Do not log submission mode rewrites if they do not change the address.
3030 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
3032 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
3033 log files in place. Contributed by Roberto Lima.
3035 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
3037 TF/03 Bugzilla 615: When checking the local_parts router precondition
3038 after a local_part_suffix or local_part_prefix option, Exim now
3039 does not use the address's named list lookup cache, since this
3040 contains cached lookups for the whole local part.
3042 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
3043 Robert Millan. Documentation is in experimental-spec.txt.
3045 TF/04 Bugzilla 668: Fix parallel build (make -j).
3047 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
3049 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
3050 Patch provided by Jan Srzednicki.
3052 TF/05 Leading white space used to be stripped from $spam_report which
3053 wrecked the formatting. Now it is preserved.
3055 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
3056 that they are available at delivery time.
3058 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
3060 TF/08 TLS error reporting now respects the incoming_interface and
3061 incoming_port log selectors.
3063 TF/09 Produce a more useful error message if an SMTP transport's hosts
3064 setting expands to an empty string.
3066 NM/06 Bugzilla 744: EXPN did not work under TLS.
3067 Patch provided by Phil Pennock.
3069 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
3070 Patch provided by Richard Godbee.
3072 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
3073 acl_smtp_notquit, added index entry.
3075 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
3076 Patch provided by Eugene Bujak.
3078 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
3079 accept(). Patch provided by Maxim Dounin.
3081 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
3082 Patch provided by Phil Pennock.
3084 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
3086 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
3087 Patch provided by Brad "anomie" Jorsch.
3089 NM/14 Bugzilla 622: Added timeout setting to transport filter.
3090 Patch provided by Dean Brooks.
3092 TK/05 Add native DKIM support (does not depend on external libraries).
3094 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
3095 Patch provided by Graeme Fowler.
3097 NM/16 Bugzilla 851: Documentation example syntax fix.
3099 NM/17 Changed NOTICE file to remove references to embedded PCRE.
3101 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
3104 NM/19 Bugzilla 745: TLS version reporting.
3105 Patch provided by Phil Pennock.
3107 NM/20 Bugzilla 167: bool: condition support.
3108 Patch provided by Phil Pennock.
3110 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
3111 clients. Patch provided by Phil Pennock.
3113 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
3114 Patch provided by Brad "anomie" Jorsch.
3116 NM/23 Bugzilla 687: Fix misparses in eximstats.
3117 Patch provided by Heiko Schlittermann.
3119 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
3120 Patch provided by Heiko Schlittermann.
3122 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
3123 plus update to original patch.
3125 NM/26 Bugzilla 799: Documentation correction for ratelimit.
3127 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
3128 Patch provided by David Brownlee.
3130 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
3132 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
3134 NM/30 Bugzilla 888: TLS documentation bugfixes.
3136 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
3138 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
3139 Unlike the original bugzilla I have changed all shell scripts in src tree.
3141 NM/33 Bugzilla 898: Transport filter timeout fix.
3142 Patch by Todd Rinaldo.
3144 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
3145 Patch by Serge Demonchaux.
3147 NM/35 Bugzilla 39: Base64 decode bug fixes.
3148 Patch by Jakob Hirsch.
3150 NM/36 Bugzilla 909: Correct connect() call in dcc code.
3152 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
3154 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
3156 NM/39 Bugzilla 911: Fixed MakeLinks build script.
3162 TK/01 Add preliminary DKIM support. Currently requires a forked version of
3163 ALT-N's libdkim that I have put here:
3164 http://duncanthrax.net/exim-experimental/
3166 Note to Michael Haardt: I had to rename some vars in sieve.c. They
3167 were called 'true' and it seems that C99 defines that as a reserved
3168 keyword to be used with 'bool' variable types. That means you could
3169 not include C99-style headers which use bools without triggering
3170 build errors in sieve.c.
3172 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
3173 as mailq or other aliases. Changed the --help handling significantly
3174 to do whats expected. exim_usage() emits usage/help information.
3176 SC/01 Added the -bylocaldomain option to eximstats.
3178 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
3180 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
3182 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
3188 PH/01 Another patch from the Sieve maintainer.
3190 PH/02 When an IPv6 address is converted to a string for single-key lookup
3191 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
3192 dots are used instead of colons so that keys in lsearch files need not
3193 contain colons. This was done some time before quoting was made available
3194 in lsearch files. However, iplsearch files do require colons in IPv6 keys
3195 (notated using the quote facility) so as to distinguish them from IPv4
3196 keys. This meant that lookups for IP addresses in host lists did not work
3197 for iplsearch lookups.
3199 This has been fixed by arranging for IPv6 addresses to be expressed with
3200 colons if the lookup type is iplsearch. This is not incompatible, because
3201 previously such lookups could never work.
3203 The situation is now rather anomalous, since one *can* have colons in
3204 ordinary lsearch keys. However, making the change in all cases is
3205 incompatible and would probably break a number of configurations.
3207 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
3210 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
3211 conversion specification without a maximum field width, thereby enabling
3212 a rogue spamd server to cause a buffer overflow. While nobody in their
3213 right mind would setup Exim to query an untrusted spamd server, an
3214 attacker that gains access to a server running spamd could potentially
3215 exploit this vulnerability to run arbitrary code as the Exim user.
3217 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
3218 $primary_hostname instead of what libspf2 thinks the hosts name is.
3220 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
3221 a directory entry by the name of the lookup key. Previously, if a
3222 symlink pointed to a non-existing file or a file in a directory that
3223 Exim lacked permissions to read, a lookup for a key matching that
3224 symlink would fail. Now it is enough that a matching directory entry
3225 exists, symlink or not. (Bugzilla 503.)
3227 PH/03 The body_linecount and body_zerocount variables are now exported in the
3230 PH/04 Added the $dnslist_matched variable.
3232 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
3233 This means they are set thereafter only if the connection becomes
3236 PH/06 Added the client_condition to authenticators so that some can be skipped
3237 by clients under certain conditions.
3239 PH/07 The error message for a badly-placed control=no_multiline_responses left
3240 "_responses" off the end of the name.
3242 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
3244 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
3245 (without spaces) instead of just copying the configuration text.
3247 PH/10 Added the /noupdate option to the ratelimit ACL condition.
3249 PH/11 Added $max_received_linelength.
3251 PH/12 Added +ignore_defer and +include_defer to host lists.
3253 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
3254 way in which PCRE > 7.0 is built.
3256 PH/14 Implemented queue_only_load_latch.
3258 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
3259 MAIL command. The effect was to mangle the value on 64-bit systems.
3261 PH/16 Another patch from the Sieve maintainer.
3263 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
3265 PH/18 If a system quota error occurred while trying to create the file for
3266 a maildir delivery, the message "Mailbox is full" was not appended to the
3267 bounce if the delivery eventually timed out. Change 4.67/27 below applied
3268 only to a quota excession during the actual writing of the file.
3270 PH/19 It seems that peer DN values may contain newlines (and other non-printing
3271 characters?) which causes problems in log lines. The DN values are now
3272 passed through string_printing() before being added to log lines.
3274 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
3275 and InterBase are left for another time.)
3277 PH/21 Added message_body_newlines option.
3279 PH/22 Guard against possible overflow in moan_check_errorcopy().
3281 PH/23 POSIX allows open() to be a macro; guard against that.
3283 PH/24 If the recipient of an error message contained an @ in the local part
3284 (suitably quoted, of course), incorrect values were put in $domain and
3285 $local_part during the evaluation of errors_copy.
3291 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
3292 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
3295 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
3296 issue a MAIL command.
3298 PH/02 In an ACL statement such as
3300 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
3302 if a client was not listed at all, or was listed with a value other than
3303 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
3304 the condition was not true (as it should be), so access was not denied.
3305 The bug was that the ! inversion was incorrectly passed on to the second
3306 item. This has been fixed.
3308 PH/03 Added additional dnslists conditions == and =& which are different from
3309 = and & when the dns lookup returns more than one IP address.
3311 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
3312 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
3314 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
3315 FSYNC, which compiles an option called disable_fsync that allows for
3316 bypassing fsync(). The documentation is heavily laced with warnings.
3318 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
3320 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
3321 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
3322 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
3323 including adding "make clean"; (3) Added -fPIC when compiling the test
3324 dynamically loaded module, to get rid of a warning.
3326 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
3327 message fails, move_frozen_messages = true and ignore_bounce_errors_after
3328 = 0s. The bug is otherwise harmless.
3330 PH/07 There was a bug in the dovecot authenticator such that the value of
3331 $auth1 could be overwritten, and so not correctly preserved, after a
3332 successful authentication. This usually meant that the value preserved by
3333 the server_setid option was incorrect.
3335 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
3337 PH/09 Installed PCRE release 7.0.
3339 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
3340 run for batched SMTP input. It is now run at the start of every message
3341 in the batch. While fixing this I discovered that the process information
3342 (output by running exiwhat) was not always getting set for -bs and -bS
3343 input. This is fixed, and it now also says "batched" for BSMTP.
3345 PH/11 Added control=no_pipelining.
3347 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
3348 patch, slightly modified), and move the expansion of helo_data till after
3349 the connection is made in the smtp transport (so it can use these
3352 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
3354 PH/14 Added log_selector = +pid.
3356 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
3358 PH/16 Add ${if forany and ${if forall.
3360 PH/17 Added dsn_from option to vary the From: line in DSNs.
3362 PH/18 Flush SMTP output before performing a callout, unless control =
3363 no_callout_flush is set.
3365 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
3366 was true (the default) a successful delivery failed to delete the retry
3367 item, thus causing premature timeout of the address. The bug is now
3370 PH/20 Added hosts_avoid_pipelining to the smtp transport.
3372 PH/21 Long custom messages for fakedefer and fakereject are now split up
3373 into multiline responses in the same way that messages for "deny" and
3374 other ACL rejections are.
3376 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
3377 with slight modification.
3379 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
3380 draft, changing the syntax and factoring some duplicate code".
3382 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
3383 for deliveries of the second and subsequent messages over the same SMTP
3386 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
3387 ${reduce, with only minor "tidies".
3389 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
3391 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
3392 expansion side effects.
3394 PH/27 When a message times out after an over-quota error from an Exim-imposed
3395 quota, the bounce message says "mailbox is full". This message was not
3396 being given when it was a system quota that was exceeded. It now should
3399 MH/03 Made $recipients available in local_scan(). local_scan() already has
3400 better access to the recipient list through recipients_list[], but
3401 $recipients can be useful in postmaster-provided expansion strings.
3403 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
3404 in the case of a MAIL command with additional options following the
3405 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
3406 were accidentally chopped off.
3408 PH/29 SMTP synchronization checks are implemented when a command is read -
3409 there is a check that no more input is waiting when there shouldn't be
3410 any. However, for some commands, a delay in an ACL can mean that it is
3411 some time before the response is written. In this time, more input might
3412 arrive, invalidly. So now there are extra checks after an ACL has run for
3413 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
3414 pipelining has not been advertised.
3416 PH/30 MH's patch to allow iscntrl() characters to be list separators.
3418 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
3419 returned in the SMTP response when smtp_return_error_details was false.
3420 This has been fixed.
3422 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
3423 instead of the C I/O that was originally supplied, because problems were
3424 reported on Solaris.
3426 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
3427 Exim which did not show up earlier: it was assuming that a call to
3428 SSL_CTX_set_info_callback() might give an error value. In fact, there is
3429 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
3430 was a macro that became an assignment, so it seemed to work. This has
3431 changed to a proper function call with a void return, hence the compile
3432 error. Exim's code has been fixed.
3434 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
3437 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
3439 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
3441 PH/37 If a message is not accepted after it has had an id assigned (e.g.
3442 because it turns out to be too big or there is a timeout) there is no
3443 "Completed" line in the log. When some messages of this type were
3444 selected by exigrep, they were listed as "not completed". Others were
3445 picked up by some special patterns. I have improved the selection
3446 criteria to be more general.
3448 PH/38 The host_find_failed option in the manualroute router can now be set
3449 to "ignore", to completely ignore a host whose IP address cannot be
3450 found. If all hosts are ignored, the behaviour is controlled by the new
3451 host_all_ignored option.
3453 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
3454 homing or because of multiple MX records with /mx) generated more than
3455 one IP address, and the following item turned out to be the local host,
3456 all the secondary addresses of the first item were incorrectly removed
3457 from the list, along with the local host and any following hosts (which
3458 is what is supposed to happen).
3460 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
3461 whoever called Exim into the -H file. In the case of the daemon it was
3462 behaving confusingly. When first started, it used values for whoever
3463 started the daemon, but after a SIGHUP it used the Exim user (because it
3464 calls itself on a restart). I have changed the code so that it now always
3467 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
3468 message are rejected with the same error (e.g. no authentication or bad
3469 sender address), and a DATA command is nevertheless sent (as can happen
3470 with PIPELINING or a stupid MUA), the error message that was given to the
3471 RCPT commands is included in the rejection of the DATA command. This is
3472 intended to be helpful for MUAs that show only the final error to their
3475 PH/42 Another patch from the Sieve maintainer.
3477 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
3478 Eximstats - Fixed some broken HTML links and added missing column headers
3480 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
3481 columns for Rejects, Temp Rejects, Ham, and Spam rows.
3483 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
3485 PH/43 Yet another patch from the Sieve maintainer.
3487 PH/44 I found a way to check for a TCP/IP connection going away before sending
3488 the response to the final '.' that terminates a message, but only in the
3489 case where the client has not sent further data following the '.'
3490 (unfortunately, this is allowed). However, in many cases there won't be
3491 any further data because there won't be any more messages to send. A call
3492 to select() can be used: if it shows that the input is "ready", there is
3493 either input waiting, or the socket has been closed. An attempt to read
3494 the next input character can distinguish the two cases. Previously, Exim
3495 would have sent an OK response which the client would never have see.
3496 This could lead to message repetition. This fix should cure that, at
3497 least in a lot of common cases.
3499 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
3500 advertised in response to EHLO.
3506 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
3507 fixed by 4.65/MH/01 (is this a record?) are fixed:
3509 (i) An empty string was always treated as zero by the numeric comparison
3510 operators. This behaviour has been restored.
3512 (ii) It is documented that the numeric comparison operators always treat
3513 their arguments as decimal numbers. This was broken in that numbers
3514 starting with 0 were being interpreted as octal.
3516 While fixing these problems I realized that there was another issue that
3517 hadn't been noticed. Values of message_size_limit (both the global option
3518 and the transport option) were treated as octal if they started with 0.
3519 The documentation was vague. These values are now always treated as
3520 decimal, and I will make that clear in the documentation.
3526 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
3527 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
3530 MH/01 Don't check that the operands of numeric comparison operators are
3531 integers when their expansion is in "skipping" mode (fixes bug
3532 introduced by 4.64-PH/07).
3534 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
3535 child addresses, Exim now panics and dies. Previously, because the count
3536 is held in a short int, deliveries were likely to be lost. As such a
3537 large number of recipients for a single message is ridiculous
3538 (performance will be very, very poor), I have chosen to impose a limit
3539 rather than extend the field.
3545 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
3546 leftover -K file (the existence of which was triggered by #402).
3547 While we were at it, introduced process PID as part of the -K
3548 filename. This should rule out race conditions when creating
3551 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
3552 processing considerably. Previous code took too long for large mails,
3553 triggering a timeout which in turn triggers #401.
3555 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
3556 in the DK code in transports.c. sendfile() is not really portable,
3557 hence the _LINUX specificness.
3559 TF/01 In the add_headers option to the mail command in an Exim filter,
3560 there was a bug that Exim would claim a syntax error in any
3561 header after the first one which had an odd number of characters
3564 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
3565 callout verification, Exim cached a "reject" for the entire domain. This
3566 is correct for most verifications, but it is not correct for a recipient
3567 verification with use_sender or use_postmaster set, because in that case
3568 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
3569 case of MAIL FROM:<> rejection from other early rejections (e.g.
3570 rejection of HELO). When verifying a recipient using a non-null MAIL
3571 address, the cache is ignored if it shows MAIL FROM:<> rejection.
3572 Whatever the result of the callout, the value of the domain cache is
3573 left unchanged (for any other kind of callout, getting as far as trying
3574 RCPT means that the domain itself is ok).
3576 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
3579 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
3580 manifest itself as EPIPE rather than ECONNECT. When tidying away a
3581 session, the daemon ignores ECONNECT errors and logs others; it now
3582 ignores EPIPE as well.
3584 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
3585 (quoted-printable decoding).
3587 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
3588 later the small subsequent patch to fix an introduced bug.
3590 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
3592 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
3594 PH/08 An error is now given if message_size_limit is specified negative.
3596 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
3597 to be given (somewhat) arbitrary names.
3599 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
3602 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
3603 miscellaneous code fixes
3605 PH/10 Added the log_reject_target ACL modifier to specify where to log
3608 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
3609 hostname. This is wrong, because it relates to the incoming message (and
3610 probably the interface on which it is arriving) and not to the outgoing
3611 callout (which could be using a different interface). This has been
3612 changed to use the value of the helo_data option from the smtp transport
3613 instead - this is what is used when a message is actually being sent. If
3614 there is no remote transport (possible with a router that sets up host
3615 addresses), $smtp_active_hostname is used.
3617 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
3618 tweaks were necessary in order to get it to work (see also 21 below):
3619 (a) The code assumed that strncpy() returns a negative number on buffer
3620 overflow, which isn't the case. Replaced with Exim's string_format()
3622 (b) There were several signed/unsigned issues. I just did the minimum
3623 hacking in of casts. There is scope for a larger refactoring.
3624 (c) The code used strcasecmp() which is not a standard C function.
3625 Replaced with Exim's strcmpic() function.
3626 (d) The code set only $1; it now sets $auth1 as well.
3627 (e) A simple test gave the error "authentication client didn't specify
3628 service in request". It would seem that Dovecot has changed its
3629 interface. Fortunately there's a specification; I followed it and
3630 changed what the client sends and it appears to be working now.
3632 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
3635 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
3636 address A is aliased to B and C, where B exists and C does not. Without
3637 -v the output is "A verified" because verification stops after a
3638 successful redirection if more than one address is generated. However,
3639 with -v the child addresses are also verified. Exim was outputting "A
3640 failed to verify" and then showing the successful verification for C,
3641 with its parentage. It now outputs "B failed to verify", showing B's
3642 parentage before showing the successful verification of C.
3644 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
3645 look up a TXT record in a specific list after matching in a combined
3648 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
3649 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
3650 they consult the DNS. I had assumed they would set it the way they
3651 wanted; and indeed my experiments on Linux seem to show that in some
3652 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
3653 To be on the safe side, however, I have now made the interface to
3654 host_find_byname() similar to host_find_bydns(), with an argument
3655 containing the DNS resolver options. The host_find_byname() function now
3656 sets these options at its start, just as host_find_bydns() does. The smtp
3657 transport options dns_qualify_single and dns_search_parents are passed to
3658 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
3659 of host_find_byname() use the default settings of RES_DEFNAMES
3660 (qualify_single) but not RES_DNSRCH (search_parents).
3662 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
3663 spool_read_header() do less string testing, by means of a preliminary
3664 switch on the second character of optional "-foo" lines. (This is
3665 overdue, caused by the large number of possibilities that now exist.
3666 Originally there were few.) While I was there, I also converted the
3667 str(n)cmp tests so they don't re-test the leading "-" and the first
3668 character, in the hope this might squeeze out yet more improvement.
3670 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
3671 flag allowing group syntax was set by the header_syntax check but not
3672 turned off, possible causing trouble later; (2) The flag was not being
3673 set at all for the header_verify test, causing "group"-style headers to
3674 be rejected. I have now set it in this case, and also caused header_
3675 verify to ignore an empty address taken from a group. While doing this, I
3676 came across some other cases where the code for allowing group syntax
3677 while scanning a header line wasn't quite right (mostly, not resetting
3678 the flag correctly in the right place). These bugs could have caused
3679 trouble for malformed header lines. I hope it is now all correct.
3681 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
3682 with the "reply" argument non-NULL. The code, however (which originally
3683 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
3684 but it didn't always do it. This confused somebody who was copying the
3685 code for some other use. I have removed all the tests.
3687 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
3688 feature that was used to support insecure browsers during the U.S. crypto
3689 embargo. It requires special client support, and Exim is probably the
3690 only MTA that supported it -- and would never use it because real RSA is
3691 always available. This code has been removed, because it had the bad
3692 effect of slowing Exim down by computing (never used) parameters for the
3693 RSA_EXPORT functionality.
3695 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
3696 authenticator to fail if there's a tab character in the incoming data
3697 (there should never be unless someone is messing about, as it's supposed
3698 to be base64-encoded). Also added, on Timo's advice, the "secured" option
3699 if the connection is using TLS or if the remote IP is the same as the
3700 local IP, and the "valid-client-cert option" if a client certificate has
3703 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
3704 authenticators. This can be used for authorization after authentication
3705 succeeds. (In the case of plaintext, it servers for both authentication
3708 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
3709 if any retry times were supplied.
3711 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
3712 connection, where there is no client IP address to check. In this
3713 situation, the verify now always succeeds.
3715 PH/25 Applied John Jetmore's -Mset patch.
3717 PH/26 Added -bem to be like -Mset, but loading a message from a file.
3719 PH/27 In a string expansion for a processed (not raw) header when multiple
3720 headers of the same name were present, leading whitespace was being
3721 removed from all of them, but trailing whitespace was being removed only
3722 from the last one. Now trailing whitespace is removed from each header
3723 before concatenation. Completely empty headers in a concatenation (as
3724 before) are ignored.
3726 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
3727 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
3729 PH/29 [Removed. This was a change that I later backed out, and forgot to
3730 correct the ChangeLog entry (that I had efficiently created) before
3731 committing the later change.]
3733 PH/30 Exim was sometimes attempting to deliver messages that had suffered
3734 address errors (4xx response to RCPT) over the same connection as other
3735 messages routed to the same hosts. Such deliveries are always "forced",
3736 so retry times are not inspected. This resulted in far too many retries
3737 for the affected addresses. The effect occurred only when there were more
3738 hosts than the hosts_max_try setting in the smtp transport when it had
3739 the 4xx errors. Those hosts that it had tried were not added to the list
3740 of hosts for which the message was waiting, so if all were tried, there
3741 was no problem. Two fixes have been applied:
3743 (i) If there are any address or message errors in an SMTP delivery, none
3744 of the hosts (tried or untried) are now added to the list of hosts
3745 for which the message is waiting, so the message should not be a
3746 candidate for sending over the same connection that was used for a
3747 successful delivery of some other message. This seems entirely
3748 reasonable: after all the message is NOT "waiting for some host".
3749 This is so "obvious" that I'm not sure why it wasn't done
3750 previously. Hope I haven't missed anything, but it can't do any
3751 harm, as the worst effect is to miss an optimization.
3753 (ii) If, despite (i), such a delivery is accidentally attempted, the
3754 routing retry time is respected, so at least it doesn't keep
3755 hammering the server.
3757 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
3758 in ${readsocket because some servers need this prod.
3760 PH/32 Added some extra debug output when updating a wait-xxx database.
3762 PH/33 The hint "could be header name not terminated by colon", which has been
3763 given for certain expansion errors for a long time, was not being given
3764 for the ${if def:h_colon_omitted{... case.
3766 PH/34 The spec says: "With one important exception, whenever a domain list is
3767 being scanned, $domain contains the subject domain." There was at least
3768 one case where this was not true.
3770 PH/35 The error "getsockname() failed: connection reset by peer" was being
3771 written to the panic log as well as the main log, but it isn't really
3772 panic-worthy as it just means the connection died rather early on. I have
3773 removed the panic log writing for the ECONNRESET error when getsockname()
3776 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
3777 runs only) independently of the message's sender address. This meant
3778 that, if the 4xx error was in fact related to the sender, a different
3779 message to the same recipient with a different sender could confuse
3780 things. In particular, this can happen when sending to a greylisting
3781 server, but other circumstances could also provoke similar problems.
3782 I have changed the default so that the retry time for these errors is now
3783 based a combination of the sender and recipient addresses. This change
3784 can be overridden by setting address_retry_include_sender=false in the
3787 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
3788 remote server are returned as part of bounce messages. This was not
3789 happening for LMTP over a pipe (the lmtp transport), but now it is the
3790 same for both kinds of LMTP.
3792 PH/38 Despite being documented as not happening, Exim was rewriting addresses
3793 in header lines that were in fact CNAMEs. This is no longer the case.
3795 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
3796 and queue runs started by the daemon processed all messages. This has
3797 been fixed so that -R and -S can now usefully be given with -q<time>.
3799 PH/40 Import PCRE release 6.7 (fixes some bugs).
3801 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
3803 PH/42 Give an error if -q is specified more than once.
3805 PH/43 Renamed the variables $interface_address and $interface_port as
3806 $received_ip_address and $received_port, to make it clear that these
3807 values apply to message reception, and not to the outgoing interface when
3808 a message is delivered. (The old names remain recognized, of course.)
3810 PH/44 There was no timeout on the connect() call when using a Unix domain
3811 socket in the ${readsocket expansion. There now is.
3813 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
3814 be meaningful with "accept".
3816 SC/01 Eximstats V1.43
3817 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
3819 SC/02 Eximstats V1.44
3820 Use a glob alias rather than an array ref in the generated
3821 parser. This improves both readability and performance.
3823 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
3824 Collect SpamAssassin and rejection statistics.
3825 Don't display local sender or destination tables unless
3826 there is data to show.
3827 Added average volumes into the top table text output.
3829 SC/04 Eximstats V1.46
3830 Collect data on the number of addresses (recipients)
3831 as well as the number of messages.
3833 SC/05 Eximstats V1.47
3834 Added 'Message too big' to the list of mail rejection
3835 reasons (thanks to Marco Gaiarin).
3837 SC/06 Eximstats V1.48
3838 Mainlog lines which have GMT offsets and are too short to
3839 have a flag are now skipped.
3841 SC/07 Eximstats V1.49 (Alain Williams)
3842 Added the -emptyok flag.
3844 SC/08 Eximstats V1.50
3845 Fixes for obtaining the IP address from reject messages.
3847 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
3848 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
3849 whitespace changes from 4.64-PH/27
3851 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
3854 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
3855 are found, allow negative numbers in numeric criteria)
3857 JJ/06 exipick.20061117.2, added new $message_body_missing variable
3859 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
3860 to match changes made in 4.64-PH/43
3862 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
3864 PH/47 Put in an explicit test for a DNS lookup of an address record where the
3865 "domain" is actually an IP address, and force a failure. This locks out
3866 those revolvers/nameservers that support "A-for-A" lookups, in
3867 contravention of the specifications.
3869 PH/48 When a host name was looked up from an IP address, and the subsequent
3870 forward lookup of the name timed out, the host name was left in
3871 $sender_host_name, contrary to the specification.
3873 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3874 restricted to single-key lookups, Exim was not diagnosing an error if
3875 * or *@ was used with a query-style lookup.
3877 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3879 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3880 long ago, but noone interested enough thought of it. Let's just say that
3881 the "1.1" means that there are some new functions that weren't there at
3882 some point in the past.
3884 PH/51 Error processing for expansion failure of helo_data from an smtp
3885 transport during callout processing was broken.
3887 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3888 tested/used via the -bh/-bhc/-bs options.
3890 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3891 bug, fixed in subsequent PCRE releases).
3893 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3894 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3896 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3902 SC/01 Use a glob alias rather than an array ref in eximstats generated
3903 parser. This improves both readability and performance.
3905 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3906 Don't display local sender or destination tables in eximstats unless
3907 there is data to show.
3908 Added average volumes into the eximstats top table text output.
3910 SC/03 Collect data on the number of addresses (recipients) as well
3911 as the number of messages in eximstats.
3913 TF/01 Correct an error in the documentation for the redirect router. Exim
3914 does (usually) call initgroups() when daemonizing.
3916 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3917 with consistent privilege compared to when running as a daemon.
3919 TF/03 Note in the spec that $authenticated_id is not set for local
3920 submissions from trusted users.
3922 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3923 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3925 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3926 by adding some example configuration directives to the default
3927 configuration file. A little bit of work is required to uncomment the
3928 directives and define how usernames and passwords are checked, but
3929 there is now a framework to start from.
3931 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3932 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3933 without this. I don't know how relevant this is to other LDAP libraries.
3935 PH/02 Add the verb name to the "unknown ACL verb" error.
3937 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3939 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3941 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3942 directory not expanded when it should be if an expanded home directory
3943 was set for the address (which is overridden by the transport).
3945 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3948 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3949 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3950 because it is too late at that time, and has no effect.
3952 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3953 security issue with \' (bugzilla #107). I could not use the
3954 PQescapeStringConn() function, because it needs a PGconn value as one of
3957 PH/08 When testing addresses using -bt, indicate those final addresses that
3958 are duplicates that would not cause an additional delivery. At least one
3959 person was confused, thinking that -bt output corresponded to deliveries.
3960 (Suppressing duplicates isn't a good idea as you lose the information
3961 about possibly different redirections that led to the duplicates.)
3963 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3964 systems where poll() doesn't work, in particular OS X.
3966 PH/10 Added more information to debugging output for retry time not reached.
3968 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3969 operations in malware.c.
3971 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3974 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3975 syslog_duplication set false, log lines that would normally be written
3976 both the the main log and to the reject log were not written to syslog at
3979 PH/14 In the default configuration, change the use of "message" in ACL warn
3980 statements to "add_header".
3982 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3983 not followed by a command (e.g. "seen endif").
3985 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3986 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3989 PH/17 Added extra conditions to the default value of delay_warning_condition
3993 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3994 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3995 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3998 The Auto-Submitted: and various List- headers are standardised, whereas I
3999 don't think Precedence: ever was.
4001 PH/18 Refactored debugging code in route_finduser() to show more information,
4002 in particular, the error code if getpwnam() issues one.
4004 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
4005 This is apparently needed in addition to the PH/07 change above to avoid
4006 any possible encoding problems.
4008 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
4009 but not after initializing Perl.
4011 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
4012 output them only if debugging. By default they are written stderr,
4013 apparently, which is not desirable.
4015 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
4018 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
4021 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
4023 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
4024 authenticated or an ident call has been made. Suppress the default
4025 values for $authenticated_id and $authenticated_sender (but permit -oMai
4026 and -oMas) when testing with -bh.
4028 PH/24 Re-jigged the order of the tests in the default configuration so that the
4029 tests for valid domains and recipients precede the DNS black list and CSA
4030 tests, on the grounds that those ones are more expensive.
4032 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
4033 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
4034 This bug exists in every version of Exim that I still have, right back to
4037 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
4038 However, an attempt to turn on case-sensitivity in a regex key by
4039 including (?-i) didn't work because the subject string was already
4040 lowercased, and the effects were non-intuitive. It turns out that a
4041 one-line patch can be used to allow (?-i) to work as expected.
4047 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
4048 other effects) broke the use of negated acl sub-conditions.
4050 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
4053 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
4054 "Deny" causes Exim to reject the incoming connection with a 554 error.
4055 Unfortunately, if there is a major crisis, such as a disk failure,
4056 tcp-wrappers gives "deny", whereas what one would like would be some
4057 kind of temporary error. A kludge has been added to help with this.
4058 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
4059 554 error is used if errno is still zero or contains ENOENT (which occurs
4060 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
4063 PH/03 Add -lutil to the default FreeBSD LIBS setting.
4065 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
4066 errors. Otherwise a message that provokes a temporary error (when other
4067 messages do not) can cause a whole host to time out.
4069 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
4070 addresses were routed directly to files or pipes from a redirect router.
4071 File deliveries just didn't batch; pipe deliveries might have suffered
4074 PH/06 A failure to get a lock for a hints database would erroneously always say
4075 "Failed to get write lock", even when it was really a read lock.
4077 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
4078 of 0600. This has been changed to use the value of the lockfile_mode
4079 option (which defaults to 0600).
4081 PH/08 Applied small patch from the Sieve maintainer.
4083 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
4084 folder from quota calculations, a direct delivery into this folder messed
4085 up the contents of the maildirsize file. This was because the regex was
4086 used only to exclude .Trash (or whatever) when the size of the mailbox
4087 was calculated. There was no check that a delivery was happening into an
4088 excluded directory. This bug has been fixed by ignoring all quota
4089 processing for deliveries into excluded directories.
4091 PH/10 Added the maildirfolder_create_regex option to appendfile.
4097 PH/01 The code for finding all the local interface addresses on a FreeBSD
4098 system running IPv6 was broken. This may well have applied to all BSD
4099 systems, as well as to others that have similar system calls. The broken
4100 code found IPv4 interfaces correctly, but gave incorrect values for the
4101 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
4102 that it would not match correctly against @[] and not recognize the IPv6
4105 PH/02 The ipliteral router was not recognizing addresses of the form user@
4106 [ipv6:....] because it didn't know about the "ipv6:" prefix.
4108 PH/03 Added disable_ipv6.
4110 PH/04 Changed $reply_address to use the raw form of the headers instead of the
4111 decoded form, because it is most often used to construct To: headers
4112 lines in autoreplies, and the decoded form may well be syntactically
4113 invalid. However, $reply_address has leading white space removed, and all
4114 newlines turned into spaces so that the autoreply transport does not
4117 PH/05 If group was specified without a user on a router, and no group or user
4118 was specified on a transport, the group from the router was ignored.
4120 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
4121 for visible compile-time settings that can be used to change these
4122 numbers, for those that want even more. Backwards compatibility with old
4123 spool files has been maintained. However, going back to a previous Exim
4124 release will lost any variables that are in spool files.
4126 PH/07 Two small changes when running in the test harness: increase delay when
4127 passing a TCP/IP connection to a new process, in case the original
4128 process has to generate a bounce, and remove special handling of
4129 127.0.0.2 (sic), which is no longer necessary.
4131 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
4132 be the same on different OS.
4134 PH/09 Moved a debug statement in filter processing to avoid a race problem when
4137 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
4138 whether --show-vars was specified or not
4140 JJ/02 exipick: Added support for new ACL variable spool format introduced
4143 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
4144 syntactically invalid From: or Reply-to: line, and a filter used this to
4145 generate an autoreply, and therefore failed to obtain an address for the
4146 autoreply, Exim could try to deliver to a non-existent relative file
4147 name, causing unrelated and misleading errors. What now happens is that
4148 it logs this as a hard delivery error, but does not attempt to create a
4151 PH/11 The exinext utility has a -C option for testing purposes, but although
4152 the given file was scanned by exinext itself; it wasn't being passed on
4153 when Exim was called.
4155 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
4156 an end-of-file indication when reading a command response.
4158 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
4159 compiled. In many other places in Exim, IPv6 addresses are always
4160 recognized, so I have changed this. It also means that IPv4 domain
4161 literals of the form [IPV4:n.n.n.n] are now always recognized.
4163 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
4164 used if the router is not running as root, for example, when verifying at
4165 ACL time, or when using -bh. The debugging output from this situation was
4166 non-existent - all you got was a failure to exec. I have made two
4169 (a) Failures to set uid/gid, the current directory, or a process leader
4170 in a subprocess such as that created by queryprogram now generate
4171 suitable debugging output when -d is set.
4173 (b) The queryprogram router detects when it is not running as root,
4174 outputs suitable debugging information if -d is set, and then runs
4175 the subprocess without attempting to change uid/gid.
4177 PH/15 Minor change to Makefile for building test_host (undocumented testing
4180 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
4181 additional section of a DNS packet that returns MX or SRV records.
4182 Instead, it always explicitly searches for A/AAAA records. This avoids
4183 major problems that occur when a DNS server includes only records of one
4184 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
4185 fixed another bug: if SRV records were looked up and the corresponding
4186 address records were *not* found in the additional section, the port
4187 values from the SRV records were lost.
4189 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
4190 using the correct key (the original address) when searching the retry
4191 rules in order to find which one to use for generating the retry hint.
4193 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
4194 adding the default one. Similarly, if it contains a Reply-To: header, the
4195 errors_reply_to option, if set, is not used.
4197 PH/19 When calculating a retry time, Exim used to measure the "time since
4198 failure" by looking at the "first failed" field in the retry record. Now
4199 it does not use this if it is later than than the arrival time of the
4200 message. Instead it uses the arrival time. This makes for better
4201 behaviour in cases where some deliveries succeed, thus re-setting the
4202 "first failed" field. An example is a quota failure for a huge message
4203 when small messages continue to be delivered. Without this change, the
4204 "time since failure" will always be short, possible causing more frequent
4205 delivery attempts for the huge message than are intended.
4206 [Note: This change was subsequently modified - see PH/04 for 4.62.]
4208 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
4209 $1, $2, $3) because the numerical variables can be reset during some
4210 expansion items (e.g. "match"), thereby losing the authentication data.
4212 PH/21 Make -bV show the size of off_t variables so that the test suite can
4213 decide whether to run tests for quotas > 2G.
4215 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
4216 mailbox_size, and mailbox_filecount in the appendfile transport. If a
4217 filecount value is greater than 2G or if a quota value is greater than 2G
4218 on a system where the size of off_t is not greater than 4, a panic error
4221 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
4222 never match. The debug and -bh output now contains an explicit error
4223 message indicating a malformed IPv4 address or mask.
4225 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
4226 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
4227 PH/23 above applies.
4229 PH/25 Do not write to syslog when running in the test harness. The only
4230 occasion when this arises is a failure to open the main or panic logs
4231 (for which there is an explicit test).
4233 PH/26 Added the /no_tell option to "control=freeze".
4235 PH/27 If a host name lookup failed very early in a connection, for example, if
4236 the IP address matched host_lookup and the reverse lookup yielded a name
4237 that did not have a forward lookup, an error message of the form "no IP
4238 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
4239 could be logged. Now it outputs the IP address instead of "NULL".
4241 PH/28 An enabling patch from MH: add new function child_open_exim2() which
4242 allows the sender and the authenticated sender to be set when
4243 submitting a message from within Exim. Since child_open_exim() is
4244 documented for local_scan(), the new function should be too.
4246 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
4247 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
4248 results in an empty string is now treated as unset.
4250 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
4252 PH/31 Added sender_verify_fail logging option.
4254 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
4255 needed by RFC 822 but not by RFC 2822 was commented out. I have now
4256 tidied the source and removed it altogether.
4258 PH/33 When a queue run was abandoned because the load average was too high, a
4259 log line was always written; now it is written only if the queue_run log
4260 selector is set. In addition, the log line for abandonment now contains
4261 information about the queue run such as the pid. This is always present
4262 in "start" and "stop" lines but was omitted from the "abandon" line.
4264 PH/34 Omit spaces between a header name and the colon in the error message that
4265 is given when verify = headers_syntax fails (if there are lots of them,
4266 the message gets confusing).
4268 PH/35 Change the default for dns_check_names_pattern to allow slashes within
4269 names, as there are now some PTR records that contain slashes. This check
4270 is only to protect against broken name servers that fall over on strange
4271 characters, so the fact that it applies to all lookups doesn't matter.
4273 PH/36 Now that the new test suite is complete, we can remove some of the
4274 special code in Exim that was needed for the old test suite. For example,
4275 sorting DNS records because real resolvers return them in an arbitrary
4276 order. The new test suite's fake resolver always returns records in the
4279 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
4280 bounces) except when queue_only is set, to avoid logging races between
4281 the different processes.
4283 PH/38 Panic-die if .include specifies a non-absolute path.
4285 PH/39 A tweak to the "H" retry rule from its user.
4287 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
4288 a label. They prevented compilation on older perls.
4290 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
4291 a warning to be raised on newish perls.
4293 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
4294 on queue. Changes to match documented behaviour of showing count of
4295 messages matching specified criteria.
4297 PH/40 Changed the default ident timeout from 30s to 5s.
4299 PH/41 Added support for the use of login_cap features, on those BSD systems
4300 that have them, for controlling the resources used by pipe deliveries.
4302 PH/42 The content-scanning code uses fopen() to create files in which to put
4303 message data. Previously it was not paying any attention to the mode of
4304 the files. Exim runs with umask(0) because the rest of the code creates
4305 files with open(), and sets the required mode explicitly. Thus, these
4306 files were ending up world-writeable. This was not a big issue, because,
4307 being within the spool directory, they were not world-accessible. I have
4308 created a function called modefopen, which takes an additional mode
4309 argument. It sets umask(777), creates the file, chmods it to the required
4310 mode, then resets the umask. All the relevant calls to fopen() in the
4311 content scanning code have been changed to use this function.
4313 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
4314 to 24 hours. This avoids potential overflow problems when processing G
4315 and H retry rules. I suspect nobody ever tinkers with this value.
4317 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
4319 PH/45 When the plaintext authenticator is running as a client, the server's
4320 challenges are checked to ensure they are valid base64 strings. By
4321 default, the authentication attempt is cancelled if an invalid string is
4322 received. Setting client_ignore_invalid_base64 true ignores these errors.
4323 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
4324 they are received. Thus, the responses can be made to depend on the
4325 challenges. If an invalid string is ignored, an empty string is placed in
4328 PH/46 Messages that are created by the autoreply transport now contains a
4329 References: header, in accordance with RFCs 2822 and 3834.
4331 PH/47 Added authenticated_sender_force to the smtp transport.
4333 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
4335 PH/49 Installed latest patch from the Sieve maintainer.
4337 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
4338 was also set, the appendfile transport was unnecessarily scanning a
4339 directory of message files (e.g. for maildir delivery) to find the count
4340 of files (along with the size), even though it did not need this
4341 information. It now does the scan only if it needs to find either the
4342 size of the count of files.
4344 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
4346 PH/52 Two bugs concerned with error handling when the smtp transport is
4349 (i) Exim was not creating retry information for temporary errors given
4350 for individual recipients after the DATA command when the smtp transport
4351 was used in LMTP mode. This meant that they could be retried too
4352 frequently, and not timed out correctly.
4354 (ii) Exim was setting the flag that allows error details to be returned
4355 for LMTP errors on RCPT commands, but not for LMTP errors for individual
4356 recipients that were returned after the DATA command.
4358 PH/53 This is related to PH/52, but is more general: for any failing address,
4359 when detailed error information was permitted to be returned to the
4360 sender, but the error was temporary, then after the final timeout, only
4361 "retry timeout exceeded" was returned. Now it returns the full error as
4362 well as "retry timeout exceeded".
4364 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
4365 do this, and (what is worse) MTAs that accept it.
4367 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
4368 will now be deprecated.
4370 PH/56 New os.c-cygwin from the Cygwin maintainer.
4372 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
4373 formats (previously only available in exim formats via -bpr, -bpru,
4374 and -bpra. Now also available in native and exiqgrep formats)
4376 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
4377 with very large, slow to parse queues
4379 JJ/08 exipick: added ! as generic prefix to negate any criteria format
4381 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
4383 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
4384 responses to authentication challenges, though it was showing the
4385 challenges; (ii) I've removed the CR characters from the debug output for
4388 PH/58 Allow for the insertion of a newline as well as a space when a string
4389 is turned into more than one encoded-word during RFC 2047 encoding. The
4390 Sieve code now uses this.
4392 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
4393 data_4xx, lost_connection, tls_required.
4395 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
4396 message was being sent as an SMTP response.
4398 PH/61 Add -l and -k options to exicyclog.
4400 PH/62 When verifying, if an address was redirected to one new address, so that
4401 verification continued, and the new address failed or deferred after
4402 having set something in $address_data, the value of $address_data was not
4403 passed back to the ACL. This was different to the case when no
4404 redirection occurred. The value is now passed back in both cases.
4406 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
4407 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
4408 use login_cap.h, so on its own it isn't the distinguishing feature. The
4409 new name refers directly to the setclassresources() function.
4411 PH/65 Added configuration files for NetBSD3.
4413 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
4415 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
4416 is preferred over IPv4.
4418 PH/68 The bounce_return_message and bounce_return_body options were not being
4419 honoured for bounces generated during the reception of non-SMTP messages.
4420 In particular, this applied to messages rejected by the ACL. This bug has
4421 been fixed. However, if bounce_return_message is true and bounce_return_
4422 body is false, the headers that are returned for a non-SMTP message
4423 include only those that have been read before the error was detected.
4424 (In the case of an ACL rejection, they have all been read.)
4426 PH/69 The HTML version of the specification is now built in a directory called
4427 spec_html instead of spec.html, because the latter looks like a path with
4428 a MIME-type, and this confuses some software.
4430 PH/70 Catch two compiler warnings in sieve.c.
4432 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
4433 function verify_get_ident() calls ip_connect() to connect a socket, but
4434 if the "connect()" function timed out, ip_connect() used to close the
4435 socket. However, verify_get_ident() also closes the socket later, and in
4436 between Exim writes to the log, which may get opened at this point. When
4437 the socket was closed in ip_connect(), the log could get the same file
4438 descriptor number as the socket. This naturally causes chaos. The fix is
4439 not to close the socket in ip_connect(); the socket should be closed by
4440 the function that creates it. There was only one place in the code where
4441 this was missing, in the iplookup router, which I don't think anybody now
4442 uses, but I've fixed it anyway.
4444 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
4445 well as to direct DNS lookups. Otherwise the handling of names in host
4446 lists is inconsistent and therefore confusing.
4452 PH/01 Two changes to the default runtime configuration:
4454 (1) Move the checks for relay_from_hosts and authenticated clients from
4455 after to before the (commented out) DNS black list checks.
4457 (2) Add control=submission to the relay_from_hosts and authenticated
4458 clients checks, on the grounds that messages accepted by these
4459 statements are most likely to be submissions.
4461 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
4463 (1) Generate an error if the third argument for the ${prvs expansion is
4466 (2) Treat a missing third argument of ${prvscheck as if it were an empty
4469 (3) Reset the variables that are obtained from the first argument of
4470 ${prvscheck and used in the second argument before leaving the code,
4471 because their memory is reclaimed, so using them afterwards may do
4474 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
4475 one (it's much easier than Tom thought :-).
4477 (5) Because of (4), we can now allow for the use of $prvscheck_result
4478 inside the third argument.
4480 PH/03 For some reason, the default setting of PATH when running a command from
4481 a pipe transport was just "/usr/bin". I have changed it to
4484 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
4485 anything to be listed in the output from -bV.
4487 PH/05 When a filter generated an autoreply, the entire To: header line was
4488 quoted in the delivery log line, like this:
4490 => >A.N.Other <ano@some.domain> <original@ddress> ...
4492 This has been changed so that it extracts the operative address. There
4493 may be more than one such address. If so, they are comma-separated, like
4496 => >ano@some.domain,ona@other.domain <original@ddress> ...
4498 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
4499 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
4500 not being looked up in the rDNS to get a host name, Exim was showing the
4501 IP address twice in Received: lines, even though the IP addresses were
4502 identical. For example:
4504 Received: from [1.2.3.4] (helo=[1.2.3.4])
4506 However, if the real host name was known, it was omitting the HELO data
4507 if it matched the actual IP address. This has been tidied up so that it
4508 doesn't show the same IP address twice.
4510 PH/07 When both +timestamp and +memory debugging was on, the value given by
4511 $tod_xxx expansions could be wrong, because the tod_stamp() function was
4512 called by the debug printing, thereby overwriting the timestamp buffer.
4513 Debugging no longer uses the tod_stamp() function when +timestamp is set.
4515 PH/08 When the original message was included in an autoreply transport, it
4516 always said "this is a copy of the message, including all the headers",
4517 even if body_only or headers_only was set. It now gives an appropriate
4520 PH/09 Applied a patch from the Sieve maintainer which:
4522 o fixes some comments
4523 o adds the (disabled) notify extension core
4524 o adds some debug output for the result of if/elsif tests
4525 o points to the current vacation draft in the documentation
4526 and documents the missing references header update
4530 o fixes a bug in processing the envelope test (when testing
4531 multiple envelope elements, the last element determined the
4534 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
4535 Electronic Mail") by including:
4537 Auto-submitted: auto-generated
4539 in the messages that it generates (bounce messages and others, such as
4540 warnings). In the case of bounce messages for non-SMTP messages, there was
4541 also a typo: it was using "Auto_submitted" (underscore instead of
4542 hyphen). Since every message generated by Exim is necessarily in response
4543 to another message, thes have all been changed to:
4545 Auto-Submitted: auto-replied
4547 in accordance with these statements in the RFC:
4549 The auto-replied keyword:
4551 - SHOULD be used on messages sent in direct response to another
4552 message by an automatic process,
4554 - MUST NOT be used on manually-generated messages,
4556 - MAY be used on Delivery Status Notifications (DSNs) and Message
4557 Disposition Notifications (MDNs),
4559 - MUST NOT be used on messages generated by automatic or periodic
4560 processes, except for messages which are automatic responses to
4563 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
4564 to the default Received: header definition.
4566 PH/12 Added log selector acl_warn_skipped (default on).
4568 PH/13 After a successful wildlsearch lookup, discard the values of numeric
4569 variables because (a) they are in the wrong storage pool and (b) even if
4570 they were copied, it wouldn't work properly because of the caching.
4572 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
4573 checking when decoding. Apparently there are clients that generate
4574 overlong encoded strings. Why am I not surprised?
4576 PH/15 If the first argument of "${if match_address" was not empty, but did not
4577 contain an "@" character, Exim crashed. Now it writes a panic log message
4578 and treats the condition as false.
4580 PH/16 In autoreply, treat an empty string for "once" the same as unset.
4582 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
4583 extension "envelope-auth". The code is finished and in agreement with
4584 other implementations, but there is no documentation so far and in fact,
4585 nobody wrote the draft yet. This extension is currently #undef'ed, thus
4586 not changing the active code.
4588 Print executed "if" and "elsif" statements when debugging is used. This
4589 helps a great deal to understand what a filter does.
4591 Document more things not specified clearly in RFC3028. I had all this
4592 sorted out, when out of a sudden new issues came to my mind. Oops."
4594 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
4597 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
4598 canonical form (as documented). However, after a host name lookup from
4599 the IP address, check_host() was doing a simple string comparison with
4600 addresses acquired from the DNS when checking that the found name did
4601 have the original IP as one of its addresses. Since any found IPv6
4602 addresses are likely to be in abbreviated form, the comparison could
4603 fail. Luckily, there already exists a function for doing the comparison
4604 by converting both addresses to binary, so now that is used instead of
4605 the text comparison.
4607 PH/20 There was another similar case to PH/19, when a complete host name was
4608 given in a host list; looking up its IP address could give an abbreviated
4609 form, whereas the current host's name might or might not be abbreviated.
4610 The same fix has been applied.
4616 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
4617 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
4620 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
4621 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
4623 PH/03 Typo: missing ".o" in src/pcre/Makefile.
4625 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
4626 header line, restrict the check to what is listed in RFCs 2369 and 2929.
4627 Also, for "Auto-Submitted", treat anything other than "no" as
4628 non-personal, in accordance with RFC 3834. (Previously it treated
4629 anything starting "auto-" as non-personal.)
4631 TF/01 The control=submission/name=... option had a problem with syntax
4632 errors if the name included a slash character. The /name= option
4633 now slurps the rest of the string, so it can include any characters
4634 but it must come last in the list of options (after /sender_retain
4637 PH/05 Some modifications to the interface to the fake nameserver for the new
4645 TK/01 Added the "success_on_redirect" address verification option. See
4646 NewStuff for rationale and an example.
4648 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
4650 PH/02 Patch to exigrep to allow it to work on syslog lines.
4652 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
4653 fread() to skip over the body file's header line, because in Cygwin the
4654 header line is locked and is inaccessible.
4656 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
4657 co-exist for some time) to make it clear that it is the Exim ID that is
4658 referenced, not the Message-ID: header line.
4660 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
4661 string_format() function, because snprintf() does not exist on all
4664 PH/06 The use of forbid_filter_existstest now also locks out the use of the
4665 ${stat: expansion item.
4667 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
4668 protocol synchronization error", to keep the pedants happy.
4670 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
4671 well as for IRIX systems, when gcc is being used. See the host.c source
4674 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
4676 PH/10 Named domain lists were not working if used in a queue_smtp_domains
4679 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
4680 transport and to the smtp transport in LMTP mode.
4682 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
4684 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
4685 run a filter in a subprocess. This could lead to confusion in subsequent
4686 lookups in the parent process. There should also be a search_tidyup() at
4687 the end of the subprocess.
4689 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
4690 only if the host matched helo_try_verify_hosts, which caused the
4691 verification to occur when the EHLO/HELO command was issued. The ACL just
4692 tested the remembered result. Now, if a previous verification attempt has
4693 not happened, "verify = helo" does it there and then.
4695 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
4697 TK/03 Fix log output including CR from clamd.
4699 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
4700 exist provoked a memory error which could cause a segfault.
4702 PH/15 Installed PCRE 6.2
4704 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
4706 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
4707 of the problem. Specifically, suggested +O2 rather than +O1 for the
4710 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
4712 PH/20 If a delivery was routed to a non-standard port by means of an SRV
4713 record, the port was not correctly logged when the outgoing_port log
4714 selector was set (it logged the transort's default port).
4716 PH/21 Added support for host-specific ports to manualroute, queryprogram,
4717 fallback_hosts, and "hosts" in the smtp transport.
4719 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
4720 host errors such as "Connection refused".
4722 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
4723 authentication with radiusclient 0.4.9:
4725 - Error returned from rc_read_config was caught wrongly
4726 - Username/password not passed on to radius server due to wrong length.
4728 The presumption is that some radiusclient API changes for 4.51/PH/17
4729 were not taken care of correctly. The code is still untested by me (my
4730 Linux distribution still has 0.3.2 of radiusclient), but it was
4731 contributed by a Radius user.
4733 PH/24 When doing a callout, the value of $domain wasn't set correctly when
4734 expanding the "port" option of the smtp transport.
4736 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
4737 while reading a MIME header. Thanks to Tom Hughes for a patch.
4739 PH/24 Include config.h inside local_scan.h so that configuration settings are
4742 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
4743 that in an ACL for RCPT (for example), you can examine exactly what was
4746 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
4747 commands, but it was not correctly comparing the address with the actual
4748 client host address. Thus, it would show the EHLO address in Received:
4749 header lines when this was not necessary.
4751 PH/27 Added the % operator to ${eval:}.
4753 PH/28 Exim tries to create and chdir to its spool directory when it starts;
4754 it should be ignoring failures (because with -C, for example, it has lost
4755 privilege). It wasn't ignoring creation failures other than "already
4758 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
4761 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
4762 because an input line was too long, either on its own, or by virtue of
4763 too many continuations, the temporary file was not being removed, and the
4764 return code was incorrect.
4766 PH/31 Missing "BOOL" in function definition in filtertest.c.
4768 PH/32 Applied Sieve patches from the maintainer.
4770 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
4772 PH/33 Added "verify = not_blind".
4774 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
4775 Local/Makefile (with some defaults set). These are used in built scripts
4776 such as exicyclog, but they have never been used in the exim_install
4777 script (though there are many overriding facilities there). I have
4778 arranged that the exim_install script now takes note of these two
4781 PH/35 Installed configuration files for Dragonfly.
4783 PH/36 When a locally submitted message by a trusted user did not contain a
4784 From: header, and the sender address was obtained from -f or from an SMTP
4785 MAIL command, and the trusted user did not use -F to supply a sender
4786 name, $originator_name was incorrectly used when constructing a From:
4787 header. Furthermore, $originator_name was used for submission mode
4788 messages from external hosts without From: headers in a similar way,
4789 which is clearly wrong.
4791 PH/37 Added control=suppress_local_fixups.
4793 PH/38 When log_selector = +received_sender was set, and the addition of the
4794 sender made the log line's construction buffer exactly full, or one byte
4795 less than full, an overflow happened when the terminating "\n" was
4798 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
4799 when the result of a list match is failure because a DNS lookup failed.
4801 PH/40 RM_COMMAND is now used in the building process.
4803 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
4804 the "build-* directories that it finds.
4806 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
4807 address in a domain literal was a prefix of an interface address.
4809 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
4810 when verifying a sender address, unless rewrite_headers is false.
4812 PH/44 Wrote a long comment about why errors_to addresses are verified as
4813 recipients, not senders.
4815 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
4816 the ratelimit ACL was added.
4818 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
4820 PH/46 Added extra information about PostgreSQL errors to the error string.
4822 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
4823 avoiding the need to install special zones in a real server. This is
4824 backwards compatible; if it can't find the fake resolver, it drops back.
4825 Thus, both old and new test suites can be run.
4827 TF/02 Added util/ratelimit.pl
4829 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
4830 clock is set back in time.
4832 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
4833 Candler <B.Candler@pobox.com>.
4835 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
4836 OK for addresses that are the result of redirections.
4838 PH/48 A number of further additions for the benefit of the new test suite,
4839 including a fake gethostbyname() that interfaces to the fake DNS resolver
4842 TF/06 The fix for widen_domains has also been applied to qualify_single and
4843 search_parents which are the other dnslookup options that can cause
4846 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
4849 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
4851 TF/07 Exim produced the error message "an SRV record indicated no SMTP
4852 service" if it encountered an MX record with an empty target hostname.
4853 The message is now "an MX or SRV record indicated no SMTP service".
4855 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
4856 if the DNS of the sending site is misconfigured. This is quite a
4857 common situation. This change restores the behaviour of treating a
4858 helo verification defer as a failure.
4860 PH/51 If self=fail was set on a router, the bounce message did not include the
4861 actual error message.
4867 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
4869 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4870 command itself ended in error, the underlying message about the transport
4871 filter timeout was being overwritten with the pipe command error. Now the
4872 underlying error message should be appended to the second error message.
4874 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4876 PH/02 Reduce the amount of output that "make" produces by default. Full output
4877 can still be requested.
4879 PH/03 The warning log line about a condition test deferring for a "warn" verb
4880 was being output only once per connection, rather than after each
4881 occurrence (because it was using the same function as for successful
4882 "warn" verbs). This seems wrong, so I have changed it.
4884 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4885 it should not have, which might have caused a crash in the right
4886 circumstances, but probably never did.
4888 PH/04 Installed a modified version of Tony Finch's patch to make submission
4889 mode fix the return path as well as the Sender: header line, and to
4890 add a /name= option so that you can make the user's friendly name appear
4893 TF/03 Added the control = fakedefer ACL modifier.
4895 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4896 Mark Lowes for thorough testing.
4898 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4900 TK/03 Merged latest SRS patch from Miles Wilton.
4902 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4903 with the definition in sysexits.h (which is #included earlier).
4904 Fortunately, Exim does not actually use EX_OK. The code used to try to
4905 preserve the sysexits.h value, by assuming that macro definitions were
4906 scanned for macro replacements. I have been disabused of this notion,
4907 so now the code just undefines EX_OK before #including unistd.h.
4909 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4910 in the smtp transport. When a block could not be written in a single
4911 write() function, the timeout was being re-applied to each part-write.
4912 This seems wrong - if the receiver was accepting one byte at a time it
4913 would take for ever. The timeout is now adjusted when this happens. It
4914 doesn't have to be particularly precise.
4916 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4917 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4919 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4920 without a domain if the check to <postmaster@domain> fails.
4922 SC/01 Eximstats: added -xls and the ability to specify output files
4923 (patch written by Frank Heydlauf).
4925 SC/02 Eximstats: use FileHandles for outputting results.
4927 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4929 SC/04 Eximstats: fixed display of large numbers with -nvr option
4931 SC/05 Eximstats: fixed merging of reports with empty tables.
4933 SC/06 Eximstats: added the -include_original_destination flag
4935 SC/07 Eximstats: removed tabs and trailing whitespace.
4937 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4939 TK/06 MBOX spool code: Add real "From " MBOX separator line
4940 so the .eml file is really in mbox format (even though
4941 most programs do not really care). Patch from Alex Miller.
4943 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4944 The latter is generated from $received_to and is only set if the
4945 message has one envelope recipient. SA can use these headers,
4946 obviously out-of-the-box. Patch from Alex Miller.
4948 PH/08 The ${def test on a variable was returning false if the variable's
4949 value was "0", contrary to what the specification has always said!
4950 The result should be true unless the variable is empty.
4952 PH/09 The syntax error of a character other than { following "${if
4953 def:variable_name" (after optional whitespace) was not being diagnosed.
4954 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4955 accidental colon was present, for example, could give incorrect results.
4957 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4958 result is used (not including appendfile, where other changes are about
4961 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4962 This involved changing a lot of size variables from int to off_t. It
4963 should work with maildirs and everything.
4965 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4966 spamd dying while we are connected to it.
4968 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4971 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4972 function for BDB 4.3.
4974 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4976 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4977 the sources for PCRE have been re-arranged and more files are now
4980 PH/15 The code I had for printing potentially long long variables in PH/11
4981 above was not the best (it lost precision). The length of off_t variables
4982 is now inspected at build time, and an appropriate printing format (%ld
4983 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4984 to be "long long int" or "long int". This is needed for the internal
4985 formatting function string_vformat().
4987 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4988 the configuration file to be ":syslog", then the script "guesses" where
4989 the logs files are, rather than using the compiled in default. In our
4990 case the guess is not the same as the compiled default, so the script
4991 suddenly stopped working when I started to use syslog. The patch checks
4992 to see if log_file_path is "". If so, it attempts to read it from exim
4993 with no configuration file to get the compiled in version, before it
4994 falls back to the previous guessing code."
4996 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4997 implementing BATV in an Exim configuration. See NewStuff for the gory
5000 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
5001 Makefile that are specific to HP-UX.
5003 PH/18 If the "use_postmaster" option was set for a recipient callout together
5004 with the "random" option, the postmaster address was used as the MAIL
5005 FROM address for the random test, but not for the subsequent recipient
5006 test. It is now used for both.
5008 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
5009 patch removes a few documentation additions to RFC 3028, because the
5010 latest draft now contains them. It adds the new en;ascii-case comparator
5011 and a new error check for 8bit text in MIME parts. Comparator and
5012 require names are now matched exactly. I enabled the subaddress
5013 extension, but it is not well tested yet (read: it works for me)."
5015 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
5016 rework some of the code of TK/09 above to avoid the hardwired use of
5017 "%lld" and "long long". Replaced the call to snprintf() with a call to
5020 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
5021 records point to non-existent hosts", "retry timeout exceeded", and
5022 "retry time not reached for any host after a long failure period".
5024 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
5025 experimental DomainKeys support:
5027 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
5028 (2) On an error such as an illegally used "control", the wrong name for
5029 the control was given.
5031 These problems did NOT occur unless DomainKeys support was compiled.
5033 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
5035 PH/24 Added ${if match_ip condition.
5037 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
5038 databases so that it will be absolutely obvious if a crash occurs in the
5039 DB library. This is a regular occurrence (often caused by mis-matched
5042 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
5043 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
5044 picked up on a user's system that detects such things. There doesn't seem
5045 to be a gcc warning option for this - only an attribute that has to be
5046 put on the function's prototype. It seems that in Fedora Core 4 they have
5047 set this on a number of new functions. No doubt there will be more in due
5050 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
5051 specify a transport. However, if an address that was verified by such a
5052 router was the subject of a callout, Exim crashed because it tried to
5053 read the rcpt_include_affixes from the non-existent transport. Now it
5054 just assumes that the setting of that option is false. This bug was
5055 introduced by 4.51/PH/31.
5057 PH/28 Changed -d+all to exclude +memory, because that information is very
5058 rarely of interest, but it makes the output a lot bigger. People tend to
5059 do -d+all out of habit.
5061 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
5062 code in os-type was giving problems when libc.so lives in lib64, like on
5065 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
5066 aren't the modern standard, and it seems that some systems' include files
5067 don't always have them. Exim was already checking for some of the newer
5068 ones like T_AAAA, and defining it itself. I've added checks for all the
5069 record types that Exim uses.
5071 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
5072 not automatically generating a new one, as it is supposed to. This
5073 prevented TLS from working. If the file did exist, but contained invalid
5074 data, a new version was generated, as expected. It was only the case of a
5075 non-existent file that was broken.
5077 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
5078 with a change in libdomainkeys > 0.64.
5080 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
5081 from DNS. If the selector record carries the flag, it now has
5082 precedence over the domain-wide flag.
5084 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
5086 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
5087 the use of a port name that isn't defined in /etc/services) occurred, the
5088 message was deferred as in a normal delivery, and thus remained on the
5089 spool, instead of being failed because of the mua_wrapper setting. This
5090 is now fixed, and I tidied up some of the mua_wrapper messages at the
5093 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
5094 the messages in a hash of arrays rather than using individual hashes.
5095 This is a bit cleaner and results in dramatic memory savings, albeit
5096 at a slight CPU cost.
5098 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
5099 as requested by Marc Sherman.
5101 SC/10 Eximstats: added histograms for user specified patterns as requested
5104 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
5106 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
5107 fopen() in the content-scanning modules that did not already have it.
5113 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
5114 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
5116 TK/02 Fix ACL "control" statement not being available in MIME ACL.
5118 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
5120 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
5121 to test Sieve filters that use "vacation".
5123 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
5124 that changes the way the GnuTLS parameters are stored in the cache file.
5125 The new format can be generated externally. For backward compatibility,
5126 if the data in the cache doesn't make sense, Exim assumes it has read an
5127 old-format file, and it generates new data and writes a new file. This
5128 means that you can't go back to an older release without removing the
5131 PH/03 A redirect router that has both "unseen" and "one_time" set does not
5132 work if there are any delivery delays because "one_time" forces the
5133 parent to be marked "delivered", so its unseen clone is never tried
5134 again. For this reason, Exim now forbids the simultaneous setting of
5137 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
5138 redirected to themselves ("homonym" addresses). Read the long ChangeLog
5139 entry if you want to know the details. The fix, however, neglected to
5140 consider the case when local delivery batching is involved. The test for
5141 "previously delivered" was not happening when checking to see if an
5142 address could be batched with a previous (undelivered) one; under
5143 certain circumstances this could lead to multiple deliveries to the same
5146 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
5147 in its include files, and this causes problems building Exim.
5149 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
5150 header_syntax) but Exim was just ignoring anything given after a slash.
5151 In particular, this caused confusion with an attempt to use "verify =
5152 reverse_host_lookup/defer_ok". An error is now given when options are
5153 supplied for verify items that do not have them. (Maybe reverse_host_
5154 lookup should have a defer_ok option, but that's a different point.)
5156 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
5157 defined by RFC 821) to 2048, because there were problems with some AUTH
5158 commands, and RFC 1869 says the size should be increased for extended
5159 SMTP commands that take arguments.
5161 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
5164 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
5165 "unknown" error; now it says that the functionality isn't in the binary.
5167 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
5168 an address' error message when a string expansion fails (syntax or
5169 whatever). Otherwise the password may appear in the log. Following change
5170 PH/42 below, there is no longer a chance of it appearing in a bounce
5173 PH/11 Installed exipick version 20050225.0 from John Jetmore.
5175 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
5176 of its addresses was ever tried. (Bugzilla bug #2.)
5178 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
5179 the result incorrectly in the debug output. (It correctly added a newline
5180 to what was transported.)
5182 TF/01 Added $received_time.
5184 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
5185 commented out examples of how to interface to a virus scanner and to
5186 SpamAssassin. Also added commented examples of av_scanner and
5187 spamd_address settings.
5189 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
5190 and controls are allowed in which ACLs. There were a couple of minor
5191 errors. Some of the entries in the conditions table (which is a table of
5192 where they are NOT allowed) were getting very unwieldy; rewrote them as a
5193 negation of where the condition IS allowed.
5195 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
5197 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
5198 header file does not have a version number, so I've had to invent a new
5199 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
5200 API. The code is untested by me (my Linux distribution still has 0.3.2 of
5201 radiusclient), but it was contributed by a Radius user.
5203 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
5204 files or directories, for OpenSSL.
5206 PH/19 When an Exim process that is running as root has to create an Exim log
5207 file, it does so in a subprocess that runs as exim:exim so as to get the
5208 ownership right at creation (otherwise, other Exim processes might see
5209 the file with the wrong ownership). There was no test for failure of this
5210 fork() call, which would lead to the process getting stuck as it waited
5211 for a non-existent subprocess. Forks do occasionally fail when resources
5212 run out. I reviewed all the other calls to fork(); they all seem to check
5215 PH/20 When checking for unexpected SMTP input at connect time (before writing
5216 the banner), Exim was not dealing correctly with a non-positive return
5217 from the read() function. If the client had disconnected by this time,
5218 the result was a log entry for a synchronization error with an empty
5219 string after "input=" when read() returned zero. If read() returned -1
5220 (an event I could not check), uninitialized data bytes were printed.
5221 There were reports of junk text (parts of files, etc) appearing after
5224 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
5226 PH/22 Added support for macro redefinition, and (re)definition in between
5227 driver and ACL definitions.
5229 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
5230 forgetting to use the resulting value; it was using the unexpanded value.
5232 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
5233 hadn't been configured. The fix is from Juergen Kreileder, who
5234 understands it better than I do:
5236 "Here's what I see happening with three configured cyrus_sasl
5237 authenticators configured (plain, login, cram-md5):
5239 On startup auth_cyrus_sasl_init() gets called for each of these.
5240 This means three calls to sasl_listmech() without a specified mech_list.
5241 => SASL tests which mechs of all available mechs actually work
5242 => three warnings about OTP not working
5243 => the returned list contains: plain, login, cram-md5, digest-md5, ...
5245 With the patch, sasl_listmech() also gets called three times. But now
5246 SASL's mech_list option is set to the server_mech specified in the the
5247 authenticator. Or in other words, the answer from sasl_listmech()
5248 gets limited to just the mech you're testing for (which is different
5250 => the return list contains just 'plain' or 'login', 'cram-md5' or
5251 nothing depending on the value of ob->server_mech.
5253 I've just tested the patch: Authentication still works fine,
5254 unavailable mechs specified in the exim configuration are still
5255 caught, and the auth.log warnings about OTP are gone."
5257 PH/25 When debugging is enabled, the contents of the command line are added
5258 to the debugging output, even when log_selector=+arguments is not
5261 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
5262 answer is "GNU", and only if the return is "GNU/something" is the answer
5265 PH/27 $acl_verify_message is now set immediately after the failure of a
5266 verification in an ACL, and so is available in subsequent modifiers. In
5267 particular, the message can be preserved by coding like this:
5269 warn !verify = sender
5270 set acl_m0 = $acl_verify_message
5272 Previously, $acl_verify_message was set only while expanding "message"
5273 and "log_message" when a very denied access.
5275 PH/28 Modified OS/os.c-Linux with
5277 -#ifndef OS_LOAD_AVERAGE
5278 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
5280 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
5281 nomenclature these days.)
5283 PH/29 Installed patch from the Sieve maintainer that adds the options
5284 sieve_useraddress and sieve_subaddress to the redirect router.
5286 PH/30 In these circumstances:
5287 . Two addresses routed to the same list of hosts;
5288 . First host does not offer TLS;
5289 . First host accepts first address;
5290 . First host gives temporary error to second address;
5291 . Second host offers TLS and a TLS session is established;
5292 . Second host accepts second address.
5293 Exim incorrectly logged both deliveries with the TLS parameters (cipher
5294 and peerdn, if requested) that were in fact used only for the second
5297 PH/31 When doing a callout as part of verifying an address, Exim was not paying
5298 attention to any local part prefix or suffix that was matched by the
5299 router that accepted the address. It now behaves in the same way as it
5300 does for delivery: the affixes are removed from the local part unless
5301 rcpt_include_affixes is set on the transport.
5303 PH/32 Add the sender address, as F=<...>, to the log line when logging a
5304 timeout during the DATA phase of an incoming message.
5306 PH/33 Sieve envelope tests were broken for match types other than :is. I have
5307 applied a patch sanctioned by the Sieve maintainer.
5309 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
5310 the uid or gid is negative. A case of a negative gid caused this to be
5311 noticed. The fix allows for either to be negative.
5313 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
5314 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
5317 PH/36 The change PH/12 above was broken. Fixed it.
5319 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
5320 the grounds that routing the same address twice would always produce the
5321 same answer. This might have been true once, but it is certainly no
5322 longer true now. Routing a child address may depend on the previous
5323 routing that produced that child. Some complicated redirection strategies
5324 went wrong when messages had multiple recipients, and made Exim's
5325 behaviour dependent on the order in which the addresses were given.
5327 I have moved the duplicate checking until after the routing is complete.
5328 Exim scans the addresses that are assigned to local and remote
5329 transports, and removes any duplicates. This means that more work will be
5330 done, as duplicates will always all be routed, but duplicates are
5331 presumably rare, so I don't expect this is of any significance.
5333 For deliveries to pipes, files, and autoreplies, the duplicate checking
5334 still happens during the routing process, since they are not going to be
5337 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
5338 It corrects a timeout issue with spamd. This is Ian's comment: "The
5339 background is that sometimes spamd either never reads data from a
5340 connection it has accepted, or it never writes response data. The exiscan
5341 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
5342 blindly assumes that writes won't block so it may never time out."
5344 PH/39 Allow G after quota size as well as K and M.
5346 PH/40 The value set for $authenticated_id in an authenticator may not contain
5347 binary zeroes or newlines because the value is written to log lines and
5348 to spool files. There was no check on this. Now the value is run through
5349 the string_printing() function so that such characters are converted to
5350 printable escape sequences.
5352 PH/41 $message_linecount is a new variable that contains the total number of
5353 lines in the message. Compare $body_linecount, which is the count for the
5356 PH/42 Exim no longer gives details of delivery errors for specific addresses in
5357 bounce and delay warning messages, except in certain special cases, which
5360 (a) An SMTP error message from a remote host;
5361 (b) A message specified in a :fail: redirection;
5362 (c) A message specified in a "fail" command in a system filter;
5363 (d) A message specified in a FAIL return from the queryprogram router;
5364 (e) A message specified by the cannot_route_message router option.
5366 In these cases only, Exim does include the error details in bounce and
5367 warning messages. There are also a few cases where bland messages such
5368 as "unrouteable address" or "local delivery error" are given.
5370 PH/43 $value is now also set for the "else" part of a ${run expansion.
5372 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
5373 being worked on, but at least Exim now implements the latest version to
5376 PH/45 In a pipe transport, although a timeout while waiting for the pipe
5377 process to complete was treated as a delivery failure, a timeout while
5378 writing the message to the pipe was logged, but erroneously treated as a
5379 successful delivery. Such timeouts include transport filter timeouts. For
5380 consistency with the overall process timeout, these timeouts are now
5381 treated as errors, giving rise to delivery failures by default. However,
5382 there is now a new Boolean option for the pipe transport called
5383 timeout_defer, which, if set TRUE, converts the failures into defers for
5384 both kinds of timeout. A transport filter timeout is now identified in
5387 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
5388 systems where "make" and "gmake" are different, calling "gmake" at top
5389 level broke things. I've arranged for the value of $(MAKE) to be passed
5390 from the Makefile to this script so that it can call the same version of
5394 A note about Exim versions 4.44 and 4.50
5395 ----------------------------------------
5397 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
5398 changes of various kinds. As a consequence, a big documentation update was
5399 needed. This delayed the release for rather longer than seemed good, especially
5400 in the light of a couple of (minor) security issues. Therefore, the changes
5401 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
5402 release. So 4.44 and 4.50 are in effect two different branches that both start
5405 I have left the 4.50 change log unchanged; it contains all the changes since
5406 4.43. The change log for 4.44 is below; many of its items are identical to
5407 those for 4.50. This seems to be the most sensible way to preserve the
5408 historical information.
5414 1. Minor wording change to the doc/README.SIEVE file.
5416 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
5417 computation of the current number of files was incorrect.
5419 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5420 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5423 4. Give more explanation in the error message when the command for a transport
5424 filter fails to execute.
5426 5. There are several places where Exim runs a non-Exim command in a
5427 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5428 was being done only for the command run by the queryprogram router. It is
5429 now done for all such subprocesses. The other cases are: ${run, transport
5430 filters, and the commands run by the lmtp and pipe transports.
5432 6. Added CONFIGURE_GROUP build-time option.
5434 7. Some older OS have a limit of 256 on the maximum number of file
5435 descriptors. Exim was using setrlimit() to set 1000 as a large value
5436 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5437 systems. I've change it so that if it can't get 1000, it tries for 256.
5439 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5440 was an oversight, and furthermore, ever since the addition of extra
5441 controls (e.g. 4.43/32), the checks on when to allow different forms of
5442 "control" were broken. There should now be diagnostics for all cases when a
5443 control that does not make sense is encountered.
5445 9. Added the /retain_sender option to "control=submission".
5447 10. $recipients is now available in the predata ACL (oversight).
5449 11. Tidy the search cache before the fork to do a delivery from a message
5450 received from the command line. Otherwise the child will trigger a lookup
5451 failure and thereby defer the delivery if it tries to use (for example) a
5452 cached ldap connection that the parent has called unbind on.
5454 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5455 of $address_data from the recipient verification was clobbered by the
5456 sender verification.
5458 13. The value of address_data from a sender verification is now available in
5459 $sender_address_data in subsequent conditions in the ACL statement.
5461 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
5463 15. Added a new option "connect=<time>" to callout options, to set a different
5466 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5467 was its contents. (It was OK if the option was not defined at all.)
5469 17. A "Completed" log line is now written for messages that are removed from
5470 the spool by the -Mrm option.
5472 18. New variables $sender_verify_failure and $recipient_verify_failure contain
5473 information about exactly what failed.
5475 19. Added -dd to debug only the daemon process.
5477 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
5478 handles timeouts, both on the server side and network timeouts. Renamed the
5479 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
5481 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
5482 It is now set to "smtps".
5484 22. $host_address is now set to the target address during the checking of
5485 ignore_target_hosts.
5487 23. When checking ignore_target_hosts for an ipliteral router, no host name was
5488 being passed; this would have caused $sender_host_name to have been used if
5489 matching the list had actually called for a host name (not very likely,
5490 since this list is usually IP addresses). A host name is now passed as
5493 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5494 code that specifies a non-restarting handler (typically sigaction() in
5495 modern systems) in an attempt to fix a rare and obscure crash bug.
5497 25. Narrowed the window for a race in the daemon that could cause it to ignore
5498 SIGCHLD signals. This is not a major problem, because they are used only to
5499 wake it up if nothing else does.
5501 26. A malformed maildirsize file could cause Exim to calculate negative values
5502 for the mailbox size or file count. Odd effects could occur as a result.
5503 The maildirsize information is now recalculated if the size or filecount
5506 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5507 support for a long time. Removed HAVE_SYS_VFS_H.
5509 28. Installed the latest version of exipick from John Jetmore.
5511 29. In an address list, if the pattern was not a regular expression, an empty
5512 subject address (from a bounce message) matched only if the pattern was an
5513 empty string. Non-empty patterns were not even tested. This was the wrong
5514 because it is perfectly reasonable to use an empty address as part of a
5515 database query. An empty address is now tested by patterns that are
5516 lookups. However, all the other forms of pattern expect the subject to
5517 contain a local part and a domain, and therefore, for them, an empty
5518 address still always fails if the pattern is not itself empty.
5520 30. Exim went into a mad DNS loop when attempting to do a callout where the
5521 host was specified on an smtp transport, and looking it up yielded more
5522 than one IP address.
5524 31. Re-factored the code for checking spool and log partition space into a
5525 function that finds that data and another that does the check. The former
5526 is then used to implement four new variables: $spool_space, $log_space,
5527 $spool_inodes, and $log_inodes.
5529 32. The RFC2047 encoding function was originally intended for short strings
5530 such as real names; it was not keeping to the 75-character limit for
5531 encoded words that the RFC imposes. It now respects the limit, and
5532 generates multiple encoded words if necessary. To be on the safe side, I
5533 have increased the buffer size for the ${rfc2047: expansion operator from
5536 33. It is now permitted to omit both strings after an "if" condition; if the
5537 condition is true, the result is "true". As before, when the second string
5538 is omitted, a false condition yields an empty string. This makes it less
5539 cumbersome to write custom ACL and router conditions.
5541 34. Failure to deliver a bounce message always caused it to be frozen, even if
5542 there was an errors_to setting on the router. The errors_to setting is now
5545 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5546 canonical form (fully expanded) before being placed in
5547 $sender_host_address.
5549 36. The table in the code that translates DNS record types into text (T_A to
5550 "A" for instance) was missing entries for NS and CNAME. It is just possible
5551 that this could have caused confusion if both these types were looked up
5552 for the same domain, because the text type is used as part of Exim's
5553 per-process caching. But the chance of anyone hitting this buglet seems
5556 37. The dnsdb lookup has been extended in a number of ways.
5558 (1) There is a new type, "zns", which walks up the domain tree until it
5559 finds some nameserver records. It should be used with care.
5561 (2) There is a new type, "mxh", which is like "mx" except that it returns
5562 just the host names, not the priorities.
5564 (3) It is now possible to give a list of domains (or IP addresses) to be
5565 looked up. The behaviour when one of the lookups defers can be
5566 controlled by a keyword.
5568 (4) It is now possible to specify the separator character for use when
5569 multiple records are returned.
5571 38. The dnslists ACL condition has been extended: it is now possible to supply
5572 a list of IP addresses and/or domains to be looked up in a particular DNS
5575 39. Added log_selector=+queue_time_overall.
5577 40. When running the queue in the test harness, wait just a tad after forking a
5578 delivery process, to get repeatability of debugging output.
5580 41. Include certificate and key file names in error message when GnuTLS fails
5581 to set them up, because the GnuTLS error message doesn't include the name
5582 of the failing file when there is a problem reading it.
5584 42. Allow both -bf and -bF in the same test run.
5586 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
5588 44. The "Exiscan patch" is now merged into the mainline Exim source.
5590 45. Sometimes the final signoff response after QUIT could fail to get
5591 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5592 < 0 before doing a fflush(). This bug looks as though it goes back to the
5593 introduction of TLS in release 3.20, but "sometimes" must have been rare
5594 because the tests only now provoked it.
5596 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
5597 (this can affect the format of dates).
5599 47. exim_tidydb, when checking for the continued existence of a message for
5600 which it has found a message-specific retry record, was not finding
5601 messages that were in split spool directories. Consequently, it was
5602 deleting retry records that should have stayed in existence.
5604 48. Steve fixed some bugs in eximstats.
5606 49. The SPA authentication driver was not abandoning authentication and moving
5607 on to the next authenticator when an expansion was forced to fail,
5608 contradicting the general specification for all authenticators. Instead it
5609 was generating a temporary error. It now behaves as specified.
5611 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
5612 (the order specifies the preference for clients). The order is now AES256,
5613 AES128, 3DES, ARCFOUR128.
5615 51. Small patch to Sieve code - explicitly set From: when generating an
5618 52. Exim crashed if a remote delivery caused a very long error message to be
5619 recorded - for instance if somebody sent an entire SpamAssassin report back
5620 as a large number of 550 error lines. This bug was coincidentally fixed by
5621 increasing the size of one of Exim's internal buffers (big_buffer) that
5622 happened as part of the Exiscan merge. However, to be on the safe side, I
5623 have made the code more robust (and fixed the comments that describe what
5626 53. Now that there can be additional text after "Completed" in log lines (if
5627 the queue_time_overall log selector is set), a one-byte patch to exigrep
5628 was needed to allow it to recognize "Completed" as not the last thing in
5631 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
5632 patch that reportedly fixes this has been added. I am not expert enough to
5633 create a test for it. This is what the patch creator wrote:
5635 "I found a little strange behaviour of ldap code when working with
5636 Windows 2003 AD Domain, where users was placed in more than one
5637 Organization Units. When I tried to give exim partial DN, the exit code
5638 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
5639 But simultaneously result of request was absolutely normal ldap result,
5640 so I produce this patch..."
5642 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
5643 REFERENCE, so I have modified the code to exclude the patch when that macro
5646 55. Some experimental protocols are using DNS PTR records for new purposes. The
5647 keys for these records are domain names, not reversed IP addresses. The
5648 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5649 leaves it alone. Component reversal etc. now happens only for IP addresses.
5652 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
5654 57. Double the size of the debug message buffer (to 2048) so that more of very
5655 long debug lines gets shown.
5657 58. The exicyclog utility now does better if the number of log files to keep
5658 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
5660 59. Two changes related to the smtp_active_hostname option:
5662 (1) $smtp_active_hostname is now available as a variable.
5663 (2) The default for smtp_banner uses $smtp_active_hostname instead
5664 of $primary_hostname.
5666 60. The host_aton() function is supposed to be passed a string that is known
5667 to be a valid IP address. However, in the case of IPv6 addresses, it was
5668 not checking this. This is a hostage to fortune. Exim now panics and dies
5669 if the condition is not met. A case was found where this could be provoked
5670 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5671 components; fortuitously, this particular loophole had already been fixed
5672 by change 4.50/55 above.
5674 If there are any other similar loopholes, the new check in host_aton()
5675 itself should stop them being exploited. The report I received stated that
5676 data on the command line could provoke the exploit when Exim was running as
5677 exim, but did not say which command line option was involved. All I could
5678 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5679 running as the user.
5682 61. There was a buffer overflow vulnerability in the SPA authentication code
5683 (which came originally from the Samba project). I have added a test to the
5684 spa_base64_to_bits() function which I hope fixes it.
5687 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
5688 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
5690 63. The daemon start-up calls getloadavg() while still root for those OS that
5691 need the first call to be done as root, but it missed one case: when
5692 deliver_queue_load_max is set with deliver_drop_privilege. This is
5693 necessary for the benefit of the queue runner, because there is no re-exec
5694 when deliver_drop_privilege is set.
5696 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
5697 This has been fixed.
5699 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5700 was in use, was not putting the data itself into the right store pool;
5701 consequently, it could be overwritten for a subsequent message in the same
5702 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5705 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
5707 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
5708 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
5709 the function were treating the return as a boolean value, which happened to
5710 work because 0=false and not-0=true, but is not correct code.
5712 68. The host_aton() function was not handling scoped IPv6 addresses (those
5713 with, for example, "%eth0" on the end) correctly.
5715 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
5716 negated items (that is, ~something) in unsigned ints. Some compilers
5717 apparently mutter when there is no cast.
5719 70. If an address verification called from an ACL failed, and did not produce a
5720 user-specific message (i.e. there was only a "system" message), nothing was
5721 put in $acl_verify_message. In this situation, it now puts the system
5724 71. Change 4.23/11 added synchronization checking at the start of an SMTP
5725 session; change 4.31/43 added the unwanted input to the log line - except
5726 that it did not do this in the start of session case. It now does.
5728 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
5729 This is wrong and can cause the other end to generate a synchronization
5730 error if it is another Exim or anything else that does the synchronization
5731 check. A QUIT command is no longer sent after a timeout.
5733 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
5734 during host lookups.
5736 74. The defer_ok option of callout verification was not working if it was used
5737 when verifying addresses in header lines, that is, for this case:
5739 verify = header_sender/callout=defer_ok
5741 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
5742 those file descriptors could be used for SMTP connections. If anything
5743 wrote to stderr (the example that came up was "warn" in embedded Perl), it
5744 could be sent to the SMTP client, causing chaos. The daemon now opens
5745 stdin, stdout, and stderr to /dev/null when it puts itself into the
5748 76. Arrange for output from Perl's "warn" command to be written to Exim's main
5749 log by default. The user can override this with suitable Perl magic.
5751 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
5752 the log message when discard triggers, was not working for the DATA ACL or
5753 for the non-SMTP ACL.
5755 78. Error message wording change in sieve.c.
5757 79. If smtp_accept_max_per_host was set, the number of connections could be
5758 restricted to fewer than expected, because the daemon was trying to set up
5759 a new connection before checking whether the processes handling previous
5760 connections had finished. The check for completed processes is now done
5761 earlier. On busy systems, this bug wouldn't be noticed because something
5762 else would have woken the daemon, and it would have reaped the completed
5765 80. If a message was submitted locally by a user whose login name contained one
5766 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
5767 It caused a spool format error. I have fixed the spool reading code. A
5768 related problem was that the "from" clause in the Received: line became
5769 illegal because of the space(s). It is now covered by ${quote_local_part.
5771 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
5774 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
5776 83. Patch from Sieve maintainer to fix unterminated string problem in
5777 "vacation" handling.
5779 84. Some minor changes to the Linux configuration files to help with other
5780 OS variants using glibc.
5782 85. One more patch for Sieve to update vacation handling to latest spec.
5785 ----------------------------------------------------
5786 See the note above about the 4.44 and 4.50 releases.
5787 ----------------------------------------------------
5793 1. Change 4.43/35 introduced a bug that caused file counts to be
5794 incorrectly computed when quota_filecount was set in an appendfile
5797 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5798 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5801 3. Give more explanation in the error message when the command for a transport
5802 filter fails to execute.
5804 4. There are several places where Exim runs a non-Exim command in a
5805 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5806 was being done only for the command run by the queryprogram router. It is
5807 now done for all such subprocesses. The other cases are: ${run, transport
5808 filters, and the commands run by the lmtp and pipe transports.
5810 5. Some older OS have a limit of 256 on the maximum number of file
5811 descriptors. Exim was using setrlimit() to set 1000 as a large value
5812 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5813 systems. I've change it so that if it can't get 1000, it tries for 256.
5815 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5816 was an oversight, and furthermore, ever since the addition of extra
5817 controls (e.g. 4.43/32), the checks on when to allow different forms of
5818 "control" were broken. There should now be diagnostics for all cases when a
5819 control that does not make sense is encountered.
5821 7. $recipients is now available in the predata ACL (oversight).
5823 8. Tidy the search cache before the fork to do a delivery from a message
5824 received from the command line. Otherwise the child will trigger a lookup
5825 failure and thereby defer the delivery if it tries to use (for example) a
5826 cached ldap connection that the parent has called unbind on.
5828 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5829 of $address_data from the recipient verification was clobbered by the
5830 sender verification.
5832 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5833 was its contents. (It was OK if the option was not defined at all.)
5835 11. A "Completed" log line is now written for messages that are removed from
5836 the spool by the -Mrm option.
5838 12. $host_address is now set to the target address during the checking of
5839 ignore_target_hosts.
5841 13. When checking ignore_target_hosts for an ipliteral router, no host name was
5842 being passed; this would have caused $sender_host_name to have been used if
5843 matching the list had actually called for a host name (not very likely,
5844 since this list is usually IP addresses). A host name is now passed as
5847 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5848 code that specifies a non-restarting handler (typically sigaction() in
5849 modern systems) in an attempt to fix a rare and obscure crash bug.
5851 15. Narrowed the window for a race in the daemon that could cause it to ignore
5852 SIGCHLD signals. This is not a major problem, because they are used only to
5853 wake it up if nothing else does.
5855 16. A malformed maildirsize file could cause Exim to calculate negative values
5856 for the mailbox size or file count. Odd effects could occur as a result.
5857 The maildirsize information is now recalculated if the size or filecount
5860 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5861 support for a long time. Removed HAVE_SYS_VFS_H.
5863 18. Updated exipick to current release from John Jetmore.
5865 19. Allow an empty sender to be matched against a lookup in an address list.
5866 Previously the only cases considered were a regular expression, or an
5869 20. Exim went into a mad DNS lookup loop when doing a callout where the
5870 host was specified on the transport, if the DNS lookup yielded more than
5873 21. The RFC2047 encoding function was originally intended for short strings
5874 such as real names; it was not keeping to the 75-character limit for
5875 encoded words that the RFC imposes. It now respects the limit, and
5876 generates multiple encoded words if necessary. To be on the safe side, I
5877 have increased the buffer size for the ${rfc2047: expansion operator from
5880 22. Failure to deliver a bounce message always caused it to be frozen, even if
5881 there was an errors_to setting on the router. The errors_to setting is now
5884 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5885 canonical form (fully expanded) before being placed in
5886 $sender_host_address.
5888 24. Updated eximstats to version 1.33
5890 25. Include certificate and key file names in error message when GnuTLS fails
5891 to set them up, because the GnuTLS error message doesn't include the name
5892 of the failing file when there is a problem reading it.
5894 26. Expand error message when OpenSSL has problems setting up cert/key files.
5897 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5898 (this can affect the format of dates).
5900 28. exim_tidydb, when checking for the continued existence of a message for
5901 which it has found a message-specific retry record, was not finding
5902 messages that were in split spool directories. Consequently, it was
5903 deleting retry records that should have stayed in existence.
5905 29. eximstats updated to version 1.35
5906 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5907 1.35 - bugfix such that pie charts by volume are generated correctly
5909 30. The SPA authentication driver was not abandoning authentication and moving
5910 on to the next authenticator when an expansion was forced to fail,
5911 contradicting the general specification for all authenticators. Instead it
5912 was generating a temporary error. It now behaves as specified.
5914 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5915 (the order specifies the preference for clients). The order is now AES256,
5916 AES128, 3DES, ARCFOUR128.
5918 31. Small patch to Sieve code - explicitly set From: when generating an
5921 32. Exim crashed if a remote delivery caused a very long error message to be
5922 recorded - for instance if somebody sent an entire SpamAssassin report back
5923 as a large number of 550 error lines. This bug was coincidentally fixed by
5924 increasing the size of one of Exim's internal buffers (big_buffer) that
5925 happened as part of the Exiscan merge. However, to be on the safe side, I
5926 have made the code more robust (and fixed the comments that describe what
5929 33. Some experimental protocols are using DNS PTR records for new purposes. The
5930 keys for these records are domain names, not reversed IP addresses. The
5931 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5932 leaves it alone. Component reversal etc. now happens only for IP addresses.
5935 34. The host_aton() function is supposed to be passed a string that is known
5936 to be a valid IP address. However, in the case of IPv6 addresses, it was
5937 not checking this. This is a hostage to fortune. Exim now panics and dies
5938 if the condition is not met. A case was found where this could be provoked
5939 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5940 components; fortuitously, this particular loophole had already been fixed
5941 by change 4.50/55 or 4.44/33 above.
5943 If there are any other similar loopholes, the new check in host_aton()
5944 itself should stop them being exploited. The report I received stated that
5945 data on the command line could provoke the exploit when Exim was running as
5946 exim, but did not say which command line option was involved. All I could
5947 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5948 running as the user.
5951 35. There was a buffer overflow vulnerability in the SPA authentication code
5952 (which came originally from the Samba project). I have added a test to the
5953 spa_base64_to_bits() function which I hope fixes it.
5956 36. The daemon start-up calls getloadavg() while still root for those OS that
5957 need the first call to be done as root, but it missed one case: when
5958 deliver_queue_load_max is set with deliver_drop_privilege. This is
5959 necessary for the benefit of the queue runner, because there is no re-exec
5960 when deliver_drop_privilege is set.
5962 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5963 was in use, was not putting the data itself into the right store pool;
5964 consequently, it could be overwritten for a subsequent message in the same
5965 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5968 38. Sometimes the final signoff response after QUIT could fail to get
5969 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5970 < 0 before doing a fflush(). This bug looks as though it goes back to the
5971 introduction of TLS in release 3.20, but "sometimes" must have been rare
5972 because the tests only now provoked it.
5978 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5979 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5980 Now it returns an integer. A number of calls were still expecting a T/F
5981 return. Fortuitously, in all cases, the tests worked in OK situations,
5982 which is the norm. However, things would have gone wrong on any write
5983 failures on the smtp file descriptor. This function is used when sending
5984 messages over SMTP and also when doing verify callouts.
5986 2. When Exim is called to do synchronous delivery of a locally submitted
5987 message (the -odf or -odi options), it no longer closes stderr before doing
5990 3. Implemented the mua_wrapper option.
5992 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5994 5. Implemented the functions header_remove(), header_testname(),
5995 header_add_at_position(), and receive_remove_recipient(), and exported them
5998 6. If an ACL "warn" statement specified the addition of headers, Exim already
5999 inserted X-ACL-Warn: at the start if there was no header name. However, it
6000 was not making this test for the second and subsequent header lines if
6001 there were newlines in the string. This meant that an invalid header could
6002 be inserted if Exim was badly configured.
6004 7. Allow an ACL "warn" statement to add header lines at the start or after all
6005 the Received: headers, as well as at the end.
6007 8. Added the rcpt_4xx retry error code.
6009 9. Added postmaster_mailfrom=xxx to callout verification option.
6011 10. Added mailfrom=xxxx to the callout verification option, for verify=
6014 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
6015 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
6017 12. Inserted some casts to stop certain compilers warning when using pointer
6018 differences as field lengths or precisions in printf-type calls (mostly
6019 affecting debugging statements).
6021 13. Added optional readline() support for -be (dynamically loaded).
6023 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
6024 same clock tick as a message's arrival, so that its received time was the
6025 same as the "first fail" time on the retry record, and that message
6026 remained on the queue past the ultimate address timeout, every queue runner
6027 would try a delivery (because it was past the ultimate address timeout) but
6028 after another failure, the ultimate address timeout, which should have then
6029 bounced the address, did not kick in. This was a "< instead of <=" error;
6030 in most cases the first failure would have been in the next clock tick
6031 after the received time, and all would be well.
6033 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
6034 being recognized when the domain list was tested by the match_domain
6035 condition in an expansion string.
6037 16. Added the ${str2b64: operator.
6039 17. Exim was always calling setrlimit() to set a large limit for the number of
6040 processes, without checking whether the existing limit was already
6041 adequate. (It did check for the limit on file descriptors.) Furthermore,
6042 errors from getrlimit() and setrlimit() were being ignored. Now they are
6043 logged to the main and panic logs, but Exim does carry on, to try to do its
6044 job under whatever limits there are.
6046 18. Imported PCRE 5.0.
6048 19. Trivial typo in log message " temporarily refused connection" (the leading
6051 20. If the log selector return_path_on_delivery was set and an address was
6052 redirected to /dev/null, the delivery process crashed because it assumed
6053 that a return path would always be set for a "successful" delivery. In this
6054 case, the whole delivery is bypassed as an optimization, and therefore no
6057 21. Internal re-arrangement: the function for sending a challenge and reading
6058 a response while authentication was assuming a zero-terminated challenge
6059 string. It's now changed to take a pointer and a length, to allow for
6060 binary data in such strings.
6062 22. Added the cyrus_sasl authenticator (code supplied by MBM).
6064 23. Exim was not respecting finduser_retries when seeking the login of the
6065 uid under which it was called; it was always trying 10 times. (The default
6066 setting of finduser_retries is zero.) Also, it was sleeping after the final
6067 failure, which is pointless.
6069 24. Implemented tls_on_connect_ports.
6071 25. Implemented acl_smtp_predata.
6073 26. If the domain in control=submission is set empty, Exim assumes that the
6074 authenticated id is a complete email address when it generates From: or
6075 Sender: header lines.
6077 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
6078 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
6079 chown and chgrp in /bin and hostname in /usr/bin.
6081 28. Exim was keeping the "process log" file open after each use, just as it
6082 does for the main log. This opens the possibility of it remaining open for
6083 long periods when the USR1 signal hits a daemon. Occasional processlog
6084 errors were reported, that could have been caused by this. Anyway, it seems
6085 much more sensible not to leave this file open at all, so that is what now
6088 29. The long-running daemon process does not normally write to the log once it
6089 has entered its main loop, and it closes the log before doing so. This is
6090 so that log files can straightforwardly be renamed and moved. However,
6091 there are a couple of unusual error situations where the daemon does write
6092 log entries, and I had neglected to close the log afterwards.
6094 30. The text of an SMTP error response that was received during a remote
6095 delivery was being truncated at 512 bytes. This is too short for some of
6096 the long messages that one sometimes sees. I've increased the limit to
6099 31. It is now possible to make retry rules that apply only when a message has a
6100 specific sender, in particular, an empty sender.
6102 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
6103 it possible to be selective about when SMTP synchronization is enforced.
6105 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
6107 32. Implemented hosts_connection_nolog.
6109 33. Added an ACL for QUIT.
6111 34. Setting "delay_warning=" to disable warnings was not working; it gave a
6114 35. Added mailbox_size and mailbox_filecount to appendfile.
6116 36. Added control = no_multiline_responses to ACLs.
6118 37. There was a bug in the logic of the code that waits for the clock to tick
6119 in the case where the clock went backwards by a substantial amount such
6120 that the microsecond fraction of "now" was more than the microsecond
6121 fraction of "then" (but the whole seconds number was less).
6123 38. Added support for the libradius Radius client library this is found on
6124 FreeBSD (previously only the radiusclient library was supported).
6130 1. When certain lookups returned multiple values in the form name=value, the
6131 quoting of the values was not always being done properly. Specifically:
6132 (a) If the value started with a double quote, but contained no whitespace,
6134 (b) If the value contained whitespace other than a space character (i.e.
6135 tabs or newlines or carriage returns) it was not quoted.
6136 This fix has been applied to the mysql and pgsql lookups by writing a
6137 separate quoting function and calling it from the lookup code. The fix
6138 should probably also be applied to nisplus, ibase and oracle lookups, but
6139 since I cannot test any of those, I have not disturbed their existing code.
6141 2. A hit in the callout cache for a specific address caused a log line with no
6142 reason for rejecting RCPT. Now it says "Previous (cached) callout
6143 verification failure".
6145 3. There was an off-by-one bug in the queryprogram router. An over-long
6146 return line was truncated at 256 instead of 255 characters, thereby
6147 overflowing its buffer with the terminating zero. As well as fixing this, I
6148 have increased the buffer size to 1024 (and made a note to document this).
6150 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
6151 when Exim is waiting for an SMTP response from a remote server, Exim
6152 restarts its select() call on the socket, thereby resetting its timeout.
6153 This is not a problem when such interrupts are rare. Somebody set up a cron
6154 job to run exiwhat every 2 minutes, which is less than the normal select()
6155 timeout (5 or 10 minutes). This meant that the select() timeout never
6156 kicked in because it was always reset. I have fixed this by comparing the
6157 time when an interrupt arrives with the time at the start of the first call
6158 to select(). If more time than the timeout has elapsed, the interrupt is
6159 treated as a timeout.
6161 5. Some internal re-factoring in preparation for the addition of Sieve
6162 extensions (by MH). In particular, the "personal" test is moved to a
6163 separate function, and given an option for scanning Cc: and Bcc: (which is
6164 not set for Exim filters).
6166 6. When Exim created an email address using the login of the caller as the
6167 local part (e.g. when creating a From: or Sender: header line), it was not
6168 quoting the local part when it contained special characters such as @.
6170 7. Installed new OpenBSD configuration files.
6172 8. Reworded some messages for syntax errors in "and" and "or" conditions to
6173 try to make them clearer.
6175 9. Callout options, other than the timeout value, were being ignored when
6176 verifying sender addresses in header lines. For example, when using
6178 verify = header_sender/callout=no_cache
6180 the cache was (incorrectly) being used.
6182 10. Added a missing instance of ${EXE} to the exim_install script; this affects
6183 only the Cygwin environment.
6185 11. When return_path_on_delivery was set as a log selector, if different remote
6186 addresses in the same message used different return paths and parallel
6187 remote delivery occurred, the wrong values would sometimes be logged.
6188 (Whenever a remote delivery process finished, the return path value from
6189 the most recently started remote delivery process was logged.)
6191 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
6192 lines when AUTH and/or TLS are in use. This is the "received protocol"
6193 field. Exim used to use "asmtp" for authenticated SMTP, without any
6194 indication (in the protocol name) for TLS use. Now it follows the RFC and
6195 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
6196 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
6197 names appear in log lines as well as in Received: header lines.
6199 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
6200 extensions, and comparison tests, and to fix some bugs.
6202 14. Changes to the "personal" filter test:
6204 (1) The test was buggy in that it was just doing the equivalent of
6205 "contains" tests on header lines. For example, if a user's address was
6206 anne@some.where, the "personal" test would incorrectly be true for
6208 To: susanne@some.where
6210 This test is now done by extracting each address from the header in turn,
6211 and checking the entire address. Other tests that are part of "personal"
6212 are now done using regular expressions (for example, to check local parts
6213 of addresses in From: header lines).
6215 (2) The list of non-personal local parts in From: addresses has been
6216 extended to include "listserv", "majordomo", "*-request", and "owner-*",
6217 taken from the Sieve specification recommendations.
6219 (3) If the message contains any header line starting with "List-" it is
6220 treated as non-personal.
6222 (4) The test for "circular" in the Subject: header line has been removed
6223 because it now seems ill-conceived.
6225 15. Minor typos in src/EDITME comments corrected.
6227 16. Installed latest exipick from John Jetmore.
6229 17. If headers_add on a router specified a text string that was too long for
6230 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
6231 of string_sprintf() is now avoided.
6233 18. $message_body_size was not set (it was always zero) when running the DATA
6234 ACL and the local_scan() function.
6236 19. For the "mail" command in an Exim filter, no default was being set for
6237 the once_repeat time, causing a random time value to be used if "once" was
6238 specified. (If the value happened to be <= 0, no repeat happened.) The
6239 default is now 0s, meaning "never repeat". The "vacation" command was OK
6240 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
6241 (I found it when inspecting the code).
6243 20. There is now an overall timeout for performing a callout verification. It
6244 defaults to 4 times the callout timeout, which applies to individual SMTP
6245 commands during the callout. The overall timeout applies when there is more
6246 than one host that can be tried. The timeout is checked before trying the
6247 next host. This prevents very long delays if there are a large number of
6248 hosts and all are timing out (e.g. when the network connections are timing
6249 out). The value of the overall timeout can be changed by specifying an
6250 additional sub-option for "callout", called "maxwait". For example:
6252 verify = sender/callout=5s,maxwait=20s
6254 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
6255 to the end before writing, but this should make it even safer).
6257 22. Exim was forgetting that it had advertised PIPELINING for the second and
6258 subsequent messages on an SMTP connection. It was also not resetting its
6259 memory on STARTTLS and an internal HELO.
6261 23. When Exim logs an SMTP synchronization error within a session, it now
6262 records whether PIPELINING has been advertised or not.
6264 24. Added 3 instances of "(long int)" casts to time_t variables that were being
6265 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
6266 rather than long int.
6268 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
6270 26. Added the never_mail option to autoreply.
6276 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
6277 crash if the getsockname() call failed; this can happen if a connection is
6278 closed very soon after it is established. The problem was simply in the
6279 order in which certain operations were done, causing Exim to try to write
6280 to the SMTP stream before it had set up the file descriptor. The bug has
6281 been fixed by making things happen in the correct order.
6287 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
6288 before the connection was closed, thus losing the rejection response.
6290 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
6291 some early Solaris releases, but causes trouble in current releases where
6292 socklen_t is defined.
6294 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
6297 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
6300 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
6301 that releasing the top of it at the end releases what was used for sub-
6302 expansions (unless the block got too big). However, discard this block if
6303 the first thing is a variable or header, so that we can use its block when
6304 it is dynamic (useful for very large $message_headers, for example).
6306 6. Lookups now cache *every* query, not just the most recent. A new, separate
6307 store pool is used for this. It can be recovered when all lookup caches are
6308 flushed. Lookups now release memory at the end of their result strings.
6309 This has involved some general refactoring of the lookup sources.
6311 7. Some code has been added to the store_xxx() functions to reduce the amount
6312 of flapping under certain conditions.
6314 8. log_incoming_interface used to affect only the <= reception log lines. Now
6315 it causes the local interface and port to be added to several more SMTP log
6316 lines, for example "SMTP connection from", and rejection lines.
6318 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
6320 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
6322 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
6323 could be overwritten at the end of the current message (or the start of a
6324 new message if it was set in a HELO ACL). The value is now preserved for
6325 the duration of the SMTP connection.
6327 12. If a transport had a headers_rewrite setting, and a matching header line
6328 contained an unqualified address, that address was qualified, even if it
6329 did not match any rewriting rules. The underlying bug was that the values
6330 of the flags that permit the existence of unqualified sender and recipient
6331 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
6332 non-local messages, and by -bnq for local messages) were not being
6333 preserved with the message after it was received.
6335 13. When Exim was logging an SMTP synchronization error, it could sometimes log
6336 "next input=" as part of the text comprising the host identity instead of
6337 the correct text. The code was using the same buffer for two different
6338 strings. However, depending on which order the printing function evaluated
6339 its arguments, the bug did not always show up. Under Linux, for example, my
6340 test suite worked just fine.
6342 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
6343 doesn't work with some older versions of Perl. It has been changed to "my",
6344 which in any case is probably the better facility to use.
6346 15. A really picky compiler found some instances of statements for creating
6347 error messages that either had too many or two few arguments for the format
6350 16. The size of the buffer for calls to the DNS resolver has been increased
6351 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
6352 for addresses that have a lot of PTR records. This alleviates a problem; it
6353 does not fully solve it.
6355 17. A dnsdb lookup for PTR records that receives more data than will fit in the
6356 buffer now truncates the list and logs the incident, which is the same
6357 action as happens when Exim is looking up a host name and its aliases.
6358 Previously in this situation something unpredictable would happen;
6359 sometimes it was "internal error: store_reset failed".
6361 18. If a server dropped the connection unexpectedly when an Exim client was
6362 using GnuTLS and trying to read a response, the client delivery process
6363 crashed while trying to generate an error log message.
6365 19. If a "warn" verb in an ACL added multiple headers to a message in a single
6366 string, for example:
6368 warn message = H1: something\nH2: something
6370 the text was added as a single header line from Exim's point of view
6371 though it ended up OK in the delivered message. However, searching for the
6372 second and subsequent header lines using $h_h2: did not work. This has been
6373 fixed. Similarly, if a system filter added multiple headers in this way,
6374 the routers could not see them.
6376 20. Expanded the error message when iplsearch is called with an invalid key to
6377 suggest using net-iplsearch in a host list.
6379 21. When running tests using -bh, any delays imposed by "delay" modifiers in
6380 ACLs are no longer actually imposed (and a message to that effect is
6383 22. If a "gecos" field in a passwd entry contained escaped characters, in
6384 particular, if it contained a \" sequence, Exim got it wrong when building
6385 a From: or a Sender: header from that name. A second bug also caused
6386 incorrect handling when an unquoted " was present following a character
6387 that needed quoting.
6389 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
6390 was not being matched caselessly.
6392 24. Arranged for all hyphens in the exim.8 source to be escaped with
6395 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
6396 properly. Recipient callout cache records were still being keyed to include
6397 the sender, even when use_sender was set false. This led to far more
6398 callouts that were necessary. The sender is no longer included in the key
6399 when use_sender is false.
6401 26. Added "control = submission" modifier to ACLs.
6403 27. Added the ${base62d: operator to decode base 62 numbers.
6405 28. dnsdb lookups can now access SRV records.
6407 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
6408 the configuration file.
6410 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
6411 (-v) mode. This makes the output for a verbose queue run more intelligible.
6413 31. Added a use_postmaster feature to recipient callouts.
6415 32. Added the $body_zerocount variable, containing the number of binary zero
6416 bytes in the message body.
6418 33. The time of last modification of the "new" subdirectory is now used as the
6419 "mailbox time last read" when there is a quota error for a maildir
6422 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
6424 35. Added +ignore_unknown as a special item in host lists.
6426 36. Code for decoding IPv6 addresses in host lists is now included, even if
6427 IPv6 support is not being compiled. This fixes a bug in which an IPv6
6428 address was recognized as an IP address, but was then not correctly decoded
6429 into binary, causing unexpected and incorrect effects when compared with
6436 1. Very minor rewording of debugging text in manualroute to say "list of
6437 hosts" instead of "hostlist".
6439 2. If verify=header_syntax was set, and a header line with an unqualified
6440 address (no domain) and a large number of spaces between the end of the
6441 name and the colon was received, the reception process suffered a buffer
6442 overflow, and (when I tested it) crashed. This was caused by some obsolete
6443 code that should have been removed. The fix is to remove it!
6445 3. When running in the test harness, delay a bit after writing a bounce
6446 message to get a bit more predictability in the log output.
6448 4. Added a call to search_tidyup() just before forking a reception process. In
6449 theory, someone could use a lookup in the expansion of smtp_accept_max_
6450 per_host which, without the tidyup, could leave open a database connection.
6452 5. Added the variables $recipient_data and $sender_data which get set from a
6453 lookup success in an ACL "recipients" or "senders" condition, or a router
6454 "senders" option, similar to $domain_data and $local_part_data.
6456 6. Moved the writing of debug_print from before to after the "senders" test
6459 7. Change 4.31/66 (moving the time when the Received: is generated) caused
6460 problems for message scanning, either using a data ACL, or using
6461 local_scan() because the Received: header was not generated till after they
6462 were called (in order to set the time as the time of reception completion).
6463 I have revised the way this works. The header is now generated after the
6464 body is received, but before the ACL or local_scan() are called. After they
6465 are run, the timestamp in the header is updated.
6471 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
6472 before starting a queue runner without re-exec. This happened only when
6473 deliver_drop_privilege was set or when the Exim user was set to root. The
6474 effect of the bug was that timeouts during subsequent deliveries caused
6475 crashes instead of being properly handled. The handler is now left at its
6476 default (and expected) setting.
6478 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
6479 message, again when deliver_drop_privilege is set or Exim is run as root.
6480 The bug described in (1) was not present in this case, but the tidying up
6481 of the other signals was missing. I have made the two cases consistent.
6483 3. The ignore_target_hosts setting on a manualroute router was being ignored
6484 for hosts that were looked up using the /MX notation.
6486 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
6489 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
6490 operated on the sender address. After changing the $sender_address to <>
6491 for the sender address verify, Exim was re-instated it as the original
6492 (before rewriting) address, but remembering that it had rewritten it, so it
6493 wasn't rewriting it again. This bug also had the effect of breaking the
6494 sender address verification caching when the sender address was rewritten.
6496 6. The ignore_target_hosts option was being ignored by the ipliteral router.
6497 This has been changed so that if the ip literal address matches
6498 ignore_target_hosts, the router declines.
6500 7. Added expansion conditions match_domain, match_address, and match_local_
6501 part (NOT match_host).
6503 8. The placeholder for the Received: header didn't have a length field set.
6505 9. Added code to Exim itself and to exim_lock to test for a specific race
6506 condition that could lead to file corruption when using MBX delivery. The
6507 issue is with the lockfile that is created in /tmp. If this file is removed
6508 after a process has opened it but before that process has acquired a lock,
6509 there is the potential for a second process to recreate the file and also
6510 acquire a lock. This could lead to two Exim processes writing to the file
6511 at the same time. The added code performs the same test as UW imapd; it
6512 checks after acquiring the lock that its file descriptor still refers to
6513 the same named file.
6515 10. The buffer for building added header lines was of fixed size, 8192 bytes.
6516 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
6519 11. Added the smtp_active_hostname option. If used, this will typically be made
6520 to depend on the incoming interface address. Because $interface_address is
6521 not set up until the daemon has forked a reception process, error responses
6522 that can happen earlier (such as "too many connections") no longer contain
6525 12. If an expansion in a condition on a "warn" statement fails because a lookup
6526 defers, the "warn" statement is abandoned, and the next ACL statement is
6527 processed. Previously this caused the whole ACL to be aborted.
6529 13. Added the iplsearch lookup type.
6531 14. Added ident_timeout as a log selector.
6533 15. Added tls_certificate_verified as a log selector.
6535 16. Added a global option tls_require_ciphers (compare the smtp transport
6536 option of the same name). This controls incoming TLS connections.
6538 17. I finally figured out how to make tls_require_ciphers do a similar thing
6539 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
6540 before starting the TLS session.
6542 18. Tabs are now shown as \t in -bP output.
6544 19. If the log selector return_path_on_delivery was set, Exim crashed when
6545 bouncing a message because it had too many Received: header lines.
6547 20. If two routers both had headers_remove settings, and the first one included
6548 a superfluous trailing colon, the final name in the first list and the
6549 first name in the second list were incorrectly joined into one item (with a
6550 colon in the middle).
6556 1. Added -C and -D options to the exinext utility, mainly to make it easier
6557 to include in the automated testing, but these could be helpful when
6558 multiple configurations are in use.
6560 2. The exinext utility was not formatting the output nicely when there was
6561 an alternate port involved in the retry record key, nor when there was a
6562 message id as well (for retries that were specific to a specific message
6563 and a specific host). It was also confused by IPv6 addresses, because of
6564 the additional colons they contain. I have fixed the IPv4 problem, and
6565 patched it up to do a reasonable job for IPv6.
6567 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
6568 delivery, the log line now contains "pipelined" if PIPELINING was used.
6570 4. An SMTP transport process used to panic and die if the bind() call to set
6571 an explicit outgoing interface failed. This has been changed; it is now
6572 treated in the same way as a connect() failure.
6574 5. A reference to $sender_host_name in the part of a conditional expansion
6575 that was being skipped was still causing a DNS lookup. This no longer
6578 6. The def: expansion condition was not recognizing references to header lines
6579 that used bh_ and bheader_.
6581 7. Added the _cache feature to named lists.
6583 8. The code for checking quota_filecount in the appendfile transport was
6584 allowing one more file than it should have been.
6586 9. For compatibility with Sendmail, the command line option
6594 and sets the incoming protocol and host name (for trusted callers). The
6595 host name and its colon can be omitted when only the protocol is to be set.
6596 Note the Exim already has two private options, -pd and -ps, that refer to
6597 embedded Perl. It is therefore impossible to set a protocol value of "d" or
6598 "s", but I don't think that's a major issue.
6600 10. A number of refactoring changes to the code, none of which should affect
6603 (a) The number of logging options was getting close to filling up the
6604 32-bit word that was used as a bit map. I have split them into two classes:
6605 those that are passed in the argument to log_write(), and those that are
6606 only ever tested independently outside of that function. These are now in
6607 separate 32-bit words, so there is plenty of room for expansion again.
6608 There is no change in the user interface or the logging behaviour.
6610 (b) When building, for example, log lines, the code previously used a
6611 macro that called string_cat() twice, in order to add two strings. This is
6612 not really sufficiently general. Furthermore, there was one instance where
6613 it was actually wrong because one of the argument was used twice, and in
6614 one call a function was used. (As it happened, calling the function twice
6615 did not affect the overall behaviour.) The macro has been replaced by a
6616 function that can join an arbitrary number of extra strings onto a growing
6619 (c) The code for expansion conditions now uses a table and a binary chop
6620 instead of a serial search (which was left over from when there were very
6621 few conditions). Also, it now recognizes conditions like "pam" even when
6622 the relevant support is not compiled in: a suitably worded error message is
6623 given if an attempt is made to use such a condition.
6625 11. Added ${time_interval:xxxxx}.
6627 12. A bug was causing one of the ddress fields not to be passed back correctly
6628 from remote delivery subprocesses. The field in question was not being
6629 subsequently used, so this caused to problems in practice.
6631 13. Added new log selectors queue_time and deliver_time.
6633 14. Might have fixed a bug in maildirsizefile handling that threw up
6634 "unexpected character" debug warnings, and recalculated the data
6635 unnecessarily. In any case, I expanded the warning message to give more
6638 15. Added the message "Restricted characters in address" to the statements in
6639 the default ACL that block characters like @ and % in local parts.
6641 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
6642 Three changes have been made:
6644 (a) There was a serious bug; a negative response to MAIL caused the whole
6645 recipient domain to be cached as invalid, thereby blocking all messages
6646 to all local parts at the same domain, from all senders. This bug has
6647 been fixed. The domain is no longer cached after a negative response to
6648 MAIL if the sender used is not empty.
6650 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
6653 (c) A new callout option, "use_sender" has been added for people who want
6654 the modified behaviour.
6660 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
6663 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
6664 indeed breaks things for older releases.
6666 3. Added additional logging to the case where there is a problem reading data
6667 from a filter that is running in a subprocess using a pipe, in order to
6668 try to track down a specific problem.
6670 4. Testing facility fudge: when running in the test harness and attempting
6671 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
6672 getting "No route to host". Convert this to a timeout.
6674 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
6677 6. Some OS don't have socklen_t but use size_t instead. This affects the
6678 fifth argument of getsockopt() amongst other things. This is now
6679 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
6680 can be set for individual OS. I have set it for SunOS5, OSF1, and
6681 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
6682 some earlier ones do not.
6684 7. Change 4.30/15 was not doing the test caselessly.
6686 8. The standard form for an IPv6 address literal was being rejected by address
6687 parsing in, for example, MAIL and RCPT commands. An example of this kind of
6688 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6689 this, as well as the form without the "IPv6" on the front (but only when
6690 address literals are enabled, of course).
6692 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
6694 10. Exim crashed if a message with an empty sender address specified by -f
6695 encountered a router with an errors_to setting. This could be provoked only
6696 by a command such as
6700 where an empty string was supplied; "<>" did not hit this bug.
6702 11. Installed PCRE release 4.5.
6704 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
6705 remained set. It is now erased.
6707 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
6708 times from message ids (which are base 36 rather than the normal 62).
6710 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
6711 were being counted as actual protocol errors, and logged if the log
6712 selector +smtp_protocol_error was set. One cannot be perfect in this test,
6713 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
6714 and DATA following a set of rejected RCPTs do not count as protocol errors.
6715 In other words, Exim assumes they were pipelined, though this may not
6716 actually be the case. Of course, in all cases the client gets an
6717 appropriate error code.
6719 15. If a lookup fails in an ACL condition, a message about the failure may
6720 be available; it is used if testing the ACL cannot continue, because most
6721 such messages specify what the cause of the deferral is. However, some
6722 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
6723 that caused an old message to be retained and used if a later statement
6724 caused a defer, replacing the real cause of the deferral.
6726 16. If an IP address had so many PTR records that the DNS lookup buffer
6727 was not large enough to hold them, Exim could crash while trying to process
6728 the truncated data. It now detects and logs this case.
6730 17. Further to 4.21/58, another change has been made: if (and only if) the
6731 first line of a message (the first header line) ends with CRLF, a bare LF
6732 in a subsequent header line has a space inserted after it, so as not to
6733 terminate the header.
6735 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
6736 unnecessarily, used atoi() instead of strtol(), and didn't check the
6737 termination when getting file sizes from file names by regex.
6739 19. Completely re-implemented the support for maildirsize files, in the light
6740 of a number of problems with the previous contributed implementation
6741 (4.30/29). In particular:
6743 . If the quota is zero, the maildirsize file is maintained, but no quota is
6746 . If the maildir directory does not exist, it is created before any attempt
6747 to write a maildirsize file.
6749 . The quota value in the file is just a cache; if the quota is changed in
6750 the transport, the new value overrides.
6752 . A regular expression is available for excluding directories from the
6755 20. The autoreply transport checks the characters in options that define the
6756 message's headers; it allows continued headers, but it was checking with
6757 isspace() after an embedded newline instead of explicitly looking for a
6760 21. If all the "regular" hosts to which an address was routed had passed their
6761 expiry times, and had not reached their retry times, the address was
6762 bounced, even if fallback hosts were defined. Now Exim should go on to try
6765 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
6766 equivalent code in the SMTP transport. Some hosts send humungous responses
6767 to HELO/EHLO, more than 1024 it seems.
6769 23. Refactoring: code in filter.c used (void *) for "any old type" but this
6770 gives compiler warnings in some environments. I've now done it "properly",
6773 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
6774 (because of problems with the built-in one) was declared to return uschar *
6775 instead of char *, causing compiler failure.
6777 25. Fixed a file descriptor leak when processing alias/forward files.
6779 26. Fixed a minor format string issue in dbfn.c.
6781 27. Typo in exim.c: ("dmbnz" for "dbmnz").
6783 28. If a filter file refered to $h_xxx or $message_headers, and the headers
6784 contained RFC 2047 "words", Exim's memory could, under certain conditions,
6787 29. When a sender address is verified, it is cached, to save repeating the test
6788 when there is more than one recipient in a message. However, when the
6789 verification involves a callout, it is possible for different callout
6790 options to be set for different recipients. It is too complicated to keep
6791 track of this in the cache, so now Exim always runs a verification when a
6792 callout is required, relying on the callout cache for the optimization.
6793 The overhead is duplication of the address routing, but this should not be
6796 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
6797 to be verified with callout=postmaster, and the main callout worked but the
6798 postmaster check failed, the verification correctly failed. However, if a
6799 subsequent RCPT command asked for sender verification *without* the
6800 postmaster check, incorrect caching caused this verification also to fail,
6803 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
6804 it was not caching the DNS options (qualify_single, search_parents) that
6805 were used when the lookup failed. A subsequent lookup with different
6806 options therefore always gave the same answer, though there were cases
6807 where it should not have. (Example: a "domains = !$mx_any" option on a
6808 dnslookup router: the "domains" option is always processed without any
6809 widening, but the router might have qualify_single set.) Now Exim uses the
6810 cached value only when the same options are set.
6812 32. Added John Jetmore's "exipick" utility to the distribution.
6814 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
6815 than a timeout (e.g. a certificate is required, and is not provided), an
6816 Exim server now closes the connection immediately. Previously it waited for
6817 the client to close - but if the client is SSL, it seems that they each
6818 wait for each other, leading to a delay before one of them times out.
6820 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
6821 maintained 0.8.x compatibility because I don't think many are using it, and
6822 it is clearly obsolete.
6824 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
6827 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
6828 client certificate was expired. A simple patch fixes this, though I don't
6829 understand the full logic of why the verify callback is called multiple
6832 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
6833 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
6834 which causes problems with some clients (such as the Certicom SSL Plus
6835 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
6836 disables the coutermeasure allowing Eudora to connect."
6838 38. Exim was not checking that a write() to a log file succeeded. This could
6839 lead to Bad Things if a log got too big, in particular if it hit a file
6840 size limit. Exim now panics and dies if it cannot write to a log file, just
6841 as it does if it cannot open a log file.
6843 39. Modified OS/Makefile-Linux so that it now contains
6845 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
6847 The two -D definitions ensure that Exim is compiled with large file
6848 support, which makes it possible to handle log files that are bigger than
6851 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
6852 instance) a domain was checked against a named list that involved a lookup,
6853 causing $domain_data to be set, then another domain was checked against the
6854 same list, then the first domain was re-checked, the value of $domain_data
6855 after the final check could be wrong. In particular, if the second check
6856 failed, it could be set empty. This bug probably also applied to
6859 41. The strip_trailing_dot option was not being applied to the address given
6860 with the -f command-line option.
6862 42. The code for reading a message's header from the spool was incrementing
6863 $received_count, but never initializing it. This meant that the value was
6864 incorrect (doubled) while delivering a message in the same process in which
6865 it was received. In the most common configuration of Exim, this never
6866 happens - a fresh exec is done - but it can happen when
6867 deliver_drop_privilege is set.
6869 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6870 it now includes up to 150 characters of the unexpected data in the log
6873 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6874 and building data strings. The size of both of these buffers was 10 000
6875 bytes - far larger than anybody would *ever* want, thought I. Needless to
6876 say, somebody hit the limit. I have increased the maximum line length to
6877 20 000 and the maximum data length of concatenated lines to 100 000. I have
6878 also fixed two bugs, because there was no checking on these buffers. Tsk,
6879 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6880 buffer is too small.
6882 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6883 lsearch lookups. Now it does.
6885 46. When parsing a route_list item in a manualroute router, a fixed-length
6886 buffer was used for the list of hosts. I made this 1024 bytes long,
6887 thinking that nobody would ever have a list of hosts that long. Wrong.
6888 Somebody had a whole pile of complicated expansion conditions, and the
6889 string was silently truncated, leading to an expansion error. It turns out
6890 that it is easier to change to an unlimited length (owing to other changes
6891 that have happened since this code was originally written) than to build
6892 structure for giving a limitation error. The length of the item that
6893 expands into the list of hosts is now unlimited.
6895 47. The lsearch lookup could not handle data where the length of text line was
6896 more than 4095 characters. Such lines were truncated, leading to shortened
6897 data being returned. It should now handle lines of any length.
6899 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6900 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6903 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6904 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6905 no longer happens in comment lines. A list of replacements is now placed
6906 at the head of all of the source files, except those whose only change is
6907 to replace PERL_COMMAND in the very first #! line.
6909 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6910 messages on the queue, with a bottom-up merge sort, using code contributed
6911 by Michael Haardt. This should make operations like -bp somewhat faster on
6912 large queues. It won't affect queue runners, except when queue_run_in_order
6915 51. Installed eximstats 1.31 in the distribution.
6917 52. Added support for SRV lookups to the dnslookup router.
6919 53. If an ACL referred to $message_body or $message_body_end, the value was not
6920 reset for any messages that followed in the same SMTP session.
6922 54. The store-handling optimization for building very long strings was not
6923 differentiating between the different store pools. I don't think this
6924 actually made any difference in practice, but I've tidied it.
6926 55. While running the routers to verify a sender address, $sender_address
6927 was still set to the sender address. This is wrong, because when routing to
6928 send a bounce to the sender, it would be empty. Therefore, I have changed
6929 it so that, while verifying a sender address, $sender_address is set to <>.
6930 (There is no change to what happens when verifying a recipient address.)
6932 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6933 target A or AAAA records (if not already returned) without resetting the
6934 qualify_single or search_parents options of the DNS resolver. These are
6935 inappropriate in this case because the targets of MX and SRV records must
6936 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6937 target that, when qualified, matched something in the local domain. These
6938 two options are now turned off when doing these lookups.
6940 57. It seems that at least some releases of Reiserfs (which does not have the
6941 concept of a fixed number of inodes) returns zero and not -1 for the
6942 number of available inodes. This interacted badly with check_spool_inodes,
6943 which assumed that -1 was the "no such thing" setting. What I have done is
6944 to check that the total number of inodes is greater than zero before doing
6945 the test of how many are available.
6947 58. When a "warn" ACL statement has a log_message modifier, the message is
6948 remembered, and not repeated. This is to avoid a lot of repetition when a
6949 message has many recipients that cause the same warning to be written.
6950 However, Exim was preserving the list of already written lines for an
6951 entire SMTP session, which doesn't seem right. The memory is now reset if a
6952 new message is started.
6954 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6955 debugging output unless log_rewrite was also set.
6957 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6958 of (int)(handle) when we know that handle contains (void *)(-1).
6960 61. The Exim daemon panic-logs an error return when it closes the incoming
6961 connection. However "connection reset by peer" seems to be common, and
6962 isn't really an error worthy of noting specially, so that particular error
6965 62. When Exim is trying to find all the local interfaces, it used to panic and
6966 die if the ioctl to get the interface flags failed. However, it seems that
6967 on at least one OS (Solaris 9) it is possible to have an interface that is
6968 included in the list of interfaces, but for which you get a failure error
6969 for this call. This happens when the interface is not "plumbed" into a
6970 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6971 failure of the "get flags" call assumes that the interface is down.
6973 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6974 makes life easier for people who are doing arithmetic on fields extracted
6975 from dates, where you often get leading zeros that should not be
6976 interpreted as octal.
6978 64. Added qualify_domain to the redirect router, to override the global
6981 65. If a pathologically long header line contained very many addresses (the
6982 report of this problem mentioned 10 000) and each of them was rewritten,
6983 Exim could use up a very large amount of memory. (It kept on making new
6984 copies of the header line as it rewrote, and never released the old ones.)
6985 At the expense of a bit more processing, the header rewriting function has
6986 been changed so that it no longer eats memory in this way.
6988 66. The generation of the Received: header has been moved from the time that a
6989 message starts to be received, to the time that it finishes. The timestamp
6990 in the Received: header should now be very close to that of the <= log
6991 line. There are two side-effects of this change:
6993 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6994 logged header lines no longer include the local Received: line, because
6995 it has not yet been created. The same applies to a copy of the message
6996 that is returned to a non-SMTP sender when a message is rejected.
6998 (b) When a filter file is tested using -bf, no additional Received: header
6999 is added to the test message. After some thought, I decided that this
7002 This change does not affect the value of $received_for. It is still set
7003 after address rewriting, but before local_scan() is called.
7005 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
7007 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
7008 gave an unhelpful panic error message, and a defer error. I have managed to
7009 change this behaviour so that it now rejects any supplied certificate,
7010 which seems right, as the list of acceptable certificates is empty.
7012 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
7013 gave an unhelpful defer error. I have not managed to make this reject any
7014 supplied certificates, but the error message it gives is "no certificate
7015 supplied", which is not helpful.
7017 70. exigrep's output now also includes lines that are not associated with any
7018 message, but which match the given pattern. Implemented by a patch from
7019 Martin Sluka, which also tidied up the Perl a bit.
7021 71. Recipient callout verification, like sender verification, was using <> in
7022 the MAIL FROM command. This isn't really the right thing, since the actual
7023 sender may affect whether the remote host accepts the recipient or not. I
7024 have changed it to use the actual sender in the callout; this means that
7025 the cache record is now keyed on a recipient/sender pair, not just the
7026 recipient address. There doesn't seem to be a real danger of callout loops,
7027 since a callout by the remote host to check the sender would use <>.
7028 [SEE ABOVE: changed after hitting problems.]
7030 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
7031 temporary errors. However, in the case of such a code being given after
7032 the end of a data transmission (i.e. after ".") Exim was failing to write
7033 a retry record for the message. (Yes, there was some broken host that was
7034 actually sending 8xx at this point.)
7036 73. An unknown lookup type in a host list could cause Exim to panic-die when
7037 the list was checked. (An example that provoked this was putting <; in the
7038 middle of a list instead of at the start.) If this happened during a DATA
7039 ACL check, a -D file could be left lying around. This kind of configuration
7040 error no longer causes Exim to die; instead it causes a defer error. The
7041 incident is still logged to the main and panic logs.
7043 74. Buglet left over from Exim 3 conversion. The message "too many messages
7044 in one connection" was written to the rejectlog but not the mainlog, except
7045 when address rewriting (yes!) was being logged.
7047 75. Added write_rejectlog option.
7049 76. When a system filter was run not as root (that is, when system_filter_user
7050 was set), the values of the $n variables were not being returned to the
7051 main process; thus, they were not subsequently available in the $sn
7054 77. Added +return_path_on_delivery log selector.
7056 78. A connection timeout was being treated differently from recipients deferred
7057 when testing hosts_max_try with a message that was older than the host's
7058 retry timeout. (The host should not be counted, thus allowing all hosts to
7059 be tried at least once before bouncing.) This may have been the cause of an
7060 occasionally reported bug whereby a message would remain on the queue
7061 longer than the retry timeout, but would be bounced if a delivery was
7062 forced. I say "may" because I never totally pinned down the problem;
7063 setting up timeout/retry tests is difficult. See also the next item.
7065 79. The ultimate address timeout was not being applied to errors that involved
7066 a combination of host plus message (for example, a timeout on a MAIL
7067 command). When an address resolved to a number of possible hosts, and they
7068 were not all tried for each delivery (e.g. because of hosts_max_try), a
7069 message could remain on the queue longer than the retry timeout.
7071 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
7074 81. Fixed an obscure SMTP outgoing bug which required at least the following
7075 conditions: (a) there was another message waiting for the same server;
7076 (b) the server returned 5xx to all RCPT commands in the first message so
7077 that the message was not completed; (c) the server dropped the connection
7078 or gave a negative response to the RSET that Exim sends to abort the
7079 transaction. The observed case was a dropped connection after DATA that had
7080 been sent in pipelining mode. That is, the server had advertised PIPELINING
7081 but was not implementing it correctly. The effect of the bug was incorrect
7082 behaviour, such as trying another host, and this could lead to a crash.
7088 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
7089 and daemon.c were passed as pointers to ints; they should have been
7090 pointers to socklen_t variables (which are typically unsigned ints).
7092 2. Some signed/unsigned type warnings in the os.c file for Linux have been
7095 3. Fixed a really odd bug that affected only the testing scheme; patching a
7096 certain fixed string in the binary changed the value of another string that
7097 happened to be identical to the end of the original first string.
7099 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
7100 name", it returns that address as the IP address. On some operating
7101 systems (e.g. Solaris), it also passes back the IP address string as the
7102 "host name". However, on others (e.g. Linux), it passes back an empty
7103 string. Exim wasn't checking for this, and was changing the host name to an
7104 empty string, assuming it had been canonicalized.
7106 5. Although rare, it is permitted to have more than one PTR record for a given
7107 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
7108 all the names associated with an address, because they do in Solaris.
7109 However, it seems that they do not in Linux for data that comes from the
7110 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
7111 I found this out when I moved to a new Linux workstation and tried to run
7112 the Exim test suite.
7114 To get round this problem I have changed the code so that it now does its
7115 own call to the DNS to look up PTR records when searching for a host name.
7116 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
7117 addresses that are only in /etc/hosts are still found.
7119 This behaviour is, however, controlled by an option called host_lookup_
7120 order, which defaults to "bydns:byaddr". If people want to use the other
7121 order, or indeed, just use one or the other means of lookup, they can
7122 specify it in this variable.
7124 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
7125 some operating systems, this comes back from gethostbyaddr() as an empty
7126 string, and this is what Exim used to test for. However, it seems that in
7127 other systems, "." is yielded. Exim now tests for this case too.
7129 7. The values of check_spool_space and check_log_space are now held internally
7130 as a number of kilobytes instead of an absolute number of bytes. If a
7131 numbers is specified without 'K' or 'M', it is rounded up to the nearest
7132 kilobyte. This means that much larger values can be stored.
7134 8. Exim monitor: an attempt to get the action menu when not actually pointing
7135 at a message produces an empty menu entitled "No message selected". This
7136 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
7137 no entries in it ("Shell widget menu has zero width and/or height"). So I
7138 have added a single, blank menu entry in this case.
7140 9. Added ${quote_local_part.
7142 10. MIME decoding is now applied to the contents of Subject: header lines when
7145 11. Now that a reference to $sender_host_address automatically causes a reverse
7146 lookup to occur if necessary (4.13/18), there is no need to arrange for a
7147 host lookup before query-style lookups in lists that might use this
7148 variable. This has therefore been abolished, and the "net-" prefix is no
7149 longer necessary for query-style lookups.
7151 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
7152 have been a typo for LFLAGS, so it has been changed.
7154 13. The install script calls Exim with "-C /dev/null" in order to find the
7155 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
7156 to be output. However, since Exim outputs its version number before the
7157 error, it didn't break the script. It just looked ugly. I fixed this by
7158 always allowing "-C /dev/null" if the caller is root.
7160 14. Ignore overlarge ACL variable number when reading spool file - insurance
7161 against a later release with more variables having written the file.
7163 15. The standard form for an IPv6 address literal was being rejected by EHLO.
7164 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
7165 this, as well as the form without the "IPv6" on the front.
7167 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
7168 OS/Makefile-Darwin file.
7170 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
7171 only to LDAP libraries that do not have LDAP_OPT_DEREF.
7173 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
7174 variable. However, some OS use ints for inodes. I've added cast to long int
7175 to get rid of the compiler warning.
7177 19. I had forgotten to lock out "/../" in configuration file names when
7178 ALT_CONFIG_PREFIX was set.
7180 20. Routers used for verification do not need to specify transports. However,
7181 if such a router generated a host list, and callout was configured, Exim
7182 crashed, because it could not find a port number from the (non-existent)
7183 transport. It now assumes port 25 in this circumstance.
7185 21. Added the -t option to exigrep.
7187 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
7188 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
7189 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
7190 them is set, LOOKUP_LSEARCH is forced.
7192 23. "exim -bV" now outputs a list of lookups that are included in the binary.
7194 24. Added sender and host information to the "rejected by local_scan()" log
7195 line; previously there was no indication of these.
7197 25. Added .include_if_exists.
7199 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
7200 Linux. It turns out that not all file systems support this. Apparently some
7201 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
7202 do it.) To cope with this, the error EINVAL, which means that sync-ing is
7203 not supported on the file descriptor, is now ignored when Exim is trying to
7204 sync a directory. This applies only to Linux.
7206 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
7208 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
7209 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
7212 29. Added support for maildirsize files from supplied patch (modified a bit).
7214 30. The use of :fail: followed by an empty string could lead Exim to respond to
7215 sender verification failures with (e.g.):
7217 550 Verification failed for <xxx>
7218 550 Sender verify failed
7220 where the first response line was missing the '-' that indicates it is not
7221 the final line of the response.
7223 31. The loop for finding the name of the user that called Exim had a hardwired
7224 limit of 10; it now uses the value of finduser_retries, which is used for
7225 all other user lookups.
7227 32. Added $received_count variable, available in data and not_smtp ACLs, and at
7230 33. Exim was neglecting to zero errno before one call of strtol() when
7231 expanding a string and expecting an integer value. On some systems this
7232 resulted in spurious "integer overflow" errors. Also, it was casting the
7233 result into an int without checking.
7235 34. Testing for a connection timeout using "timeout_connect" in the retry rules
7236 did not work. The code looks as if it has *never* worked, though it appears
7237 to have been documented since at least release 1.62. I have made it work.
7239 35. The "timeout_DNS" error in retry rules, also documented since at least
7240 1.62, also never worked. As it isn't clear exactly what this means, and
7241 clearly it isn't a major issue, I have abolished the feature by treating it
7242 as "timeout", and writing a warning to the main and panic logs.
7244 36. The display of retry rules for -brt wasn't always showing the error code
7247 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
7248 timeout_connect_A, timeout_connect_MX.
7250 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
7251 to the empty sender.
7253 39. The daemon was not analysing the content of -oX till after it had closed
7254 stderr and disconnected from the controlling terminal. This meant that any
7255 syntax errors were only noted on the panic log, and the return code from
7256 the command was 0. By re-arranging the code a little, I've made the
7257 decoding happen first, so such errors now appear on stderr, and the return
7258 code is 1. However, the actual setting up of the sockets still happens in
7259 the disconnected process, so errors there are still only recorded on the
7262 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
7263 connections (as happens on some IP stacks) was logged at start up time as
7264 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
7265 it from "IPv6 and IPv4", which means that two separate sockets are being
7268 41. The debug output for gethostbyname2() or getipnodebyname() failures now
7269 says whether AF_INET or AF_INET6 was passed as an argument.
7271 42. Exiwhat output was messed up when time zones were included in log
7274 43. Exiwhat now gives more information about the daemon's listening ports,
7275 and whether -tls-on-connect was used.
7277 44. The "port" option of the smtp transport is now expanded.
7279 45. A "message" modifier in a "warn" statement in a non-message ACL was being
7280 silently ignored. Now an error message is written to the main and panic
7283 46. There's a new ACL modifier called "logwrite" which writes to a log file
7284 as soon as it is encountered.
7286 47. Added $local_user_uid and $local_user_gid at routing time.
7288 48. Exim crashed when trying to verify a sender address that was being
7291 49. Exim was recognizing only a space character after ".include". It now also
7292 recognizes a tab character.
7294 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
7295 extracting the relevant information from the specification. The man page no
7296 longer contains scrambled data for the -d option, and I've added a section
7297 at the front about calling Exim under different names.
7299 51. Added "extra_headers" argument to the "mail" command in filter files.
7301 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
7304 53. Installed eximstats 1.29.
7306 54. Added transport_filter_timeout as a generic transport option.
7308 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
7309 Cc: header lines. This was required by RFC 822, but it not required by RFC
7312 56. Exim used to add From:, Date:, and Message-Id: header lines to any
7313 incoming messages that did not have them. Now it does so only if the
7314 message originates locally, that is, if there is no associated remote host
7315 address. When Resent- header lines are present, this applies to the Resent-
7316 lines rather than the non-Resent- lines.
7318 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
7319 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
7321 58. Messages for configuration errors now include the name of the main
7322 configuration file - useful now that there may be more than one file in a
7323 list (.included file names were always shown).
7325 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
7326 for those rare installations that do not start the daemon as root or run it
7327 setuid root. I've cut out the call to initgroups() if the daemon is not
7330 60. The Exim user and group can now be bound into the binary as text strings
7331 that are looked up at the start of Exim's processing.
7333 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
7335 62. Added $mailstore_basename variable.
7337 63. Installed patch to sieve.c from Michael Haardt.
7339 64. When Exim failed to open the panic log after failing to open the main log,
7340 the original message it was trying to log was written to stderr and debug
7341 output, but if they were not available (the usual case in production), it
7342 was lost. Now it is written to syslog before the two lines that record the
7343 failures to open the logs.
7345 65. Users' Exim filters run in subprocesses under the user's uid. It is
7346 possible for a "deliver" command or an alias in a "personal" command to
7347 provoke an address rewrite. If logging of address rewriting is configured,
7348 this fails because the process is not running as root or exim. There may be
7349 a better way of dealing with this, but for the moment (because 4.30 needs
7350 to be released), I have disabled address rewrite logging when running a
7351 filter in a non-root, non-exim process.
7357 1. The buildconfig auxiliary program wasn't quoting the value set for
7358 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
7359 not defined. This bug was masked in 4.22 by the effect that was fixed in
7362 2. Some messages that were rejected after a message id was allocated were
7363 shown as "incomplete" by exigrep. It no longer does this for messages that
7364 are rejected by local_scan() or the DATA or non-SMTP ACLs.
7366 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
7367 have allow_domain_literals set, the ID did not get logged in the <= line.
7368 Domain literals are now always recognized in Message-ID: header lines.
7370 4. The first argument for a ${extract expansion item is the key name or field
7371 number. Leading and trailing spaces in this item were not being ignored,
7372 causing some misleading effects.
7374 5. When deliver_drop_privilege was set, single queue runner processes started
7375 manually (i.e. by the command "exim -q") or by the daemon (which uses the
7376 same command in the process it spins off) were not dropping privilege.
7378 6. When the daemon running as "exim" started a queue runner, it always
7379 re-executed Exim in the spun-off process. This is a waste of effort when
7380 deliver_drop_privilege is set. The new process now just calls the
7381 queue-runner function directly.
7387 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
7390 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
7391 ignored. Though the use of -oP was forcing the writing of a pid file, it
7392 was always written to the default place.
7394 3. If the message "no IP address found for host xxxx" is generated during
7395 incoming verification, it is now followed by identification of the incoming
7396 connection (so you can more easily find what provoked it).
7398 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
7400 5. Added some features to "harden" Exim a bit more against certain attacks:
7402 (a) There is now a build-time option called FIXED_NEVER_USERS that can
7403 be put in Local/Makefile. This is like the never_users runtime option,
7404 but it cannot be overridden. The default setting is "root".
7406 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
7407 prefix string with which any file named in a -C command line option
7410 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
7411 is retained for -C and -D only if the caller of Exim is root. Without
7412 it, the exim user may also use -C and -D and retain privilege.
7414 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
7415 command line option is disabled.
7417 6. Macro names set by the -D option must start with an upper case letter, just
7418 like macro names defined in the configuration file.
7420 7. Added "dereference=" facility to LDAP.
7422 8. Two instances of the typo "uknown" in the source files are fixed.
7424 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
7425 the Configure-Makefile script screwed up while processing it.
7427 10. Incorporated PCRE 4.4.
7429 11. The SMTP synchronization check was not operating right at the start of an
7430 SMTP session. For example, it could not catch a HELO sent before the client
7431 waited for the greeting. There is now a check for outstanding input at the
7432 point when the greeting is written. Because of the duplex, asynchronous
7433 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
7434 way, but not yet received, when the check is performed.
7436 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
7437 on TCP/IP sockets, because this apparently causes some broken clients to
7440 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
7441 unchanged) from the Cygwin maintainer.
7443 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
7444 was supported instead of testing for mbx. Effectively a typo.
7446 15. The spa authenticator server code was not checking that the input it
7447 received was valid base64.
7449 16. The debug output line for the "set" modifier in ACLs was not showing the
7450 name of the variable that was being set.
7452 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
7454 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
7455 lookup on its own. Something else was needed to trigger the lookup. For
7456 example, a match in host_lookup or the need for a host name in a host list.
7457 Now, if $sender_host_name is referenced and the host name has not yet been
7458 looked up, a lookup is performed. If the lookup fails, the variable remains
7459 empty, and $host_lookup_failed is set to "1".
7461 19. Added "eqi" as a case-independent comparison operator.
7463 20. The saslauthd authentication condition could segfault if neither service
7464 nor realm was specified.
7466 21. If an overflowing value such as "2048M" was set for message_size_limit, the
7467 error message that was logged was misleading, and incoming SMTP
7468 connections were dropped. The message is now more accurate, and temporary
7469 errors are given to SMTP connections.
7471 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
7472 (except RSET) with a 421 error, until QUIT is received. However, it was
7473 failing to send a response to QUIT.
7475 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
7476 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
7477 ACL is now run after the helo_try_verify_hosts code.
7479 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
7480 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
7481 case-independent, so other case variants are also recognized). Apparently
7482 some systems use these upper case variants.
7484 25. If more than two messages were waiting for the same host, and a transport
7485 filter was specified for the transport, Exim sent two messages over the
7486 same TCP/IP connection, and then failed with "socket operation on non-
7487 socket" when it tried to send the third.
7489 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
7491 27. The extern definition of crypt16() in expand.c was not being excluded when
7492 the OS had its own crypt16() function.
7494 28. Added bounce_return_body as a new option, and bounce_return_size_limit
7495 as a preferred synonym for return_size_limit, both as an option and as an
7498 29. Added LIBS=-liconv to OS/Makefile-OSF1.
7500 30. Changed the default configuration ACL to relax the local part checking rule
7501 for addresses that are not in any local domains. For these addresses,
7502 slashes and pipe symbols are allowed within local parts, but the sequence
7503 /../ is explicitly forbidden.
7505 31. SPA server authentication was not clearing the challenge buffer before
7508 32. log_message in a "warn" ACL statement was writing to the reject log as
7509 well as to the main log, which contradicts the documentation and doesn't
7510 seem right (because no rejection is happening). So I have stopped it.
7512 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
7513 However, I am unable to do any testing of this.
7515 34. Fixed an infelicity in the appendfile transport. When checking directories
7516 for a mailbox, to see if any needed to be created, it was accidentally
7517 using path names with one or more superfluous leading slashes; tracing
7518 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
7520 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
7521 contents are added to the log line that is written for every discarded
7522 recipient. (Previously a log_message setting was ignored.)
7524 36. The ${quote: operator now quotes the string if it is empty.
7526 37. The install script runs exim in order to find its version number. If for
7527 some reason other than non-existence or emptiness, which it checks, it
7528 could not run './exim', it was installing it with an empty version number,
7529 i.e. as "exim-". This error state is now caught, and the installation is
7532 38. An argument was missing from the function that creates an error message
7533 when Exim fails to connect to the socket for saslauthd authentication.
7534 This could cause Exim to crash, or give a corrupted message.
7536 39. Added isip, isip4, and isip6 to ${if conditions.
7538 40. The ACL variables $acl_xx are now saved with the message, and can be
7539 accessed later in routers, transports, and filters.
7541 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
7542 strings in the file are not string-expanded.
7544 42. If a MAIL command specified a SIZE value that was too large to fit into an
7545 int variable, the check against message_size_limit failed. Such values are
7546 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
7547 day this will have to be increased, but I don't think I want to be around
7548 when emails are that large.
7555 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
7556 iconv() is not standard in FreeBSD.
7558 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
7559 IPv6 enabled. The observed symptom was a segmentation fault on return from
7560 the function os_common_find_running_interfaces() in src/os.c.
7562 3. In the check_special_case() function in daemon.c I had used "errno" as an
7563 argument name, which causes warnings on some systems. This was basically a
7564 typo, since it was named "eno" in the comments!
7566 4. The code that waits for the clock to tick (at a resolution of some fraction
7567 of a second) so as to ensure message-id uniqueness was always waiting for
7568 at least one whole tick, when it could have waited for less. [This is
7569 almost certainly not relevant at current processor speeds, where it is
7570 unlikely to ever wait at all. But we try to future-proof.]
7572 5. The function that sleeps for a time interval that includes fractions of a
7573 second contained a race. It did not block SIGALRM between setting the
7574 timer, and suspending (a couple of lines later). If the interval was short
7575 and the sigsuspend() was delayed until after it had expired, the suspension
7576 never ended. On busy systems this could lead to processes getting stuck for
7579 6. Some uncommon configurations may cause a lookup to happen in a queue runner
7580 process, before it forks any delivery processes. The open lookup caching
7581 mechanism meant that the open file or database connection was passed into
7582 the delivery process. The problem was that delivery processes always tidy
7583 up cached lookup data. This could cause a problem for the next delivery
7584 process started by the queue runner, because the external queue runner
7585 process does not know about the closure. So the next delivery process
7586 still has data in the lookup cache. In the case of a file lookup, there was
7587 no problem because closing a file descriptor in a subprocess doesn't affect
7588 the parent. However, if the lookup was caching a connection to a database,
7589 the connection was closed, and the second delivery process was likely to
7590 see errors such as "PGSQL: query failed: server closed the connection
7591 unexpectedly". The problem has been fixed by closing all cached lookups
7592 in a queue runner before running a delivery process.
7594 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
7595 seem to have the "const" qualifier which it has on other OS. I've
7598 8. Change 4.21/2 was too strict. It is only if there are two authenticators
7599 *of the same type* (client or server) with the same public name that an
7600 error should be diagnosed.
7602 9. When Exim looked up a host name for an IP address, but failed to find the
7603 original IP address when looking up the host name (a safety check), it
7604 output the message "<ip address> does not match any IP for NULL", which was
7605 confusing, to say the least. The bug was that the host name should have
7606 appeared instead of "NULL".
7608 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
7609 user that is built into the binary, and the -C or -D options is used, root
7610 privilege is dropped before the configuration file is read. In addition,
7611 logging is switched to stderr instead of the normal log files. If the
7612 configuration then re-defines the Exim user, the unprivileged environment
7613 is probably not what is expected, so Exim logs a panic warning message (but
7616 However, if deliver_drop_privilege is set, the unprivileged state may well
7617 be exactly what is intended, so the warning has been cut out in that case,
7618 and Exim is allowed to try to write to its normal log files.
7624 1. smtp_return_error_details was not giving details for temporary sender
7625 or receiver verification errors.
7627 2. Diagnose a configuration error if two authenticators have the same public
7630 3. Exim used not to create the message log file for a message until the first
7631 delivery attempt. This could be confusing when incoming messages were held
7632 for policy or load reasons. The message log file is now created at the time
7633 the message is received, and an initial "Received" line is written to it.
7635 4. The automatically generated man page for command line options had a minor
7636 bug that caused no ill effects; however, a more serious problem was that
7637 the procedure for building the man page automatically didn't always
7638 operate. Consequently, release 4.20 contains an out-of-date version. This
7639 shouldn't happen again.
7641 5. When building Exim with embedded Perl support, the script that builds the
7642 Makefile was calling 'perl' to find its compile-time parameters, ignoring
7643 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
7645 6. The freeze_tell option was not being used for messages that were frozen on
7646 arrival, either by an ACL or by local_scan().
7648 7. Added the smtp_incomplete_transaction log selector.
7650 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
7651 was accepting AUTH without a new EHLO.
7653 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
7654 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
7657 10. Logging of TCP/IP connections (when configured) now happens in the main
7658 daemon process instead of the child process, so that the TCP/IP connection
7659 count is more accurate (but it can never be perfect).
7661 11. The use of "drop" in a nested ACL was not being handled correctly in the
7662 outer ACL. Now, if condition failure induced by the nested "drop" causes
7663 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
7664 or "require"), the connection is dropped.
7666 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
7667 that yield "discard" can now be used with an "accept" or a "discard" verb,
7668 but an error is generated for any others (because I can't see a useful way
7669 to define what should happen).
7671 13. When an ACL is read dynamically from a file (or anywhere else), the lines
7672 are now processed in the same way as lines in the Exim configuration file.
7673 In particular, continuation lines are supported.
7675 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
7677 15. Added -ti meaning -t -i.
7679 16. Check for letters, digits, hyphens, and dots in the names of dnslist
7680 domains, and warn by logging if others are found.
7682 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
7683 returned from GIFCONF when Exim is trying to find the list of interfaces on
7684 a host. The code in os.c has been modified to copy each ifreq to an aligned
7685 structure in all cases.
7687 Also, in some cases, the returned ifreq's were being copied to a 'struct
7688 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
7689 means the last couple of bytes of an IPv6 address could be chopped if the
7690 ifreq contained only a normal sockaddr (14 bytes storage).
7692 18. Named domain lists were not supported in the hosts_treat_as_local option.
7693 An entry such as +xxxx was not recognized, and was treated as a literal
7696 19. Ensure that header lines added by a DATA ACL are included in the reject log
7697 if the ACL subsequently rejects the message.
7699 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
7700 MD5 (which is deprecated).
7702 21. When testing a filter file using -bf, Exim was writing a message when it
7703 took the sender from a "From " line in the message, but it was not doing so
7704 when it took $return_path from a Return-Path: header line. It now does.
7706 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
7707 with a valid header line field name (a series of printing characters
7708 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
7710 23. Changed "disc" in the source to "disk" to conform to the documentation and
7711 the book and for uniformity.
7713 24. Ignore Sendmail's -Ooption=value command line item.
7715 25. When execve() failed while trying to run a command in a pipe transport,
7716 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
7717 could be confused with a return value of 69 from the command itself. This
7718 has been changed to 127, the value the shell returns if it is asked to run
7719 a non-existent command. The wording for the related log line suggests a
7720 non-existent command as the problem.
7722 26. If received_header_text expands to an empty string, do not add a Received:
7723 header line to the message. (Well, it adds a token one on the spool, but
7724 marks it "old" so that it doesn't get used or transmitted.)
7726 27. Installed eximstats 1.28 (addition of -nt option).
7728 28. There was no check for failure on the call to getsockname() in the daemon
7729 code. This can fail if there is a shortage of resources on the system, with
7730 ENOMEM, for example. A temporary error is now given on failure.
7732 29. Contrary to the C standard, it seems that in some environments, the
7733 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
7734 program. Exim now does this explicitly; it affects the formatting of
7735 timestamps using strftime().
7737 30. If exiqsumm was given junk data, it threw up some uninitialized variable
7738 complaints. I've now initialized all the variables, to avoid this.
7740 32. Header lines added by a system filter were not being "seen" during
7741 transport-time rewrites.
7743 33. The info_callback() function passed to OpenSSL is set up with type void
7744 (*)(SSL *, int, int), as described somewhere. However, when calling the
7745 function (actually a macro) that sets it up, the type void(*)() is
7746 expected. I've put in a cast to prevent warnings from picky compilers.
7748 34. If a DNS black list lookup found a CNAME record, but there were no A
7749 records associated with the domain it pointed at, Exim crashed.
7751 35. If a DNS black list lookup returned more than one A record, Exim ignored
7752 all but the first. It now scans all returned addresses if a particular IP
7753 value is being sought. In this situation, the contents of the
7754 $dnslist_value variable are a list of all the addresses, separated by a
7757 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
7758 to accept a host name and all its aliases if the forward lookup for any of
7759 them yielded the IP address of the incoming connection. Now it accepts only
7760 those names whose forward lookup yields the correct IP address. Any other
7761 names are discarded. This closes a loophole whereby a rogue DNS
7762 administrator could create reverse DNS records to break through a
7763 wildcarded host restriction in an ACL.
7765 37. If a user filter or a system filter that ran in a subprocess used any of
7766 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
7767 the wrong values were passed to the pipe command ($thisaddress had the
7768 value of $0, $0 had the value of $1, etc). This bug was introduced by
7769 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
7771 38. Improved the line breaking for long SMTP error messages from ACLs.
7772 Previously, if there was no break point between 40 and 75 characters, Exim
7773 left the rest of the message alone. Two changes have been made: (a) I've
7774 reduced the minimum length to 35 characters; (b) if it can't find a break
7775 point between 35 and 75 characters, it looks ahead and uses the first one
7776 that it finds. This may give the occasional overlong line, but at least the
7777 remaining text gets split now.
7779 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
7780 file descriptors might be low, and that setting 1000 would always raise it.
7781 It turns out that in some environments, the limit is already over 1000 and
7782 that lowering it causes trouble. So now Exim takes care not to decrease it.
7784 40. When delivering a message, the value of $return_path is set to $sender_
7785 address at the start of routing (routers may change the value). By an
7786 oversight, this default was not being set up when an address was tested by
7787 -bt or -bv, which affected the outcome if any router or filter referred to
7790 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
7791 log only when "message" or "log_message" are set. However, if one of the
7792 conditions was an address verification, or a call to a nested ACL, the
7793 messages generated by the underlying test were being passed through. This
7794 no longer happens. The underlying message is available in $acl_verify_
7795 message for both "message" and "log_message" expansions, so it can be
7796 passed through if needed.
7798 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
7799 new expansion $bh_ to give the encoded byte string without charset
7800 translation. Translation happens only if iconv() is available; HAVE_ICONV
7801 indicates this at build time. HEADERS_CHARSET gives the charset to
7802 translate to; headers_charset can change it in the configuration, and
7803 "headers charset" can change it in an individual filter file.
7805 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
7806 that creates RFC 2047 encoded "words" labels them as that charset instead
7807 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
7808 expansion operator; (ii) when Exim creates a From: line for a local
7809 message; (iii) when a header line is rewritten to include a "phrase" part.
7811 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
7812 buggy, causing it to skip the first lines of messages whose message ID
7813 ended in 'D'. This would not have bitten before Exim release 4.14, because
7814 message IDs were unlikely to end in 'D' before then. The effect was to have
7815 incorrect size information for certain domains.
7817 45. #include "config.h" was missing at the start of the crypt16.c module. This
7818 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
7821 46. If there was a timeout during a "random" callout check, Exim treated it as
7822 a failure of the random address, and carried on sending RSET and the real
7823 address. If the delay was just some slowness somewhere, the response to the
7824 original RCPT would be taken as a response to RSET and so on, causing
7825 mayhem of various kinds.
7827 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
7828 when I implemented it. It didn't allow for the fact that some option values
7829 may legitimately be negative (e.g. size_addition), and it didn't even do
7830 the right test for positive values.
7832 48. Domain names in DNS records are case-independent. Exim always looks them up
7833 in lower case. Some resolvers return domain names in exactly the case they
7834 appear in the zone file, that is, they may contain uppercase letters. Not
7835 all resolvers do this - some return always lower case. Exim was treating a
7836 change of case by a resolver as a change of domain, similar to a widening
7837 of a domain abbreviation. This triggered its re-routing code and so it was
7838 trying to route what was effectively the same domain again. This normally
7839 caused routing to fail (because the router wouldn't handle the domain
7840 twice). Now Exim checks for this case specially, and just changes the
7841 casing of the domain that it ultimately uses when it transmits the message
7844 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
7847 50. If a filter generated a file delivery with a non-absolute name (possible if
7848 no home directory exists for the router), the forbid_file option was not
7851 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
7852 the existing equality matching.
7854 52. Exim was using ints instead of ino_t variables in some places where it was
7855 dealing with inode numbers.
7857 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
7858 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
7859 called TMPDIR, and if it finds it is different, it changes its value.
7861 54. The smtp_printf() function is now made available to local_scan() so
7862 additional output lines can be written before returning. There is also an
7863 smtp_fflush() function to enable the detection of a dropped connection.
7864 The variables smtp_input and smtp_batched_input are exported to
7867 55. Changed the default runtime configuration: the message "Unknown user"
7868 has been removed from the ACL, and instead placed on the localuser router,
7869 using the cannot_route_message feature. This means that any verification
7870 failures that generate their own messages won't get overridden. Similarly,
7871 the "Unrouteable address" message that was in the ACL for unverifiable
7872 relay addresses has also been removed.
7874 56. Added hosts_avoid_esmtp to the smtp transport.
7876 57. The exicyclog script was not checking for the esoteric option
7877 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7878 will work only if exicyclog is run under the appropriate euid.
7880 58. Following a discussion on the list, the rules by which Exim recognises line
7881 endings on incoming messages have been changed. The -dropcr and drop_cr
7882 options are now no-ops, retained only for backwards compatibility. The
7883 following line terminators are recognized: LF CRLF CR. However, special
7884 processing applies to CR:
7886 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7887 nor a local message in the state where . is a terminator.
7889 (ii) If a bare CR is encountered in a header line, an extra space is added
7890 after the line terminator so as not to end the header. The reasoning
7891 behind this is that bare CRs in header lines are most likely either
7892 to be mistakes, or people trying to play silly games.
7894 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7895 was being incorrectly given as 18 bytes larger than it should have been.
7896 This is a VOB (very old bug).
7898 60. This may never have affected anything current, but just in case it has:
7899 When the local host is found other than at the start of a list of hosts,
7900 the local host, those with the same MX, and any that follow, are discarded.
7901 When the list in question was part of a longer list of hosts, the following
7902 hosts (not currently being processed) were also being discarded. This no
7903 longer happens. I'm not sure if this situation could ever has previously
7906 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7909 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7910 header. This is something that is recommended in a new Internet Draft, and
7911 is something that is documented as being done by Sendmail. There are two
7912 possible values. For messages generated by the autoreply transport, Exim
7915 Auto-Submitted: auto-replied
7917 whereas for all other generated messages (e.g. bounces) it adds
7919 Auto-Submitted: auto-generated
7921 63. The "personal" condition in filters now includes a test for the
7922 Auto-Submitted: header. If it contains the string "auto-" the message it
7923 not considered personal.
7925 64. Added rcpt_include_affixes as a generic transport option.
7927 65. Added queue_only_override (default true).
7929 66. Added the syslog_duplication option.
7931 67. If what should have been the first header line of a message consisted of
7932 a space followed by a colon, Exim was mis-interpreting it as a header line.
7933 It isn't of course - it is syntactically invalid and should therefore be
7934 treated as the start of the message body. The misbehaviour could have
7935 caused a number of strange effects, including loss of data in subsequent
7936 header lines, and spool format errors.
7938 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7939 client host had authenticated. This control can now be exercised by an ACL
7940 for more flexibility.
7942 69. By default, callouts do not happen when testing with -bh. There is now a
7943 variant, -bhc, which does actually run the callout code, including
7944 consulting and updating the callout cache.
7946 70. Added support for saslauthd authentication, courtesy of Alexander
7949 71. If statvfs() failed on the spool or log directories while checking their
7950 size for availability, Exim confusingly gave the error "space shortage".
7951 Furthermore, in debugging mode it crashed with a floating point exception.
7952 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7953 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7954 serious problem, Exim now writes to the main and panic logs when this
7955 happens, with details of the failure. It then refuses to accept the
7956 incoming message, giving the message "spool directory problem" or "log
7957 directory problem" with a 421 code for SMTP messages.
7959 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7960 0, 1, and 2 exist, because some OS complain for execs without them (see
7961 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7962 descriptors. However, the code omitted to check that the open succeeded,
7963 causing mysterious errors if for some reason the permissions on /dev/null
7964 got screwed. Now Exim writes a message to the main and panic logs, and
7965 bombs out if it can't open /dev/null.
7967 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7968 interact so that it is all more flexible. It is supposed to remain
7969 backwards compatible. Also added extra_local_interfaces.
7971 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7972 to bomb out with an assertion failure - to the client this appears as a
7973 connection drop. This problem occurs in the part of the code that was taken
7974 from the Samba project. Fortunately, the assertion is in a very simple
7975 function, so I have fixed this by reproducing the function inline in the
7976 one place where it is called, and arranging for authentication to fail
7977 instead of killing the process with assert().
7979 75. The SPA client code was not working when the server requested OEM rather
7980 than Unicode encoding.
7982 76. Added code to make require_files with a specific uid setting more usable in
7983 the case where statting the file as root fails - usually a non-root-mounted
7984 NFS file system. When this happens and the failure is EACCES, Exim now
7985 forks a subprocess and does the per-uid checking as the relevant uid.
7987 77. Added process_log_path.
7989 78. If log_file_path was not explicitly set, a setting of check_log_space or
7990 check_log_inodes was ignored.
7992 79. If a space check for the spool or log partitions fails, the incident is now
7993 logged. Of course, in the latter case the data may get lost...
7995 80. Added the %p formatting code to string_format() so that it can be used to
7996 print addresses in debug_print(). Adjusted all the address printing in the
7997 debugging in store.c to use %p rather than %d.
7999 81. There was a concern that a line of code in smtp_in.c could overflow a
8000 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
8001 initially expressed, the concern was not well-founded, because trailing
8002 spaces are removed early. However, if the trailing spaces were followed by
8003 a NULL, they did not get removed, so the overflow was possible. Two fixes
8006 (a) I re-wrote the offending code in a cleaner fashion.
8007 (b) If an incoming SMTP command contains a NULL character, it is rejected
8010 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
8011 runs initgroups(), so that if the Exim user is in any additional groups,
8012 they will be used during message reception.
8018 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob