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.
231 JH/01 Avoid costly startup code when not strictly needed. This reduces time
232 for some exim process initialisations. It does mean that the logging
233 of TLS configuration problems is only done for the daemon startup.
235 JH/02 Early-pipelining support code is now included unless disabled in Makefile.
237 JH/03 DKIM verification defaults no long accept sha1 hashes, to conform to
238 RFC 8301. They can still be enabled, using the dkim_verify_hashes main
241 JH/04 Support CHUNKING from an smtp transport using a transport_filter, when
242 DKIM signing is being done. Previously a transport_filter would always
243 disable CHUNKING, falling back to traditional DATA.
245 JH/05 Regard command-line recipients as tainted.
247 JH/06 Bug 340: Remove the daemon pid file on exit, when due to SIGTERM.
249 JH/07 Bug 2489: Fix crash in the "pam" expansion condition. It seems that the
250 PAM library frees one of the arguments given to it, despite the
251 documentation. Therefore a plain malloc must be used.
253 JH/08 Bug 2491: Use tainted buffers for the transport smtp context. Previously
254 on-stack buffers were used, resulting in a taint trap when DSN information
255 copied from a received message was written into the buffer.
257 JH/09 Bug 2493: Harden ARC verify against Outlook, whick has been seen to mix
258 the ordering of its ARC headers. This caused a crash.
260 JH/10 Bug 2492: Use tainted memory for retry record when needed. Previously when
261 a new record was being constructed with information from the peer, a trap
264 JH/11 Bug 2494: Unset the default for dmarc_tld_file. Previously a naiive
265 installation would get error messages from DMARC verify, when it hit the
266 nonexistent file indicated by the default. Distros wanting DMARC enabled
267 should both provide the file and set the option.
268 Also enforce no DMARC verification for command-line sourced messages.
270 JH/12 Fix an uninitialised flag in early-pipelining. Previously connections
271 could, depending on the platform, hang at the STARTTLS response.
273 JH/13 Bug 2498: Reset a counter used for ARC verify before handling another
274 message on a connection. Previously if one message had ARC headers and
275 the following one did not, a crash could result when adding an
276 Authentication-Results: header.
278 JH/14 Bug 2500: Rewind some of the common-coding in string handling between the
279 Exim main code and Exim-related utities. The introduction of taint
280 tracking also did many adjustments to string handling. Since then, eximon
281 frequently terminated with an assert failure.
283 JH/15 When PIPELINING, synch after every hundred or so RCPT commands sent and
284 check for 452 responses. This slightly helps the inefficieny of doing
285 a large alias-expansion into a recipient-limited target. The max_rcpt
286 transport option still applies (and at the current default, will override
287 the new feature). The check is done for either cause of synch, and forces
288 a fast-retry of all 452'd recipients using a new MAIL FROM on the same
289 connection. The new facility is not tunable at this time.
291 JH/16 Fix the variables set by the gsasl authenticator. Previously a pointer to
292 library live data was being used, so the results became garbage. Make
293 copies while it is still usable.
295 JH/17 Logging: when the deliver_time selector ise set, include the DT= field
296 on delivery deferred (==) and failed (**) lines (if a delivery was
297 attemtped). Previously it was only on completion (=>) lines.
299 JH/18 Authentication: the gsasl driver not provides the $authN variables in time
300 for the expansion of the server_scram_iter and server_scram_salt options.
302 WB/01 SPF: DNS lookups for the obsolete SPF RR type done by the libspf2 library
303 are now specifically given a NO_DATA response without hitting the system
304 resolver. The library goes on to do the now-standard TXT lookup.
305 Use of dnsdb lookups is not affected.
307 JH/19 Bug 2507: Modules: on handling a dynamic-module (lookups) open failure,
308 only retrieve the errormessage once. Previously two calls to dlerror()
309 were used, and the second one (for mainlog/paniclog) retrieved null
312 JH/20 Taint checking: disallow use of tainted data for
313 - the appendfile transport file and directory options
314 - the pipe transport command
315 - the autoreply transport file, log and once options
316 - file names used by the redirect router (including filter files)
318 - paths used by single-key lookups
319 Previously this was permitted.
321 JH/21 Bug 2501: Fix init call in the heimdal authenticator. Previously it
322 adjusted the size of a major service buffer; this failed because the
323 buffer was in use at the time. Change to a compile-time increase in the
324 buffer size, when this authenticator is compiled into exim.
326 JH/22 Taint-checking: move to safe-mode taint checking on all platforms. The
327 previous fast-mode was untenable in the face of glibs using mmap to
328 support larger malloc requests.
330 PP/01 Update the openssl_options possible values through OpenSSL 1.1.1c.
331 New values supported, if defined on system where compiled:
332 allow_no_dhe_kex, cryptopro_tlsext_bug, enable_middlebox_compat,
333 no_anti_replay, no_encrypt_then_mac, prioritize_chacha, tlsext_padding
335 JH/23 Performance improvement in the initial phase of a two-pass queue run. By
336 running a limited number of proceses in parallel, a benefit is gained. The
337 amount varies with the platform hardware and load. The use of the option
338 queue_run_in_order means we cannot do this, as ordering becomes
341 JH/24 Bug 2524: fix the cyrus_sasl auth driver gssapi usage. A previous fix
342 had introduced a string-copy (for ensuring NUL-termination) which was not
343 appropriate for that case, which can include embedded NUL bytes in the
344 block of data. Investigation showed the copy to actually be needless, the
345 data being length-specified.
347 JH/25 Fix use of concurrent TLS connections under GnuTLS. When a callout was
348 done during a receiving connection, and both used TLS, global info was
349 used rather than per-connection info for tracking the state of data
350 queued for transmission. This could result in a connection hang.
352 JH/26 Fix use of the SIZE parameter on MAIL commands, on continued connections.
353 Previously, when delivering serveral messages down a single connection
354 only the first would provide a SIZE. This was due to the size information
355 not being properly tracked.
357 JH/27 Bug 2530: When operating in a timezone with sub-minute offset, such as
358 TAI (at 37 seconds currently), pretend to be in UTC for time-related
359 expansion and logging. Previously, spurious values such as a future
360 minute could be seen.
362 JH/28 Bug 2533: Fix expansion of ${tr } item. When called in some situations
363 it could crash from a null-deref. This could also affect the
364 ${addresses: } operator and ${readsock } item.
366 JH/29 Bug 2537: Fix $mime_part_count. When a single connection had a non-mime
367 message following a mime one, the variable was not reset.
369 JH/30 When an pipelined-connect fails at the first response, assume incorrect
370 cached capability (perhaps the peer reneged?) and immediately retry in
373 JH/31 Fix spurious detection of timeout while writing to transport filter.
375 JH/32 Bug 2541: Fix segfault on bad cmdline -f (sender) argument. Previously
376 an attempt to copy the string was made before checking it.
378 JH/33 Fix the dsearch lookup to return an untainted result. Previously the
379 taint of the lookup key was maintained; we now regard the presence in the
380 filesystem as sufficient validation.
382 JH/34 Fix the readsocket expansion to not segfault when an empty "options"
383 argument is supplied.
385 JH/35 The dsearch lookup now requires that the directory is an absolute path.
386 Previously this was not checked, and nonempty relative paths made an
387 access under Exim's current working directory.
389 JH/36 Bug 2554: Fix msg:defer event for the hosts_max_try_hardlimit case.
390 Previously no event was raised.
392 JH/37 Bug 2552: Fix the check on spool space during reception to use the SIZE
393 parameter supplied by the sender MAIL FROM command. Previously it was
394 ignored, and only the check_spool_space option value for the required
397 JH/38 Fix $dkim_key_length. This should, after a DKIM verification, present
398 the size of the signing public-key. Previously it was instead giving
399 the size of the signature hash.
401 JH/39 DKIM verification: the RFC 8301 restriction on sizes of RSA keys is now
402 the default. See the (new) dkim_verify_min_keysizes option.
404 JH/40 Fix a memory-handling bug: when a connection carried multiple messages
405 and an ACL use a lookup for checking either the local_part or domain,
406 stale data could be accessed. Ensure that variable references are
407 dropped between messages.
409 JH/41 Bug 2571: Fix SPA authenticator. Running as a server, an offset supplied
410 by the client was not checked as pointing within response data before
411 being used. A malicious client could thus cause an out-of-bounds read and
412 possibly gain authentication. Fix by adding the check.
414 JH/42 Internationalisation: change the default for downconversion in the smtp
415 transport to be "if needed". Previously it was "as previously set" for
416 the message, which usually meant "if needed" for message-submission but
417 "no" for everything else. However, MTAs have been seen using SMTPUTF8
418 even when the envelope addresses did not need it, resulting in forwarding
419 failures to non-supporting MTAs. A downconvert in such cases will be
420 a no-op on the addresses, merely dropping the use of SMTPUTF8 by the
421 transport. The change does mean that addresses needing conversion will
422 be converted when previously a delivery failure would occur.
424 JH/43 Fix possible long line in DSN. Previously when a very long SMTP error
425 response was received it would be used unchecked in a fail-DSN, violating
426 standards on line-length limits. Truncate if needed.
428 HS/01 Remove parameters of the link to www.open-spf.org. The linked form
429 doesn't work. (Additionally add a new main config option to configure the
436 JH/01 OpenSSL: With debug enabled output keying information sufficient, server
437 side, to decode a TLS 1.3 packet capture.
439 JH/02 OpenSSL: Suppress the sending of (stateful) TLS1.3 session tickets.
440 Previously the default library behaviour applied, sending two, each in
443 JH/03 Debug output for ACL now gives the config file name and line number for
446 JH/04 The default received_header_text now uses the RFC 8314 tls cipher clause.
448 JH/05 DKIM: ensure that dkim_domain elements are lowercased before use.
450 JH/06 Fix buggy handling of autoreply bounce_return_size_limit, and a possible
451 buffer overrun for (non-chunking) other transports.
453 JH/07 GnuTLS: Our use of late (post-handshake) certificate verification, under
454 TLS1.3, means that a server rejecting a client certificate is not visible
455 to the client until the first read of encrypted data (typically the
456 response to EHLO). Add detection for that case and treat it as a failed
457 TLS connection attempt, so that the normal retry-in-clear can work (if
458 suitably configured).
460 JB/01 Bug 2375: fix expansions of 822 addresses having comments in local-part
461 and/or domain. Found and fixed by Jason Betts.
463 JH/08 Add hardening against SRV & TLSA lookups the hit CNAMEs (a nonvalid
464 configuration). If a CNAME target was not a wellformed name pattern, a
467 JH/09 Logging: Fix initial listening-on line for multiple ports for an IP when
468 the OS reports them interleaved with other addresses.
470 JH/10 OpenSSL: Fix aggregation of messages. Previously, when PIPELINING was
471 used both for input and for a verify callout, both encrypted, SMTP
472 responses being sent by the server could be lost. This resulted in
473 dropped connections and sometimes bounces generated by a peer sending
476 JH/11 Harden plaintext authenticator against a badly misconfigured client-send
477 string. Previously it was possible to cause undefined behaviour in a
478 library routine (usually a crash). Found by "zerons".
480 JH/12 Bug 2384: fix "-bP smtp_receive_timeout". Previously it returned no
483 JH/13 Bug 2386: Fix builds with Dane under LibreSSL 2.9.0 onward. Some old
484 API was removed, so update to use the newer ones.
486 JH/14 Bug 1891: Close the log file if receiving a non-smtp message, without
487 any timeout set, is taking a long time. Previously we would hang on to a
488 rotated logfile "forever" if the input was arriving with long gaps
489 (a previous attempt to fix addressed lack, for a long time, of initial
492 HS/01 Bug 2390: Use message_id for tempfile creation to avoid races in a
493 shared (NFS) environment. The length of the tempfile name is now
494 4 + 16 ("hdr.$message_exim_id") which might break on file
495 systems which restrict the file name length to lower values.
498 HS/02 Bug 2390: Use message_id for tempfile creation to avoid races in a
499 shared (NFS) environment.
501 HS/03 Bug 2392: exigrep does case sensitive *option* processing (as it
502 did for all versions <4.90). Notably -M, -m, --invert, -I may be
505 JH/15 Use unsigned when creating bitmasks in macros, to avoid build errors
506 on some platforms for bit 31.
508 JH/16 GnuTLS: rework ciphersuite strings under recent library versions. Thanks
509 to changes apparently associated with TLS1.3 handling some of the APIs
510 previously used were either nonfunctional or inappropriate. Strings
511 like TLS1.3:ECDHE_SECP256R1__RSA_PSS_RSAE_SHA256__AES_256_GCM__AEAD:256
512 and TLS1.2:ECDHE_SECP256R1__RSA_SHA256__AES_128_CBC__SHA256:128 replace
513 the previous TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256 .
514 This affects log line X= elements, the $tls_{in,out}_cipher variables,
515 and the use of specific cipher names in the encrypted= ACL condition.
517 JH/17 OpenSSL: the default openssl_options now disables ssl_v3.
519 JH/18 GnuTLS: fix $tls_out_ocsp under hosts_request_ocsp. Previously the
520 verification result was not updated unless hosts_require_ocsp applied.
522 JH/19 Bug 2398: fix listing of a named-queue. Previously, even with the option
523 queue_list_requires_admin set to false, non-admin users were denied the
526 JH/20 Bug 2389: fix server advertising of usable certificates, under GnuTLS in
527 directory-of-certs mode. Previously they were advertised despite the
530 JH/21 The smtp transport option "hosts_noproxy_tls" is now unset by default.
531 A single TCP connection by a client will now hold a TLS connection open
532 for multiple message deliveries, by default. Previously the default was to
535 JH/22 The smtp transport option "hosts_try_dane" now enables all hosts by
536 default. If built with the facility, DANE will be used. The facility
537 SUPPORT_DANE is now enabled in the prototype build Makefile "EDITME".
539 JH/23 The build default is now for TLS to be included; the SUPPORT_TLS define
540 is replaced with DISABLE_TLS. Either USE_GNUTLS or (the new) USE_OPENSSL
541 must be defined and you must still, unless you define DISABLE_TLS, manage
542 the the include-dir and library-file requirements that go with that
543 choice. Non-TLS builds are still supported.
545 JH/24 Fix duplicated logging of peer name/address, on a transport connection-
548 JH/25 The smtp transport option "hosts_try_fastopen" now enables all hosts by
549 default. If the platform supports and has the facility enabled, it will
550 be requested on all coneections.
552 JH/26 The PIPE_CONNECT facility is promoted from experimental status and is now
553 controlled by the build-time option SUPPORT_PIPE_CONNECT.
555 PP/01 Unbreak heimdal_gssapi, broken in 4.92.
557 JH/27 Bug 2404: Use the main-section configuration option "dsn_from" for
558 success-DSN messages. Previously the From: header was always the default
559 one for these; the option was ignored.
561 JH/28 Fix the timeout on smtp response to apply to the whole response.
562 Previously it was reset for every read, so a teergrubing peer sending
563 single bytes within the time limit could extend the connection for a
564 long time. Credit to Qualsys Security Advisory Team for the discovery.
566 JH/29 Fix DSN Final-Recipient: field. Previously it was the post-routing
567 delivery address, which leaked information of the results of local
568 forwarding. Change to the original envelope recipient address, per
571 JH/30 Bug 2411: Fix DSN generation when RFC 3461 failure notification is
572 requested. Previously not bounce was generated and a log entry of
573 error ignored was made.
575 JH/31 Avoid re-expansion in ${sort } expansion. (CVE-2019-13917)
577 JH/32 Introduce a general tainting mechanism for values read from the input
578 channel, and values derived from them. Refuse to expand any tainted
579 values, to catch one form of exploit.
581 JH/33 Bug 2413: Fix dkim_strict option. Previously the expansion result
582 was unused and the unexpanded text used for the test. Found and
583 fixed by Ruben Jenster.
585 JH/34 Fix crash after TLS shutdown. When the TCP/SMTP channel was left open,
586 an attempt to use a TLS library read routine dereffed a nul pointer,
589 JH/35 Bug 2409: filter out-of-spec chars from callout response before using
590 them in our smtp response.
592 JH/36 Have the general router option retry_use_local_part default to true when
593 any of the restrictive preconditions are set (to anything). Previously it
594 was only for check_local user. The change removes one item of manual
595 configuration which is required for proper retries when a remote router
596 handles a subset of addresses for a domain.
598 JH/37 Appendfile: when evaluating quota use (non-quota_size_regex) take the file
599 link count into consideration.
601 HS/04 Fix handling of very log lines in -H files. If a -<key> <value> line
602 caused the extension of big_buffer, the following lines were ignored.
604 JH/38 Bug 1395: Teach the DNS negative-cache about TTL value from the SOA in
605 accordance with RFC 2308. Previously there was no expiry, so a longlived
606 receive process (eg. due to ACL delays) versus a short SOA value could
609 HS/05 Handle trailing backslash gracefully. (CVE-2019-15846)
611 JH/39 Promote DMARC support to mainline.
613 JH/40 Bug 2452: Add a References: header to DSNs.
615 JH/41 With GnuTLS 3.6.0 (and later) do not attempt to manage Diffie-Hellman
616 parameters. The relevant library call is documented as "Deprecated: This
617 function is unnecessary and discouraged on GnuTLS 3.6.0 or later. Since
618 3.6.0, DH parameters are negotiated following RFC7919."
620 HS/06 Change the default of dnssec_request_domains to "*"
622 JH/42 Bug 2545: Fix CHUNKING for all RCPT commands rejected. Previously we
623 carried on and emitted a BDAT command, even when PIPELINING was not
626 JH/43 Bug 2465: Fix taint-handling in dsearch lookup. Previously a nontainted
627 buffer was used for the filename, resulting in a trap when tainted
628 arguments (eg. $domain) were used.
630 JH/44 With OpenSSL 1.1.1 (onwards) disable renegotiation for TLS1.2 and below;
631 recommended to avoid a possible server-load attack. The feature can be
632 re-enabled via the openssl_options main cofiguration option.
634 JH/45 local_scan API: documented the current smtp_printf() call. This changed
635 for version 4.90 - adding a "more data" boolean to the arguments.
636 Bumped the ABI version number also, this having been missed previously;
637 release versions 4.90 to 4.92.3 inclusive were effectively broken in
638 respect of usage of smtp_printf() by either local_scan code or libraries
639 accessed via the ${dlfunc } expansion item. Both will need coding
640 adjustment for any calls to smtp_printf() to match the new function
641 signature; a FALSE value for the new argument is always safe.
643 JH/46 FreeBSD: fix use of the sendfile() syscall. The shim was not updating
644 the file-offset (which the Linux syscall does, and exim expects); this
645 resulted in an indefinite loop.
647 JH/47 ARC: fix crash in signing, triggered when a configuration error failed
648 to do ARC verification. The Authentication-Results: header line added
649 by the configuration then had no ARC item.
655 JH/01 Remove code calling the customisable local_scan function, unless a new
656 definition "HAVE_LOCAL_SCAN=yes" is present in the Local/Makefile.
658 JH/02 Bug 1007: Avoid doing logging from signal-handlers, as that can result in
659 non-signal-safe functions being used.
661 JH/03 Bug 2269: When presented with a received message having a stupidly large
662 number of DKIM-Signature headers, disable DKIM verification to avoid
663 a resource-consumption attack. The limit is set at twenty.
665 JH/04 Add variables $arc_domains, $arc_oldest_pass for ARC verify. Fix the
666 report of oldest_pass in ${authres } in consequence, and separate out
667 some descriptions of reasons for verification fail.
669 JH/05 Bug 2273: Cutthrough delivery left a window where the received messsage
670 files in the spool were present and unlocked. A queue-runner could spot
671 them, resulting in a duplicate delivery. Fix that by doing the unlock
672 after the unlink. Investigation by Tim Stewart. Take the opportunity to
673 add more error-checking on spoolfile handling while that code is being
676 PP/01 Refuse to open a spool data file (*-D) if it's a symlink.
677 No known attacks, no CVE, this is defensive hardening.
679 JH/06 Bug 2275: The MIME ACL unlocked the received message files early, and
680 a queue-runner could start a delivery while other operations were ongoing.
681 Cutthrough delivery was a common victim, resulting in duplicate delivery.
682 Found and investigated by Tim Stewart. Fix by using the open message data
683 file handle rather than opening another, and not locally closing it (which
684 releases a lock) for that case, while creating the temporary .eml format
685 file for the MIME ACL. Also applies to "regex" and "spam" ACL conditions.
687 JH/07 Bug 177: Make a random-recipient callout success visible in ACL, by setting
688 $sender_verify_failure/$recipient_verify_failure to "random".
690 JH/08 When generating a selfsigned cert, use serial number 1 since zero is not
693 JH/09 Bug 2274: Fix logging of cmdline args when starting in an unlinked cwd.
694 Previously this would segfault.
696 JH/10 Fix ARC signing for case when DKIM signing failed. Previously this would
699 JH/11 Bug 2264: Exim now only follows CNAME chains one step by default. We'd
700 like zero, since the resolver should be doing this for us, But we need one
701 as a CNAME but no MX presence gets the CNAME returned; we need to check
702 that doesn't point to an MX to declare it "no MX returned" rather than
703 "error, loop". A new main option is added so the older capability of
704 following some limited number of chain links is maintained.
706 JH/12 Add client-ip info to non-pass iprev ${authres } lines.
708 JH/13 For receent Openssl versions (1.1 onward) use modern generic protocol
709 methods. These should support TLS 1.3; they arrived with TLS 1.3 and the
710 now-deprecated earlier definitions used only specified the range up to TLS
711 1.2 (in the older-version library docs).
713 JH/14 Bug 2284: Fix DKIM signing for body lines starting with a pair of dots.
715 JH/15 Rework TLS client-side context management. Stop using a global, and
716 explicitly pass a context around. This enables future use of TLS for
717 connections to service-daemons (eg. malware scanning) while a client smtp
718 connection is using TLS; with cutthrough connections this is quite likely.
720 JH/16 Fix ARC verification to do AS checks in reverse order.
722 JH/17 Support a "tls" option on the ${readsocket } expansion item.
724 JH/18 Bug 2287: Fix the protocol name (eg utf8esmtp) for multiple messages
725 using the SMTPUTF8 option on their MAIL FROM commands, in one connection.
726 Previously the "utf8" would be re-prepended for every additional message.
728 JH/19 Reject MAIL FROM commands with SMTPUTF8 when the facility was not advertised.
729 Previously thery were accepted, resulting in issues when attempting to
730 forward messages to a non-supporting MTA.
732 PP/02 Let -n work with printing macros too, not just options.
734 JH/20 Bug 2296: Fix cutthrough for >1 address redirection. Previously only
735 one parent address was copied, and bogus data was used at delivery-logging
736 time. Either a crash (after delivery) or bogus log data could result.
737 Discovery and analysis by Tim Stewart.
739 PP/03 Make ${utf8clean:} expansion operator detect incomplete final character.
740 Previously if the string ended mid-character, we did not insert the
741 promised '?' replacement.
743 PP/04 Documentation: current string operators work on bytes, not codepoints.
745 JH/21 Change as many as possible of the global flags into one-bit bitfields; these
746 should pack well giving a smaller memory footprint so better caching and
747 therefore performance. Group the declarations where this can't be done so
748 that the byte-sized flag variables are not interspersed among pointer
749 variables, giving a better chance of good packing by the compiler.
751 JH/22 Bug 1896: Fix the envelope from for DMARC forensic reports to be possibly
752 non-null, to avoid issues with sites running BATV. Previously reports were
753 sent with an empty envelope sender so looked like bounces.
755 JH/23 Bug 2318: Fix the noerror command within filters. It wasn't working.
756 The ignore_error flag wasn't being returned from the filter subprocess so
757 was not set for later routers. Investigation and fix by Matthias Kurz.
759 JH/24 Bug 2310: Raise a msg:fail:internal event for each undelivered recipient,
760 and a msg:complete for the whole, when a message is manually removed using
761 -Mrm. Developement by Matthias Kurz, hacked on by JH.
763 JH/25 Avoid fixed-size buffers for pathnames in DB access. This required using
764 a "Gnu special" function, asprintf() in the DB utility binary builds; I
765 hope that is portable enough.
767 JH/26 Bug 2311: Fix DANE-TA verification under GnuTLS. Previously it was also
768 requiring a known-CA anchor certificate; make it now rely entirely on the
769 TLSA as an anchor. Checking the name on the leaf cert against the name
770 on the A-record for the host is still done for TA (but not for EE mode).
772 JH/27 Fix logging of proxy address. Previously, a pointless "PRX=[]:0" would be
773 included in delivery lines for non-proxied connections, when compiled with
774 SUPPORT_SOCKS and running with proxy logging enabled.
776 JH/28 Bug 2314: Fire msg:fail:delivery event even when error is being ignored.
777 Developement by Matthias Kurz, tweaked by JH. While in that bit of code,
778 move the existing event to fire before the normal logging of message
779 failure so that custom logging is bracketed by normal logging.
781 JH/29 Bug 2322: A "fail" command in a non-system filter (file) now fires the
782 msg:fail:internal event. Developement by Matthias Kurz.
784 JH/30 Bug 2329: Increase buffer size used for dns lookup from 2k, which was
785 far too small for todays use of crypto signatures stored there. Go all
786 the way to the max DNS message size of 64kB, even though this might be
787 overmuch for IOT constrained device use.
789 JH/31 Fix a bad use of a copy function, which could be used to pointlessly
790 copy a string over itself. The library routine is documented as not
791 supporting overlapping copies, and on MacOS it actually raised a SIGABRT.
793 JH/32 For main options check_spool_space and check_inode_space, where the
794 platform supports 64b integers, support more than the previous 2^31 kB
795 (i.e. more than 2 TB). Accept E, P and T multipliers in addition to
796 the previous G, M, k.
798 JH/33 Bug 2338: Fix the cyrus-sasl authenticator to fill in the
799 $authenticated_fail_id variable on authentication failure. Previously
802 JH/34 Increase RSA keysize of autogen selfsign cert from 1024 to 2048. RHEL 8.0
803 OpenSSL didn't want to use such a weak key. Do for GnuTLS also, and for
804 more-modern GnuTLS move from GNUTLS_SEC_PARAM_LOW to
805 GNUTLS_SEC_PARAM_MEDIUM.
807 JH/35 OpenSSL: fail the handshake when SNI processing hits a problem, server
808 side. Previously we would continue as if no SNI had been received.
810 JH/36 Harden the handling of string-lists. When a list consisted of a sole
811 "<" character, which should be a list-separator specification, we walked
812 off past the nul-terimation.
814 JH/37 Bug 2341: Send "message delayed" warning MDNs (restricted to external
815 causes) even when the retry time is not yet met. Previously they were
816 not, meaning that when (say) an account was over-quota and temp-rejecting,
817 and multiple senders' messages were queued, only one sender would get
818 notified on each configured delay_warning cycle.
820 JH/38 Bug 2351: Log failures to extract envelope addresses from message headers.
822 JH/39 OpenSSL: clear the error stack after an SSL_accept(). With anon-auth
823 cipher-suites, an error can be left on the stack even for a succeeding
824 accept; this results in impossible error messages when a later operation
827 AM/01 Bug 2359: GnuTLS: repeat lowlevel read and write operations while they
828 return error codes indicating retry. Under TLS1.3 this becomes required.
830 JH/40 Fix the feature-cache refresh for EXPERIMENTAL_PIPE_CONNECT. Previously
831 it only wrote the new authenticators, resulting in a lack of tracking of
832 peer changes of ESMTP extensions until the next cache flush.
834 JH/41 Fix the loop reading a message header line to check for integer overflow,
835 and more-often against header_maxsize. Previously a crafted message could
836 induce a crash of the recive process; now the message is cleanly rejected.
838 JH/42 Bug 2366: Fix the behaviour of the dkim_verify_signers option. It had
839 been totally disabled for all of 4.91. Discovery and fix by "Mad Alex".
845 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
846 When redis_servers is set to a list of > 1 element, and the Redis servers
847 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
848 case of MOVED into a DEFER case instead, thus moving the query onto the
849 next server in the list. For a cluster of N elements, all N servers must
850 be defined in redis_servers.
852 GF/02 Catch and remove uninitialized value warning in exiqsumm
853 Check for existence of @ARGV before looking at $ARGV[0]
855 JH/01 Replace the store_release() internal interface with store_newblock(),
856 which internalises the check required to safely use the old one, plus
857 the allocate and data copy operations duplicated in both (!) of the
858 extant use locations.
860 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
861 modifier. This matches the restriction on the commandline.
863 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
864 Previously only the last row was returned.
866 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
867 we assumed that tags in the header were well-formed, and parsed the
868 element content after inspecting only the first char of the tag.
869 Assumptions at that stage could crash the receive process on malformed
872 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
873 While running the DKIM ACL we operate on the Permanent memory pool so that
874 variables created with "set" persist to the DATA ACL. Also (at any time)
875 DNS lookups that fail create cache records using the Permanent pool. But
876 expansions release any allocations made on the current pool - so a dnsdb
877 lookup expansion done in the DKIM ACL releases the memory used for the
878 DNS negative-cache, and bad things result. Solution is to switch to the
879 Main pool for expansions.
880 While we're in that code, add checks on the DNS cache during store_reset,
881 active in the testsuite.
882 Problem spotted, and debugging aided, by Wolfgang Breyha.
884 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
885 When none of the hosts presented to a transport match an already-open
886 connection, close it and proceed with the list. Previously we would
887 queue the message. Spotted by Lena with Yahoo, probably involving
890 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
891 Previously a spurious "250 OK id=" response was appended to the proper
894 JH/08 The "support for" informational output now, which built with Content
895 Scanning support, has a line for the malware scanner interfaces compiled
896 in. Interface can be individually included or not at build time.
898 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
899 by the template makefile "src/EDITME". The "STREAM" support for an older
900 ClamAV interface method is removed.
902 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
903 rows affected is given instead).
905 JH/11 The runtime Berkeley DB library version is now additionally output by
906 "exim -d -bV". Previously only the compile-time version was shown.
908 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
909 SMTP connection. Previously, when one had more recipients than the
910 first, an abortive onward connection was made. Move to full support for
911 multiple onward connections in sequence, handling cutthrough connection
912 for all multi-message initiating connections.
914 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
915 routers. Previously, a multi-recipient message would fail to match the
916 onward-connection opened for the first recipient, and cause its closure.
918 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
919 a timeout on read on a GnuTLS initiating connection, resulting in the
920 initiating connection being dropped. This mattered most when the callout
921 was marked defer_ok. Fix to keep the two timeout-detection methods
924 JH/15 Relax results from ACL control request to enable cutthrough, in
925 unsupported situations, from error to silently (except under debug)
926 ignoring. This covers use with PRDR, frozen messages, queue-only and
929 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
931 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
932 metadata, resulting in a crash in free().
934 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
935 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
936 Broken also in d185889f4, with init system revamp.
938 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
939 Previously we abruptly closed the connection after reading a malware-
940 found indication; now we go on to read the "scan ok" response line,
943 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
944 ACL. Previously, a crash would result.
946 JH/19 Speed up macro lookups during configuration file read, by skipping non-
947 macro text after a replacement (previously it was only once per line) and
948 by skipping builtin macros when searching for an uppercase lead character.
950 JH/20 DANE support moved from Experimental to mainline. The Makefile control
951 for the build is renamed.
953 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
954 was allocated for every new TLS startup, meaning one per message. Fix
955 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
957 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
958 reported the original. Fix to report (as far as possible) the ACL
959 result replacing the original.
961 JH/23 Fix memory leak during multi-message connections using STARTTLS under
962 OpenSSL. Certificate information is loaded for every new TLS startup,
963 and the resources needed to be freed.
965 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
967 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
970 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
971 DATA response info to the (existing) per-recipient response info for
972 the "C=" log element. It can have useful tracking info from the
973 destination system. Patch from Simon Arlott.
975 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
976 length value. Previously this would segfault.
978 HS/02 Support Avast multiline protoocol, this allows passing flags to
979 newer versions of the scanner.
981 JH/28 Ensure that variables possibly set during message acceptance are marked
982 dead before release of memory in the daemon loop. This stops complaints
983 about them when the debug_store option is enabled. Discovered specifically
984 for sender_rate_period, but applies to a whole set of variables.
985 Do the same for the queue-runner and queue-list loops, for variables set
986 from spool message files. Do the same for the SMTP per-message loop, for
987 certain variables indirectly set in ACL operations.
989 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
990 as a multi-recipient message from a mailinglist manager). The coding had
991 an arbitrary cutoff number of characters while checking for more input;
992 enforced by writing a NUL into the buffer. This corrupted long / fast
993 input. The problem was exposed more widely when more pipelineing of SMTP
994 responses was introduced, and one Exim system was feeding another.
995 The symptom is log complaints of SMTP syntax error (NUL chars) on the
996 receiving system, and refused recipients seen by the sending system
997 (propating to people being dropped from mailing lists).
998 Discovered and pinpointed by David Carter.
1000 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
1001 replaced by the ${authresults } expansion.
1003 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
1005 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
1006 allows proper process termination in container environments.
1008 JH/32 Bug 2258: Fix spool_wireformat in combination with LMTP transport.
1009 Previously the "final dot" had a newline after it; ensure it is CR,LF.
1011 JH/33 SPF: remove support for the "spf" ACL condition outcome values "err_temp"
1012 and "err_perm", deprecated since 4.83 when the RFC-defined words
1013 "temperror" and "permerror" were introduced.
1015 JH/34 Re-introduce enforcement of no cutthrough delivery on transports having
1016 transport-filters or DKIM-signing. The restriction was lost in the
1017 consolidation of verify-callout and delivery SMTP handling.
1018 Extend the restriction to also cover ARC-signing.
1020 JH/35 Cutthrough: for a final-dot response timeout (and nonunderstood responses)
1021 in defer=pass mode supply a 450 to the initiator. Previously the message
1024 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
1025 tls_require_ciphers is used as before.
1027 HS/03 Malware Avast: Better match the Avast multiline protocol. Add
1028 "pass_unscanned". Only tmpfails from the scanner are written to
1029 the paniclog, as they may require admin intervention (permission
1030 denied, license issues). Other scanner errors (like decompression
1031 bombs) do not cause a paniclog entry.
1033 JH/36 Fix reinitialisation of DKIM logging variable between messages.
1034 Previously it was possible to log spurious information in receive log
1037 JH/37 Bug 2255: Revert the disable of the OpenSSL session caching. This
1038 triggered odd behaviour from Outlook Express clients.
1040 PP/03 Add util/renew-opendmarc-tlds.sh script for safe renewal of public
1043 JH/38 DKIM: accept Ed25519 pubkeys in SubjectPublicKeyInfo-wrapped form,
1044 since the IETF WG has not yet settled on that versus the original
1045 "bare" representation.
1047 JH/39 Fix syslog logging for syslog_timestamp=no and log_selector +millisec.
1048 Previously the millisecond value corrupted the output.
1049 Fix also for syslog_pid=no and log_selector +pid, for which the pid
1050 corrupted the output.
1056 JH/01 Rework error string handling in TLS interface so that the caller in
1057 more cases is responsible for logging. This permits library-sourced
1058 string to be attached to addresses during delivery, and collapses
1059 pairs of long lines into single ones.
1061 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
1062 during configuration. Wildcards are allowed and expanded.
1064 JH/02 Rework error string handling in DKIM to pass more info back to callers.
1065 This permits better logging.
1067 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
1068 do not close it down and have the child transport start it up again on
1069 the passed-on TCP connection. Instead, proxy the child (and any
1070 subsequent ones) for TLS via a unix-domain socket channel. Logging is
1071 affected: the continued delivery log lines do not have any DNSSEC, TLS
1072 Certificate or OCSP information. TLS cipher information is still logged.
1074 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
1075 identical IP addresses on different listening ports. Will also affect
1078 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
1079 add noisy ifdef guards to special-case this sillyness.
1080 Patch from Bernd Kuhls.
1082 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
1083 than 255 are no longer allowed.
1085 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
1086 Disable the session-cache too, which might reduce our load. Since we
1087 currrectly use a new context for every connection, both as server and
1088 client, there is no benefit for these.
1089 GnuTLS appears to not support tickets server-side by default (we don't
1090 call gnutls_session_ticket_enable_server()) but client side is enabled
1091 by default on recent versions (3.1.3 +) unless the PFS priority string
1094 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
1095 <https://reproducible-builds.org/specs/source-date-epoch/>.
1097 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
1098 the check for any unsuccessful recipients did not notice the limit, and
1099 erroneously found still-pending ones.
1101 JH/08 Pipeline CHUNKING command and data together, on kernels that support
1102 MSG_MORE. Only in-clear (not on TLS connections).
1104 JH/09 Avoid using a temporary file during transport using dkim. Unless a
1105 transport-filter is involved we can buffer the headers in memory for
1106 creating the signature, and read the spool data file once for the
1107 signature and again for transmission.
1109 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
1110 4.77 as the kernel support then wasn't solid, having issues in 64bit
1111 mode. Now, it's been long enough. Add support for FreeBSD also.
1113 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
1114 case where the routing stage had gathered several addresses to send to
1115 a host before calling the transport for the first, we previously failed
1116 to close down TLS in the old transport process before passing the TCP
1117 connection to the new process. The new one sent a STARTTLS command
1118 which naturally failed, giving a failed delivery and bloating the retry
1119 database. Investigation and fix prototype from Wolfgang Breyha.
1121 JH/12 Fix check on SMTP command input synchronisation. Previously there were
1122 false-negatives in the check that the sender had not preempted a response
1123 or prompt from Exim (running as a server), due to that code's lack of
1124 awareness of the SMTP input buffering.
1126 PP/04 Add commandline_checks_require_admin option.
1127 Exim drops privileges sanely, various checks such as -be aren't a
1128 security problem, as long as you trust local users with access to their
1129 own account. When invoked by services which pass untrusted data to
1130 Exim, this might be an issue. Set this option in main configuration
1131 AND make fixes to the calling application, such as using `--` to stop
1134 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
1135 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
1136 responses to those, into a single TLS record each way (this usually means
1137 a single packet). As a side issue, smtp_enforce_sync now works on TLS
1140 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
1141 affects you only if you're dancing at the edge of the param size limits.
1142 If you are, and this message makes sense to you, then: raise the
1143 configured limit or use OpenSSL 1.1. Nothing we can do for older
1146 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
1147 cmdline element to get the documented default one. Previously it was
1150 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
1151 are unsuable for non-message-specific reasons - eg. connection timeout,
1152 banner-time rejection.
1154 JH/16 Fix logging of delivery remote port, when specified by router, under
1157 PP/06 Repair manualroute's ability to take options in any order, even if one
1158 is the name of a transport.
1161 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
1163 JH/17 Change the list-building routines interface to use the expanding-string
1164 triplet model, for better allocation and copying behaviour.
1166 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
1167 Previously it was constructed the first time a possibly-matching string
1168 was met in the configuration file input during startup; now it is done
1171 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
1172 compatible one, to avoid the (poorly documented) possibility of a config
1173 file in the working directory redirecting the DB files, possibly correpting
1174 some existing file. CVE-2017-10140 assigned for BDB.
1176 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
1177 cache-hot. Previously, although the result was properly cached, the
1178 initial verify call returned a defer.
1180 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
1181 the main verify for receipient in uncached-mode.
1183 JH/22 Retire historical build files to an "unsupported" subdir. These are
1184 defined as "ones for which we have no current evidence of testing".
1186 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
1187 if present. Previously it was ignored.
1189 JH/24 Start using specified-initialisers in C structure init coding. This is
1190 a C99 feature (it's 2017, so now considered safe).
1192 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
1193 if was a fixed-sized field and bitmask ops via macros; it is now more
1196 PP/07 GitHub PR 56: Apply MariaDB build fix.
1197 Patch provided by Jaroslav Škarvada.
1199 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
1200 during Coverity cleanups [4.87 JH/47]
1201 Diagnosis and fix provided by Michael Fischer v. Mollard.
1203 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
1204 the right size to place the terminating semicolon on its own folded
1205 line, the header hash was calculated to an incorrect value thanks to
1206 the (relaxed) space the fold became.
1208 HS/02 Fix Bug 2130: large writes from the transport subprocess were chunked
1209 and confused the parent.
1211 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
1212 which could crash as a result. This could lead to undeliverable messages.
1214 JH/28 Logging: "next input sent too soon" now shows where input was truncated
1217 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
1218 matters on fast-turnover and PID-randomising systems, which were getting
1219 out-of-order delivery.
1221 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
1222 a possibly-overlapping copy. The symptom was that "Remote host closed
1223 connection in response to HELO" was logged instead of the actual 4xx
1226 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
1227 Previously only that bufferd was discarded, resulting in SYMTP command
1230 JH/32 DKIM: when a message has multiple signatures matching an identity given
1231 in dkim_verify_signers, run the dkim acl once for each. Previously only
1232 one run was done. Bug 2189.
1234 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
1235 "panic the current process" to "deliberately defer". The panic log is
1236 still written with the problem list name; the mail and reject logs now
1237 get a temp-reject line for the message that was being handled, saying
1238 something like "domains check lookup or other defer". The SMTP 451
1239 message is still "Temporary local problem".
1241 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
1242 A crafted sequence of BDAT commands could result in in-use memory beeing
1243 freed. CVE-2017-16943.
1245 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
1246 from SMTP input. Previously it was always done; now only done for DATA
1247 and not BDAT commands. CVE-2017-16944.
1249 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
1250 to the message (such as an overlong header line). Previously this was
1251 not done and we did not exit BDAT mode. Followon from the previous item
1252 though a different problem.
1258 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
1259 than -2003 did; needs libidn2 in addition to libidn.
1261 JH/02 The path option on a pipe transport is now expanded before use.
1263 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
1264 Patch provided by "Björn", documentation fix added too.
1266 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
1267 missing a wire-to-host endian conversion.
1269 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
1270 close after a BDAT command line could be taken as a following command,
1271 giving a synch failure. Fix by only checking for synch immediately
1272 before acknowledging the chunk.
1274 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
1275 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
1276 macro. Patches provided by Josh Soref.
1278 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
1279 Previously we did not; the RFC seems ambiguous and VRFY is not listed
1280 by IANA as a service extension. However, John Klensin suggests that we
1283 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
1284 the dkim code may be unix-mode line endings rather than smtp wire-format
1285 CRLF, so prepend a CR to any bare LF.
1287 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
1288 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
1290 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
1291 the first were themselves being wrongly included in the feed into dkim
1292 processing; with most chunk sizes in use this resulted in an incorrect
1293 body hash calculated value.
1295 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
1296 DKIM signature block, for verification. Although advised against by
1297 standards it is specifically not ruled illegal.
1299 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
1301 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
1302 missing a body hash (the bh= tag).
1304 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
1305 It seems that HAProxy sends the Proxy Protocol information in clear and
1306 only then does a TLS startup, so do the same.
1308 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
1309 TCP connections (such as for Spamd) unless the daemon successfully set
1310 Fast Open mode on its listening sockets. This fixes breakage seen on
1311 too-old kernels or those not configured for Fast Open, at the cost of
1312 requiring both directions being enabled for TFO, and TFO never being used
1313 by non-daemon-related Exim processes.
1315 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
1316 endings, at least on the first header line. Try to canonify any that get
1317 past that check, despite the cost.
1319 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
1320 now limited to an arbitrary five deep, while parsing addresses with the
1321 strip_excess_angle_brackets option enabled.
1323 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
1324 instead leave the unprompted TLS handshake in socket buffer for the
1325 TLS library to consume.
1327 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
1329 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
1331 JH/16 Drop variables when they go out of scope. Memory management drops a whole
1332 region in one operation, for speed, and this leaves assigned pointers
1333 dangling. Add checks run only under the testsuite which checks all
1334 variables at a store-reset and panics on a dangling pointer; add code
1335 explicitly nulling out all the variables discovered. Fixes one known
1336 bug: a transport crash, where a dangling pointer for $sending_ip_address
1337 originally assigned in a verify callout, is re-used.
1339 PP/06 Drop '.' from @INC in various Perl scripts.
1341 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
1343 PP/08 Reduce a number of compilation warnings under clang; building with
1344 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
1345 should be warning-free.
1347 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
1349 HS/01 Fix portability problems introduced by PP/08 for platforms where
1350 realloc(NULL) is not equivalent to malloc() [SunOS et al].
1352 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
1353 chunk. This allows us to accept broken chunked messages. We need a more
1354 general solution here.
1356 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
1357 already-broken messages in the queue.
1359 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
1361 JH/19 Fix reference counting bug in routing-generated-address tracking.
1367 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
1368 supports it and a size is available (ie. the sending peer gave us one).
1370 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
1371 years of being deprecated). The replacements are the ACLs
1372 acl_smtp_mime and acl_not_smtp_mime.
1374 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
1375 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
1376 or even an in-clear connection were permitted. Now, if the host lookup
1377 was dnssec and dane was requested then the host is only used if the
1378 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
1379 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
1380 if one fails this test.
1381 This means that a poorly-configured remote DNS will make it incommunicado;
1382 but it protects against a DNS-interception attack on it.
1384 JH/04 Bug 1810: make continued-use of an open smtp transport connection
1385 non-noisy when a race steals the message being considered.
1387 JH/05 If main configuration option tls_certificate is unset, generate a
1388 self-signed certificate for inbound TLS connections.
1390 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
1391 in rewrites and routers.
1393 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
1394 and logged a warning sing 4.83; now they are a configuration file error.
1396 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
1397 (lacking @domain). Apply the same qualification processing as RCPT.
1399 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
1401 JH/10 Support ${sha256:} applied to a string (as well as the previous
1404 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
1405 a cutthrough deliver is pending, as we always want to make a connection.
1406 This also avoids re-routing the message when later placing the cutthrough
1407 connection after a verify cache hit.
1408 Do not update it with the verify result either.
1410 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
1411 when routing results in more than one destination address.
1413 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
1414 signing (which inhibits the cutthrough capability). Previously only
1415 the presence of an option was tested; now an expansion evaluating as
1416 empty is permissible (obviously it should depend only on data available
1417 when the cutthrough connection is made).
1419 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
1420 the relevant preceding SMTP command did not note the pipelining mode.
1422 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
1423 Previously they were not counted.
1425 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
1426 as one having no matching records. Previously we deferred the message
1427 that needed the lookup.
1429 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
1430 distinguished as "(=".
1432 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
1433 for missing MX records. Previously it only worked for missing A records.
1435 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
1437 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
1438 after the data-go-ahead and data-ack. Patch from Jason Betts.
1440 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
1441 even for a "none" policy. Patch from Tony Meyer.
1443 JH/22 Fix continued use of a connection for further deliveries. If a port was
1444 specified by a router, it must also match for the delivery to be
1447 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
1448 When one of the recipients of a message was unsuitable for the connection
1449 (has no matching addresses), we lost track of needing to mark it
1450 deferred. As a result mail would be lost.
1452 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
1454 JH/25 Decoding ACL controls is now done using a binary search; the source code
1455 takes up less space and should be simpler to maintain. Merge the ACL
1456 condition decode tables also, with similar effect.
1458 JH/26 Fix problem with one_time used on a redirect router which returned the
1459 parent address unchanged. A retry would see the parent address marked as
1460 delivered, so not attempt the (identical) child. As a result mail would
1463 JH/27 Fix a possible security hole, wherein a process operating with the Exim
1464 UID can gain a root shell. Credit to http://www.halfdog.net/ for
1465 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
1468 JH/28 Enable {spool,log} filesystem space and inode checks as default.
1469 Main config options check_{log,spool}_{inodes,space} are now
1470 100 inodes, 10MB unless set otherwise in the configuration.
1472 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
1473 Previously it only applied to the main-section connection policy
1476 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
1478 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
1479 by me. Added RFC7919 DH primes as an alternative.
1481 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
1482 are not in the system include path.
1484 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
1485 GnuTLS, when a session startup failed (eg because the client disconnected)
1486 Exim did stdio operations after fclose. This was exposed by a recent
1487 change which nulled out the file handle after the fclose.
1489 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
1490 signed directly by the cert-signing cert, rather than an intermediate
1491 OCSP-signing cert. This is the model used by LetsEncrypt.
1493 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
1495 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
1496 an incoming connection.
1498 HS/02 Bug 1802: Do not half-close the connection after sending a request
1501 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
1502 fallback to "prime256v1".
1504 JH/34 SECURITY: Use proper copy of DATA command in error message.
1505 Could leak key material. Remotely exploitable. CVE-2016-9963.
1511 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
1512 and 3.4.4 - once the server is enabled to respond to an OCSP request
1513 it does even when not requested, resulting in a stapling non-aware
1514 client dropping the TLS connection.
1516 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
1517 support variable-length bit vectors. No functional change.
1519 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
1520 The I= interface field on outgoing lines is now after the H= remote
1521 host field, same as incoming lines. There is a separate
1522 outgoing_interface log selector which allows you to disable the
1525 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
1526 If not running log_selector +smtp_connection the mainlog would be held
1527 open indefinitely after a "too many connections" event, including to a
1528 deleted file after a log rotate. Leave the per net connection logging
1529 leaving it open for efficiency as that will be quickly detected by the
1530 check on the next write.
1532 HS/01 Bug 1671: Fix post transport crash.
1533 Processing the wait-<transport> messages could crash the delivery
1534 process if the message IDs didn't exist for some reason. When
1535 using 'split_spool_directory=yes' the construction of the spool
1536 file name failed already, exposing the same netto behaviour.
1538 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
1539 mime_regex ACL conditions.
1541 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
1542 to DSN fail messages (bounces): remote IP, remote greeting, remote response
1543 to HELO, local diagnostic string.
1545 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
1546 log line to debug. Even when configured with a tls authenticator many
1547 client connections are expected to not authenticate in this way, so
1548 an authenticate fail is not an error.
1550 HS/02 Add the Exim version string to the process info. This way exiwhat
1551 gives some more detail about the running daemon.
1553 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
1554 matter for fast-change records such as DNSBLs.
1556 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
1557 retry record, even if constant. There may be multiple transports with
1558 different interface settings and the retry behaviour needs to be kept
1561 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
1563 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
1565 JH/10 Bug 840: fix log_defer_output option of pipe transport
1567 JH/11 Bug 830: use same host for all RCPTS of a message, even under
1568 hosts_randomize. This matters a lot when combined with mua_wrapper.
1570 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
1571 ${quote_pgsql:<string>} operator.
1573 JH/13 Bug 1708: avoid misaligned access in cached lookup.
1575 JH/14 Change header file name for freeradius-client. Relevant if compiling
1576 with Radius support; from the Gentoo tree and checked under Fedora.
1578 JH/15 Bug 1712: Introduce $prdr_requested flag variable
1580 JH/16 Bug 1714: Permit an empty string as expansion result for transport
1581 option transport_filter, meaning no filtering.
1583 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
1585 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
1586 defaults to "*" (all hosts). The variable is now available when not built
1587 with TLS, default unset, mainly to enable keeping the testsuite sane.
1588 If a server certificate is not supplied (via tls_certificate) an error is
1589 logged, and clients will find TLS connections fail on startup. Presumably
1590 they will retry in-clear.
1591 Packagers of Exim are strongly encouraged to create a server certificate
1592 at installation time.
1594 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
1595 with the $config_file variable.
1597 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
1598 in transport context, after the attempt, and per-recipient. The latter type
1599 is per host attempted. The event data is the error message, and the errno
1600 information encodes the lookup type (A vs. MX) used for the (first) host,
1601 and the trailing two digits of the smtp 4xx response.
1603 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
1604 to write to mainlog (or rejectlog, paniclog) in the window between file
1605 creation and permissions/ownership being changed. Particularly affects
1606 installations where exicyclog is run as root, rather than exim user;
1607 result is that the running daemon panics and dies.
1609 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
1611 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
1612 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
1613 "pri" and "weight". Note that the previous implicit priority given by the
1614 list order is no longer honoured.
1616 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
1617 for DKIM processing.
1619 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
1620 by defining SUPPORT_SOCKS.
1622 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
1623 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
1624 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
1625 variables are renamed to proxy_{local,external}_{address,port}.
1627 JH/27 Move Internationalisation support from Experimental to mainline, enabled
1628 for a build by defining SUPPORT_I18N
1630 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
1631 of the query string, and make ${quote_redis:} do that quoting.
1633 JH/29 Move Events support from Experimental to mainline, enabled by default
1634 and removable for a build by defining DISABLE_EVENT.
1636 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
1638 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
1639 cached by the daemon.
1641 JH/32 Move Redis support from Experimental to mainline, enabled for a build
1642 by defining LOOKUP_REDIS. The libhiredis library is required.
1644 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
1645 keys are given for lookup.
1647 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
1648 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
1649 only supported when built with TLS support. The PolarSSL SHA routines
1650 are still used when the TLS library is too old for convenient support.
1652 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
1653 openssl_options), for security. OpenSSL forces this from version 1.1.0
1654 server-side so match that on older versions.
1656 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
1657 allocation for $value could be released as the expansion processing
1658 concluded, but leaving the global pointer active for it.
1660 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
1661 and to use the domains and local_parts ACL conditions.
1663 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
1664 incorrectly not doubled on cutthrough transmission, hence seen as a
1665 body-termination at the receiving system - resulting in truncated mails.
1666 Commonly the sender saw a TCP-level error, and retransmitted the message
1667 via the normal store-and-forward channel. This could result in duplicates
1668 received - but deduplicating mailstores were liable to retain only the
1669 initial truncated version.
1671 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
1673 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
1675 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
1676 we're in there, support oversigning also; bug 1309.
1678 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
1680 HS/04 Add support for keep_environment and add_environment options.
1682 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
1683 either intentional arithmetic overflow during PRNG, or testing config-
1686 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
1687 delivery resulted in actual delivery. Cancel cutthrough before DATA
1690 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
1691 rejects a recipient: pass the reject to the originator.
1693 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
1694 Many were false-positives and ignorable, but it's worth fixing the
1697 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
1698 for the new environment-manipulation done at startup. Move the routines
1699 from being local to tls.c to being global via the os.c file.
1701 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
1702 an extract embedded as result-arg for a map, the first arg for extract
1703 is unavailable so we cannot tell if this is a numbered or keyed
1704 extraction. Accept either.
1710 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
1713 JH/02 The smtp transport option "multi_domain" is now expanded.
1715 JH/03 The smtp transport now requests PRDR by default, if the server offers
1718 JH/04 Certificate name checking on server certificates, when exim is a client,
1719 is now done by default. The transport option tls_verify_cert_hostnames
1720 can be used to disable this per-host. The build option
1721 EXPERIMENTAL_CERTNAMES is withdrawn.
1723 JH/05 The value of the tls_verify_certificates smtp transport and main options
1724 default to the word "system" to access the system default CA bundle.
1725 For GnuTLS, only version 3.0.20 or later.
1727 JH/06 Verification of the server certificate for a TLS connection is now tried
1728 (but not required) by default. The verification status is now logged by
1729 default, for both outbound TLS and client-certificate supplying inbound
1732 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
1735 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
1736 Status Notification (bounce) messages are now MIME format per RFC 3464.
1737 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
1738 under the control of the dsn_advertise_hosts option, and routers may
1739 have a dsn_lasthop option.
1741 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
1742 default, modifiable by a malware= option. The list separator for
1743 the options can now be changed in the usual way. Bug 68.
1745 JH/10 The smtp_receive_timeout main option is now expanded before use.
1747 JH/11 The incoming_interface log option now also enables logging of the
1748 local interface on delivery outgoing connections.
1750 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
1751 if the interface and destination host and port all match.
1753 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
1756 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
1757 Patch from Andrew Lewis.
1759 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
1760 now supports optional time-restrictions, weighting, and priority
1761 modifiers per server. Patch originally by <rommer@active.by>.
1763 JH/16 The spamd_address main option now supports a mixed list of local
1764 and remote servers. Remote servers can be IPv6 addresses, and
1765 specify a port-range.
1767 JH/17 Bug 68: The spamd_address main option now supports an optional
1768 timeout value per server.
1770 JH/18 Bug 1581: Router and transport options headers_add/remove can
1771 now have the list separator specified.
1773 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
1776 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
1779 JH/21 Support for the A6 type of dns record is withdrawn.
1781 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
1782 rather than the verbs used.
1784 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
1785 from 255 to 1024 chars.
1787 JH/24 Verification callouts now attempt to use TLS by default.
1789 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
1790 are generic router options now. The defaults didn't change.
1792 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
1793 Original patch from Alexander Shikoff, worked over by JH.
1795 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
1796 files if ZCAT_COMMAND is not executable.
1798 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
1800 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
1802 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
1803 Normally benign, it bites when the pair was led to by a CNAME;
1804 modern usage is to not canonicalize the domain to a CNAME target
1805 (and we were inconsistent anyway for A-only vs AAAA+A).
1807 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
1809 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
1810 when evaluating $sender_host_dnssec.
1812 JH/31 Check the HELO verification lookup for DNSSEC, adding new
1813 $sender_helo_dnssec variable.
1815 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
1817 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
1819 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
1821 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
1822 documented as working, but never had. Support all but $spam_report.
1824 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
1825 added for tls authenticator.
1827 HS/03 Add perl_taintmode main config option
1833 TL/01 When running the test suite, the README says that variables such as
1834 no_msglog_check are global and can be placed anywhere in a specific
1835 test's script, however it was observed that placement needed to be near
1836 the beginning for it to behave that way. Changed the runtest perl
1837 script to read through the entire script once to detect and set these
1838 variables, reset to the beginning of the script, and then run through
1839 the script parsing/test process like normal.
1841 TL/02 The BSD's have an arc4random API. One of the functions to induce
1842 adding randomness was arc4random_stir(), but it has been removed in
1843 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
1844 function when detected.
1846 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
1847 cause callback expansion.
1849 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
1850 syntax errors in an expansion can be treated as a string instead of
1851 logging or causing an error, due to the internal use of bool_lax
1852 instead of bool when processing it.
1854 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
1855 server certificates when making smtp deliveries.
1857 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
1859 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
1861 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1863 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1864 Merged patch from Sebastian Wiedenroth.
1866 JH/05 Fix results-pipe from transport process. Several recipients, combined
1867 with certificate use, exposed issues where response data items split
1868 over buffer boundaries were not parsed properly. This eventually
1869 resulted in duplicates being sent. This issue only became common enough
1870 to notice due to the introduction of connection certificate information,
1871 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1873 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1874 size buffer was used, resulting in syntax errors when an expansion
1877 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1878 version 3.3.6 or later.
1880 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1881 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1882 both become "event_action", the variables become $event_name, $event_data
1883 and $event_defer_errno. There is a new variable $verify_mode, usable in
1884 routers, transports and related events. The tls:cert event is now also
1885 raised for inbound connections, if the main configuration event_action
1888 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1889 early OCSP support, but no stapling (appears to be less than 1.0.0).
1891 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1892 server certificate names available under the smtp transport option
1893 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1896 JH/10 Time-related extraction expansions from certificates now use the main
1897 option "timezone" setting for output formatting, and are consistent
1898 between OpenSSL and GnuTLS compilations. Bug 1541.
1900 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1901 encoded parameter in the incoming message. Bug 1558.
1903 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1904 include certificate info, eximon was claiming there were spoolfile
1907 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1909 JH/14 Log delivery-related information more consistently, using the sequence
1910 "H=<name> [<ip>]" wherever possible.
1912 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1913 are problematic for Debian distribution, omit them from the release
1916 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1918 JH/16 Fix string representation of time values on 64bit time_t architectures.
1921 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1922 output list separator was used.
1927 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1928 checkers that were complaining about end of non-void function with no
1931 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1932 This was a regression introduced in 4.83 by another bugfix.
1934 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1936 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1937 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1943 TF/01 Correctly close the server side of TLS when forking for delivery.
1945 When a message was received over SMTP with TLS, Exim failed to clear up
1946 the incoming connection properly after forking off the child process to
1947 deliver the message. In some situations the subsequent outgoing
1948 delivery connection happened to have the same fd number as the incoming
1949 connection previously had. Exim would try to use TLS and fail, logging
1950 a "Bad file descriptor" error.
1952 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1953 utilities have not been installed.
1955 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1956 temporary space as the ACL may create new global variables.
1958 TL/01 LDAP support uses per connection or global context settings, depending
1959 upon the detected version of the libraries at build time.
1961 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1962 to extract and use the src ip:port in logging and expansions as if it
1963 were a direct connection from the outside internet. PPv2 support was
1964 updated based on HAProxy spec change in May 2014.
1966 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1968 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1969 Properly escape header and check for NULL return.
1971 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1974 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1976 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1977 characters in header names, implemented as a verify condition.
1978 Contributed by Michael Fischer v. Mollard.
1980 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1981 results permerror and temperror. Previous values are deprecated but
1982 still accepted. In a future release, err_perm and err_temp will be
1983 completely removed, which will be a backward incompatibility if the
1984 ACL tests for either of these two old results. Patch contributed by
1985 user bes-internal on the mailing list.
1987 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1989 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1990 selectors, in both main and reject logs.
1992 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1995 JH/07 Add malware type "sock" for talking to simple daemon.
1997 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1999 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
2000 routers/transports under cutthrough routing.
2002 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
2003 numbers. Touch up "bool" conditional to keep the same definition.
2005 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
2007 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
2009 JH/12 Expand items in router/transport headers_add or headers_remove lists
2010 individually rather than the list as a whole. Bug 1452.
2012 Required for reasonable handling of multiple headers_ options when
2013 they may be empty; requires that headers_remove items with embedded
2014 colons must have them doubled (or the list-separator changed).
2016 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
2017 view the policy declared in the DMARC record. Currently, $dmarc_status
2018 is a combined value of both the record presence and the result of the
2021 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
2023 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
2024 dnslookup router and the smtp transport (applying to the forward
2027 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
2028 of ldap servers used for a specific lookup. Patch provided by Heiko
2031 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
2032 New variable $lookup_dnssec_authenticated for observability.
2034 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
2035 Patch submitted by Lars Timman.
2037 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
2039 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
2040 Requires trusted mode and valid format message id, aborts otherwise.
2041 Patch contributed by Heiko Schlichting.
2043 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
2044 certextract with support for various fields. Bug 1358.
2046 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
2047 is requested by default, modifiable by smtp transport option
2050 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
2051 operate on certificate variables to give certificate fingerprints
2052 Also new ${sha256:cert_variable}.
2054 JH/23 The PRDR feature is moved from being Experimental into the mainline.
2056 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
2057 Christian Aistleitner.
2059 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
2061 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
2062 file. Patch from Wolfgang Breyha.
2064 JH/25 Expand the coverage of the delivery $host and $host_address to
2065 client authenticators run in verify callout. Bug 1476.
2067 JH/26 Port service names are now accepted for tls_on_connect_ports, to
2068 align with daemon_smtp_ports. Bug 72.
2070 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
2071 support and error reporting did not work properly.
2073 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
2074 and is readable. Patch from Andrew Colin Kissa.
2076 TL/14 Enhance documentation of ${run expansion and how it parses the
2077 commandline after expansion, particularly in the case when an
2078 unquoted variable expansion results in an empty value.
2080 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
2082 PP/02 Fix internal collision of T_APL on systems which support RFC3123
2083 by renaming away from it. Addresses GH issue 15, reported by
2086 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
2088 TL/15 SECURITY: prevent double expansion in math comparison functions
2089 (can expand unsanitized data). Not remotely exploitable.
2096 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
2098 PP/02 Make -n do something, by making it not do something.
2099 When combined with -bP, the name of an option is not output.
2101 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
2104 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
2105 $sender_host_name and config options to manage this, and basic check
2108 PP/05 DSCP support for outbound connections and control modifier for inbound.
2110 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
2111 (Only plugin which currently uses this is kerberos4, which nobody should
2112 be using, but we should make it available and other future plugins might
2113 conceivably use it, even though it would break NAT; stuff *should* be
2114 using channel bindings instead).
2116 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
2117 name; added for Sendmail compatibility; requires admin caller.
2118 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
2119 just ignore it); requires trusted caller.
2120 Also parse but ignore: -Ac -Am -X<logfile>
2123 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
2125 TL/02 Add +smtp_confirmation as a default logging option.
2127 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
2128 Patch by Magnus Holmgren from 2007-02-20.
2130 TL/04 Bugzilla 1281 - Spec typo.
2131 Bugzilla 1283 - Spec typo.
2132 Bugzilla 1290 - Spec grammar fixes.
2134 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
2136 TL/06 Add Experimental DMARC support using libopendmarc libraries.
2138 TL/07 Fix an out of order global option causing a segfault. Reported to dev
2139 mailing list by by Dmitry Isaikin.
2141 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
2143 JH/02 Support "G" suffix to numbers in ${if comparisons.
2145 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
2147 NM/01 Bugzilla 1197 - Spec typo
2148 Bugzilla 1196 - Spec examples corrections
2150 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
2152 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
2153 gnutls_enable_pkcs11, but renamed to more accurately indicate its
2156 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
2157 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
2159 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
2160 "acl {{name}{arg}...}", and optional args on acl condition
2163 JH/05 Permit multiple router/transport headers_add/remove lines.
2165 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
2167 JH/07 Avoid using a waiting database for a single-message-only transport.
2168 Performance patch from Paul Fisher. Bugzilla 1262.
2170 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
2173 JH/09 Add $headers_added variable, with content from use of ACL modifier
2174 add_header (but not yet added to the message). Bugzilla 199.
2176 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
2177 Pulled from Bugzilla 817 by Wolfgang Breyha.
2179 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
2181 (nb: this is the same fix as in Exim 4.80.1)
2183 JH/11 Add A= logging on delivery lines, and a client_set_id option on
2186 JH/12 Add optional authenticated_sender logging to A= and a log_selector
2189 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
2191 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
2192 advertise SMTP AUTH mechanism to us, instead of a generic
2193 protocol violation error. Also, make Exim more robust to bad
2194 data from the Dovecot auth socket.
2196 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
2198 When a queue runner is handling a message, Exim first routes the
2199 recipient addresses, during which it prunes them based on the retry
2200 hints database. After that it attempts to deliver the message to
2201 any remaining recipients. It then updates the hints database using
2204 So if a recipient address works intermittently, it can get repeatedly
2205 deferred at routing time. The retry hints record remains fresh so the
2206 address never reaches the final cutoff time.
2208 This is a fairly common occurrence when a user is bumping up against
2209 their storage quota. Exim had some logic in its local delivery code
2210 to deal with this. However it did not apply to per-recipient defers
2211 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
2213 This change adds a proper retry rule check during routing so that the
2214 final cutoff time is checked against the message's age. We only do
2215 this check if there is an address retry record and there is not a
2216 domain retry record; this implies that previous attempts to handle
2217 the address had the retry_use_local_parts option turned on. We use
2218 this as an approximation for the destination being like a local
2219 delivery, as in LMTP.
2221 I suspect this new check makes the old local delivery cutoff check
2222 redundant, but I have not verified this so I left the code in place.
2224 TF/02 Correct gecos expansion when From: is a prefix of the username.
2226 Test 0254 submits a message to Exim with the header
2230 When I ran the test suite under the user fanf2, Exim expanded
2231 the header to contain my full name, whereas it should have added
2232 a Resent-Sender: header. It erroneously treats any prefix of the
2233 username as equal to the username.
2235 This change corrects that bug.
2237 GF/01 DCC debug and logging tidyup
2238 Error conditions log to paniclog rather than rejectlog.
2239 Debug lines prefixed by "DCC: " to remove any ambiguity.
2241 TF/03 Avoid unnecessary rebuilds of lookup-related code.
2243 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
2244 Bug spotted by Jeremy Harris; was flawed since initial commit.
2245 Would have resulted in OCSP responses post-SNI triggering an Exim
2246 NULL dereference and crash.
2248 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
2250 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
2251 Bug detection, analysis and fix by Samuel Thibault.
2252 Bugzilla 1331, Debian bug #698092.
2254 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
2256 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
2257 Server implementation by Todd Lyons, client by JH.
2258 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
2259 config variable "prdr_enable" controls whether the server
2260 advertises the facility. If the client requests PRDR a new
2261 acl_data_smtp_prdr ACL is called once for each recipient, after
2262 the body content is received and before the acl_smtp_data ACL.
2263 The client is controlled by both of: a hosts_try_prdr option
2264 on the smtp transport, and the server advertisement.
2265 Default client logging of deliveries and rejections involving
2266 PRDR are flagged with the string "PRDR".
2268 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
2269 fclose(). Diagnosis by Todd Lyons.
2271 PP/17 Update configure.default to handle IPv6 localhost better.
2272 Patch by Alain Williams (plus minor tweaks).
2275 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
2276 This is now consistent with GnuTLS, and is now documented: the
2277 previous undocumented portable approach to treating the option as
2278 unset was to force an expansion failure. That still works, and
2279 an empty string is now equivalent.
2281 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
2282 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
2283 not performing validation itself.
2285 PP/20 Added force_command boolean option to pipe transport.
2286 Patch from Nick Koston, of cPanel Inc.
2288 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
2291 TF/04 Added udpsend ACL modifier and hexquote expansion operator
2293 PP/21 Fix eximon continuous updating with timestamped log-files.
2294 Broken in a format-string cleanup in 4.80, missed when I repaired the
2295 other false fix of the same issue.
2296 Report and fix from Heiko Schlichting.
2299 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
2300 Report from Prashanth Katuri.
2302 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
2303 It's SecureTransport, so affects any MacOS clients which use the
2304 system-integrated TLS libraries, including email clients.
2306 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
2307 using a MIME ACL for non-SMTP local injection.
2308 Report and assistance in diagnosis by Warren Baker.
2310 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
2312 JH/16 Fix comparisons for 64b. Bugzilla 1385.
2314 TL/09 Add expansion variable $authenticated_fail_id to keep track of
2315 last id that failed so it may be referenced in subsequent ACL's.
2317 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
2320 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
2321 ldap library initialization, allowing self-signed CA's to be
2322 used. Also properly sets require_cert option later in code by
2323 using NULL (global ldap config) instead of ldap handle (per
2324 session). Bug diagnosis and testing by alxgomz.
2326 TL/12 Enhanced documentation in the ratelimit.pl script provided in
2327 the src/util/ subdirectory.
2329 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
2330 renamed to Transport Post Delivery Action by Jeremy Harris, as
2333 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
2334 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
2335 redis_servers = needs to be configured which will be used by the redis
2336 lookup. Patch from Warren Baker, of The Packet Hub.
2338 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
2340 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
2341 hostname or reverse DNS when processing a host list. Used suggestions
2342 from multiple comments on this bug.
2344 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
2346 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
2347 Missed a few lines, added it to make the runtest require no keyboard
2350 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
2351 contains upper case chars. Make router use caseful_local_part.
2353 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
2354 support when GnuTLS has been built with p11-kit.
2360 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
2362 This, or similar/improved, will also be change PP/11 of 4.82.
2368 PP/01 Handle short writes when writing local log-files.
2369 In practice, only affects FreeBSD (8 onwards).
2370 Bugzilla 1053, with thanks to Dmitry Isaikin.
2372 NM/01 Bugzilla 949 - Documentation tweak
2374 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
2377 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
2379 PP/02 Implemented gsasl authenticator.
2381 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
2383 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
2384 `pkg-config foo` for cflags/libs.
2386 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
2387 with rest of GSASL and with heimdal_gssapi.
2389 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
2390 `pkg-config foo` for cflags/libs for the TLS implementation.
2392 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
2393 properties get this fed in as external SSF. A number of robustness
2394 and debugging improvements to the cyrus_sasl authenticator.
2396 PP/08 cyrus_sasl server now expands the server_realm option.
2398 PP/09 Bugzilla 1214 - Log authentication information in reject log.
2399 Patch by Jeremy Harris.
2401 PP/10 Added dbmjz lookup type.
2403 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
2405 PP/12 MAIL args handles TAB as well as SP, for better interop with
2406 non-compliant senders.
2407 Analysis and variant patch by Todd Lyons.
2409 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
2410 Bug report from Lars Müller <lars@samba.org> (via SUSE),
2411 Patch from Dirk Mueller <dmueller@suse.com>
2413 PP/13 tls_peerdn now print-escaped for spool files.
2414 Observed some $tls_peerdn in wild which contained \n, which resulted
2415 in spool file corruption.
2417 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
2418 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
2419 or write after TLS renegotiation, which otherwise led to messages
2422 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
2423 as a tracking header (ie: a signed header comes before the signature).
2424 Patch from Wolfgang Breyha.
2426 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
2427 comma-sep list; embedded commas doubled.
2429 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
2431 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
2433 Report and patch from Dmitry Banschikov.
2435 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
2436 Removed SSL_clear() after SSL_new() which led to protocol negotiation
2437 failures. We appear to now support TLS1.1+ with Exim.
2439 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
2440 lets Exim select keys and certificates based upon TLS SNI from client.
2441 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
2442 before an outbound SMTP session. New log_selector, +tls_sni.
2444 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
2445 NULL dereference. Report and patch from Alun Jones.
2447 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
2448 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
2449 Not seeing resolver debug output on NetBSD, but suspect this is a
2450 resolver implementation change.
2452 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
2453 Left warnings. Added "eximon gdb" invocation mode.
2455 PP/21 Defaulting "accept_8bitmime" to true, not false.
2457 PP/22 Added -bw for inetd wait mode support.
2459 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
2460 locate the relevant includes and libraries. Made this the default.
2462 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
2463 Bugzilla 1246, report and most of solution from Tomasz Kusy.
2465 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
2466 This may cause build issues on older platforms.
2468 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
2469 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
2470 gnutls_require_mac & gnutls_require_protocols (no longer supported).
2471 Added SNI support via GnuTLS too.
2472 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
2474 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
2476 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
2477 Applied second patch from Janne, implementing suggestion to default
2478 multiple-strings-in-record handling to match SPF spec.
2480 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
2482 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
2483 read-only, out of scope).
2484 Patch from Wolfgang Breyha, report from Stuart Northfield.
2486 PP/29 Fix three issues highlighted by clang analyser static analysis.
2487 Only crash-plausible issue would require the Cambridge-specific
2488 iplookup router and a misconfiguration.
2489 Report from Marcin Mirosław.
2491 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
2493 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
2494 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
2495 As part of this, removing so much warning spew let me fix some minor
2496 real issues in debug logging.
2498 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
2499 assignment on my part. Fixed.
2501 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
2502 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
2503 Janne Snabb (who went above and beyond: thank you).
2505 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
2506 string otherwise requires a connection and a bunch more work and it's
2507 relatively easy to get wrong. Should also expose TLS library linkage
2510 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
2511 64-bit ${eval} (JH/03).
2513 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
2514 GNU libc to support some of the 64-bit stuff, should not lead to
2515 conflicts. Defined before os.h is pulled in, so if a given platform
2516 needs to override this, it can.
2518 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
2519 protection layer was required, which is not implemented.
2520 Bugzilla 1254, patch from Wolfgang Breyha.
2522 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
2523 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
2524 tls_dhparam take prime identifiers. Also unbreak combination of
2525 OpenSSL+DH_params+TLSSNI.
2527 PP/39 Disable SSLv2 by default in OpenSSL support.
2533 PP/01 Solaris build fix for Oracle's LDAP libraries.
2534 Bugzilla 1109, patch from Stephen Usher.
2536 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
2538 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
2541 TF/02 Fix a couple more cases where we did not log the error message
2542 when unlink() failed. See also change 4.74-TF/03.
2544 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
2545 lock up or crash if it happened to be inside a call to libc when it
2546 got a SIGUSR1 from exiwhat.
2548 The SIGUSR1 handler appends the current process status to the process
2549 log which is later printed by exiwhat. It used to use the general
2550 purpose logging code to do this, but several functions it calls are
2551 not safe for signals.
2553 The new output code in the SIGUSR1 handler is specific to the process
2554 log, and simple enough that it's easy to inspect for signal safety.
2555 Removing some special cases also simplifies the general logging code.
2556 Removing the spurious timestamps from the process log simplifies
2559 TF/04 Improved ratelimit ACL condition.
2561 The /noupdate option has been deprecated in favour of /readonly which
2562 has clearer semantics. The /leaky, /strict, and /readonly update modes
2563 are mutually exclusive. The update mode is no longer included in the
2564 database key; it just determines when the database is updated. (This
2565 means that when you upgrade Exim will forget old rate measurements.)
2567 Exim now checks that the per_* options are used with an update mode that
2568 makes sense for the current ACL. For example, when Exim is processing a
2569 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
2570 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
2571 must specify per_mail/readonly. If you omit the update mode it defaults to
2572 /leaky where that makes sense (as before) or /readonly where required.
2574 The /noupdate option is now undocumented but still supported for
2575 backwards compatibility. It is equivalent to /readonly except that in
2576 ACLs where /readonly is required you may specify /leaky/noupdate or
2577 /strict/noupdate which are treated the same as /readonly.
2579 A useful new feature is the /count= option. This is a generalization
2580 of the per_byte option, so that you can measure the throughput of other
2581 aggregate values. For example, the per_byte option is now equivalent
2582 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
2584 The per_rcpt option has been generalized using the /count= mechanism
2585 (though it's more complicated than the per_byte equivalence). When it is
2586 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
2587 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
2588 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
2589 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
2590 Note that using per_rcpt with a non-readonly update mode in more than
2591 one ACL will cause the recipients to be double-counted. (The per_mail
2592 and per_byte options don't have this problem.)
2594 The handling of very low rates has changed slightly. If the computed rate
2595 is less than the event's count (usually one) then this event is the first
2596 after a long gap. In this case the rate is set to the same as this event's
2597 count, so that the first message of a spam run is counted properly.
2599 The major new feature is a mechanism for counting the rate of unique
2600 events. The new per_addr option counts the number of different
2601 recipients that someone has sent messages to in the last time period. It
2602 behaves like per_rcpt if all the recipient addresses are different, but
2603 duplicate recipient addresses do not increase the measured rate. Like
2604 the /count= option this is a general mechanism, so the per_addr option
2605 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
2606 example, measure the rate that a client uses different sender addresses
2607 with the options per_mail/unique=$sender_address. There are further
2608 details in the main documentation.
2610 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
2612 TF/06 Removed a few PCRE remnants.
2614 TF/07 Automatically extract Exim's version number from tags in the git
2615 repository when doing development or release builds.
2617 PP/02 Raise smtp_cmd_buffer_size to 16kB.
2618 Bugzilla 879. Patch from Paul Fisher.
2620 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
2621 Heavily based on revision 40f9a89a from Simon Arlott's tree.
2624 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
2626 PP/05 Variable $av_failed, true if the AV scanner deferred.
2627 Bugzilla 1078. Patch from John Horne.
2629 PP/06 Stop make process more reliably on build failure.
2630 Bugzilla 1087. Patch from Heiko Schlittermann.
2632 PP/07 Make maildir_use_size_file an _expandable_ boolean.
2633 Bugzilla 1089. Patch from Heiko Schlittermann.
2635 PP/08 Handle ${run} returning more data than OS pipe buffer size.
2636 Bugzilla 1131. Patch from Holger Weiß.
2638 PP/09 Handle IPv6 addresses with SPF.
2639 Bugzilla 860. Patch from Wolfgang Breyha.
2641 PP/10 GnuTLS: support TLS 1.2 & 1.1.
2643 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
2646 PP/11 match_* no longer expand right-hand-side by default.
2647 New compile-time build option, EXPAND_LISTMATCH_RHS.
2648 New expansion conditions, "inlist", "inlisti".
2650 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
2652 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
2654 PP/14 fix log_write() format string regression from TF/03.
2655 Bugzilla 1152. Patch from Dmitry Isaikin.
2661 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
2663 PP/02 Harmonised TLS library version reporting; only show if debugging.
2664 Layout now matches that introduced for other libraries in 4.74 PP/03.
2666 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
2668 PP/04 New "dns_use_edns0" global option.
2670 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
2673 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
2674 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
2676 TK/01 Updated PolarSSL code to 0.14.2.
2677 Bugzilla 1097. Patch from Andreas Metzler.
2679 PP/07 Catch divide-by-zero in ${eval:...}.
2680 Fixes bugzilla 1102.
2682 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
2685 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
2686 format-string attack -- SECURITY: remote arbitrary code execution.
2688 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
2689 time unintentionally subject to list matching rules, letting the header
2690 cause arbitrary Exim lookups (of items which can occur in lists, *not*
2691 arbitrary string expansion). This allowed for information disclosure.
2693 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
2694 INT_MIN/-1 -- value coerced to INT_MAX.
2700 NM/01 Workaround for PCRE version dependency in version reporting
2703 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
2704 This fixes portability to compilers other than gcc, notably
2705 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
2707 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
2708 makefiles for portability to HP-UX and POSIX correctness.
2710 PP/01 Permit LOOKUP_foo enabling on the make command-line.
2711 Also via indented variable definition in the Makefile.
2712 (Debugging by Oliver Heesakkers).
2714 PP/02 Restore caching of spamd results with expanded spamd_address.
2715 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
2717 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
2718 Improves build reliability. Fix from: Frank Elsner
2720 NM/02 Fix wide character breakage in the rfc2047 coding
2721 Fixes bug 1064. Patch from Andrey N. Oktyabrski
2723 NM/03 Allow underscore in dnslist lookups
2724 Fixes bug 1026. Patch from Graeme Fowler
2726 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
2727 Code patches from Adam Ciarcinski of NetBSD.
2729 NM/04 Fixed exiqgrep to cope with mailq missing size issue
2732 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
2733 is logged, to avoid truncation. Patch from John Horne.
2735 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
2736 Patch from Jakob Hirsch.
2738 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
2739 SQL string expansion failure details.
2740 Patch from Andrey Oktyabrski.
2742 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
2743 Patch from Simon Arlott.
2745 PP/09 New lookups functionality failed to compile on old gcc which rejects
2746 extern declarations in function scope.
2747 Patch from Oliver Fleischmann
2749 PP/10 Use sig_atomic_t for flags set from signal handlers.
2750 Check getgroups() return and improve debugging.
2751 Fixed developed for diagnosis in bug 927 (which turned out to be
2754 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
2755 Patch from Mark Zealey.
2757 PP/12 Bugzilla 1056: Improved spamd server selection.
2758 Patch from Mark Zealey.
2760 PP/13 Bugzilla 1086: Deal with maildir quota file races.
2761 Based on patch from Heiko Schlittermann.
2763 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
2764 Patch from Uwe Doering, sign-off by Michael Haardt.
2766 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
2767 variable declaration deep within a block. Bug and patch from
2770 PP/15 lookups-Makefile IRIX compatibility coercion.
2772 PP/16 Make DISABLE_DKIM build knob functional.
2774 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
2775 Patch by Simon Arlott
2777 TF/03 Fix valgrind.h portability to C89 compilers that do not support
2778 variable argument macros. Our copy now differs from upstream.
2784 TF/01 Failure to get a lock on a hints database can have serious
2785 consequences so log it to the panic log.
2787 TF/02 Log LMTP confirmation messages in the same way as SMTP,
2788 controlled using the smtp_confirmation log selector.
2790 TF/03 Include the error message when we fail to unlink a spool file.
2792 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
2793 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
2794 for maintaining out-of-tree patches for some time.
2796 PP/01 Bugzilla 139: Documentation and portability issues.
2797 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
2798 Handle per-OS dynamic-module compilation flags.
2800 PP/02 Let /dev/null have normal permissions.
2801 The 4.73 fixes were a little too stringent and complained about the
2802 permissions on /dev/null. Exempt it from some checks.
2803 Reported by Andreas M. Kirchwitz.
2805 PP/03 Report version information for many libraries, including
2806 Exim version information for dynamically loaded libraries. Created
2807 version.h, now support a version extension string for distributors
2808 who patch heavily. Dynamic module ABI change.
2810 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
2811 privilege escalation vulnerability whereby the Exim run-time user
2812 can cause root to append content of the attacker's choosing to
2815 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
2818 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
2819 If dropping privileges for untrusted macros, we disabled normal logging
2820 on the basis that it would fail; for the Exim run-time user, this is not
2821 the case, and it resulted in successful deliveries going unlogged.
2822 Fixed. Reported by Andreas Metzler.
2828 PP/01 Date: & Message-Id: revert to normally being appended to a message,
2829 only prepend for the Resent-* case. Fixes regression introduced in
2830 Exim 4.70 by NM/22 for Bugzilla 607.
2832 PP/02 Include check_rfc2047_length in configure.default because we're seeing
2833 increasing numbers of administrators be bitten by this.
2835 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
2837 PP/03 Bugzilla 994: added openssl_options main configuration option.
2839 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
2841 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
2843 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
2845 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
2846 without a peer certificate, leading to a segfault because of an
2847 assumption that peers always have certificates. Be a little more
2848 paranoid. Problem reported by Martin Tscholak.
2850 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
2851 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
2852 NB: ClamAV planning to remove STREAM in "middle of 2010".
2853 CL also introduces -bmalware, various -d+acl logging additions and
2854 more caution in buffer sizes.
2856 PP/09 Implemented reverse_ip expansion operator.
2858 PP/10 Bugzilla 937: provide a "debug" ACL control.
2860 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2862 PP/12 Bugzilla 973: Implement --version.
2864 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2866 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2868 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2870 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2871 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2872 ignore trailing whitespace.
2874 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2876 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2879 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2880 Notification from Dr Andrew Aitchison.
2882 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2883 ExtendedDetectionInfo response format.
2884 Notification from John Horne.
2886 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2889 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2890 XSL and documented dependency on system catalogs, with examples of how
2893 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2896 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2897 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2898 configuration file which is writeable by the Exim user or group.
2900 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2901 of configuration files to cover files specified with the -C option if
2902 they are going to be used with root privileges, not just the default
2905 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2906 option (effectively making it always true).
2908 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2909 files to be used while preserving root privileges.
2911 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2912 that rogue child processes cannot use them.
2914 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2915 run-time user, instead of root.
2917 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2918 Exim run-time user without dropping privileges.
2920 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2921 result string, instead of calling string_vformat() twice with the same
2924 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2925 for other users. Others should always drop root privileges if they use
2926 -C on the command line, even for a whitelisted configure file.
2928 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2930 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2936 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2937 $data_path, and $header_path variables; fixed documentation bugs and
2940 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2941 exipick to access non-standard spools, including the "frozen" queue
2944 NM/01 Bugzilla 965: Support mysql stored procedures.
2945 Patch from Alain Williams
2947 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2949 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2950 Patch from Andreas Metzler
2952 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2953 Patch from Kirill Miazine
2955 NM/05 Bugzilla 671: Added umask to procmail example.
2957 JJ/03 installed exipick 20100323.0, fixing doc bug
2959 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2960 directory. Notification and patch from Dan Rosenberg.
2962 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2964 TK/02 Improve log output when DKIM signing operation fails.
2966 MH/01 Treat the transport option dkim_domain as a colon separated
2967 list, not as a single string, and sign the message with each element,
2968 omitting multiple occurences of the same signer.
2970 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2971 Bugzilla 985, 986. Patch by Simon Arlott
2973 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2974 Patch by Simon Arlott
2976 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2977 MBX locking. Notification from Dan Rosenberg.
2983 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2985 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2987 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2989 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2991 NM/04 Bugzilla 915: Flag broken perl installation during build.
2997 TK/01 Added patch by Johannes Berg that expands the main option
2998 "spamd_address" if it starts with a dollar sign.
3000 TK/02 Write list of recipients to X-Envelope-Sender header when building
3001 the mbox-format spool file for content scanning (suggested by Jakob
3004 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
3005 (http://www.dcc-servers.net/) support via dccifd. Activated by
3006 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
3008 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
3009 by Mark Daniel Reidel <mr@df.eu>.
3011 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
3012 When building exim an external PCRE library is now needed -
3013 PCRE is a system library on the majority of modern systems.
3014 See entry on PCRE_LIBS in EDITME file.
3016 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
3017 conversation. Added nologin parameter to request.
3018 Patch contributed by Kirill Miazine.
3020 TF/01 Do not log submission mode rewrites if they do not change the address.
3022 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
3024 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
3025 log files in place. Contributed by Roberto Lima.
3027 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
3029 TF/03 Bugzilla 615: When checking the local_parts router precondition
3030 after a local_part_suffix or local_part_prefix option, Exim now
3031 does not use the address's named list lookup cache, since this
3032 contains cached lookups for the whole local part.
3034 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
3035 Robert Millan. Documentation is in experimental-spec.txt.
3037 TF/04 Bugzilla 668: Fix parallel build (make -j).
3039 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
3041 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
3042 Patch provided by Jan Srzednicki.
3044 TF/05 Leading white space used to be stripped from $spam_report which
3045 wrecked the formatting. Now it is preserved.
3047 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
3048 that they are available at delivery time.
3050 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
3052 TF/08 TLS error reporting now respects the incoming_interface and
3053 incoming_port log selectors.
3055 TF/09 Produce a more useful error message if an SMTP transport's hosts
3056 setting expands to an empty string.
3058 NM/06 Bugzilla 744: EXPN did not work under TLS.
3059 Patch provided by Phil Pennock.
3061 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
3062 Patch provided by Richard Godbee.
3064 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
3065 acl_smtp_notquit, added index entry.
3067 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
3068 Patch provided by Eugene Bujak.
3070 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
3071 accept(). Patch provided by Maxim Dounin.
3073 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
3074 Patch provided by Phil Pennock.
3076 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
3078 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
3079 Patch provided by Brad "anomie" Jorsch.
3081 NM/14 Bugzilla 622: Added timeout setting to transport filter.
3082 Patch provided by Dean Brooks.
3084 TK/05 Add native DKIM support (does not depend on external libraries).
3086 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
3087 Patch provided by Graeme Fowler.
3089 NM/16 Bugzilla 851: Documentation example syntax fix.
3091 NM/17 Changed NOTICE file to remove references to embedded PCRE.
3093 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
3096 NM/19 Bugzilla 745: TLS version reporting.
3097 Patch provided by Phil Pennock.
3099 NM/20 Bugzilla 167: bool: condition support.
3100 Patch provided by Phil Pennock.
3102 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
3103 clients. Patch provided by Phil Pennock.
3105 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
3106 Patch provided by Brad "anomie" Jorsch.
3108 NM/23 Bugzilla 687: Fix misparses in eximstats.
3109 Patch provided by Heiko Schlittermann.
3111 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
3112 Patch provided by Heiko Schlittermann.
3114 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
3115 plus update to original patch.
3117 NM/26 Bugzilla 799: Documentation correction for ratelimit.
3119 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
3120 Patch provided by David Brownlee.
3122 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
3124 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
3126 NM/30 Bugzilla 888: TLS documentation bugfixes.
3128 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
3130 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
3131 Unlike the original bugzilla I have changed all shell scripts in src tree.
3133 NM/33 Bugzilla 898: Transport filter timeout fix.
3134 Patch by Todd Rinaldo.
3136 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
3137 Patch by Serge Demonchaux.
3139 NM/35 Bugzilla 39: Base64 decode bug fixes.
3140 Patch by Jakob Hirsch.
3142 NM/36 Bugzilla 909: Correct connect() call in dcc code.
3144 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
3146 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
3148 NM/39 Bugzilla 911: Fixed MakeLinks build script.
3154 TK/01 Add preliminary DKIM support. Currently requires a forked version of
3155 ALT-N's libdkim that I have put here:
3156 http://duncanthrax.net/exim-experimental/
3158 Note to Michael Haardt: I had to rename some vars in sieve.c. They
3159 were called 'true' and it seems that C99 defines that as a reserved
3160 keyword to be used with 'bool' variable types. That means you could
3161 not include C99-style headers which use bools without triggering
3162 build errors in sieve.c.
3164 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
3165 as mailq or other aliases. Changed the --help handling significantly
3166 to do whats expected. exim_usage() emits usage/help information.
3168 SC/01 Added the -bylocaldomain option to eximstats.
3170 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
3172 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
3174 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
3180 PH/01 Another patch from the Sieve maintainer.
3182 PH/02 When an IPv6 address is converted to a string for single-key lookup
3183 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
3184 dots are used instead of colons so that keys in lsearch files need not
3185 contain colons. This was done some time before quoting was made available
3186 in lsearch files. However, iplsearch files do require colons in IPv6 keys
3187 (notated using the quote facility) so as to distinguish them from IPv4
3188 keys. This meant that lookups for IP addresses in host lists did not work
3189 for iplsearch lookups.
3191 This has been fixed by arranging for IPv6 addresses to be expressed with
3192 colons if the lookup type is iplsearch. This is not incompatible, because
3193 previously such lookups could never work.
3195 The situation is now rather anomalous, since one *can* have colons in
3196 ordinary lsearch keys. However, making the change in all cases is
3197 incompatible and would probably break a number of configurations.
3199 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
3202 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
3203 conversion specification without a maximum field width, thereby enabling
3204 a rogue spamd server to cause a buffer overflow. While nobody in their
3205 right mind would setup Exim to query an untrusted spamd server, an
3206 attacker that gains access to a server running spamd could potentially
3207 exploit this vulnerability to run arbitrary code as the Exim user.
3209 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
3210 $primary_hostname instead of what libspf2 thinks the hosts name is.
3212 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
3213 a directory entry by the name of the lookup key. Previously, if a
3214 symlink pointed to a non-existing file or a file in a directory that
3215 Exim lacked permissions to read, a lookup for a key matching that
3216 symlink would fail. Now it is enough that a matching directory entry
3217 exists, symlink or not. (Bugzilla 503.)
3219 PH/03 The body_linecount and body_zerocount variables are now exported in the
3222 PH/04 Added the $dnslist_matched variable.
3224 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
3225 This means they are set thereafter only if the connection becomes
3228 PH/06 Added the client_condition to authenticators so that some can be skipped
3229 by clients under certain conditions.
3231 PH/07 The error message for a badly-placed control=no_multiline_responses left
3232 "_responses" off the end of the name.
3234 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
3236 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
3237 (without spaces) instead of just copying the configuration text.
3239 PH/10 Added the /noupdate option to the ratelimit ACL condition.
3241 PH/11 Added $max_received_linelength.
3243 PH/12 Added +ignore_defer and +include_defer to host lists.
3245 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
3246 way in which PCRE > 7.0 is built.
3248 PH/14 Implemented queue_only_load_latch.
3250 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
3251 MAIL command. The effect was to mangle the value on 64-bit systems.
3253 PH/16 Another patch from the Sieve maintainer.
3255 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
3257 PH/18 If a system quota error occurred while trying to create the file for
3258 a maildir delivery, the message "Mailbox is full" was not appended to the
3259 bounce if the delivery eventually timed out. Change 4.67/27 below applied
3260 only to a quota excession during the actual writing of the file.
3262 PH/19 It seems that peer DN values may contain newlines (and other non-printing
3263 characters?) which causes problems in log lines. The DN values are now
3264 passed through string_printing() before being added to log lines.
3266 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
3267 and InterBase are left for another time.)
3269 PH/21 Added message_body_newlines option.
3271 PH/22 Guard against possible overflow in moan_check_errorcopy().
3273 PH/23 POSIX allows open() to be a macro; guard against that.
3275 PH/24 If the recipient of an error message contained an @ in the local part
3276 (suitably quoted, of course), incorrect values were put in $domain and
3277 $local_part during the evaluation of errors_copy.
3283 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
3284 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
3287 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
3288 issue a MAIL command.
3290 PH/02 In an ACL statement such as
3292 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
3294 if a client was not listed at all, or was listed with a value other than
3295 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
3296 the condition was not true (as it should be), so access was not denied.
3297 The bug was that the ! inversion was incorrectly passed on to the second
3298 item. This has been fixed.
3300 PH/03 Added additional dnslists conditions == and =& which are different from
3301 = and & when the dns lookup returns more than one IP address.
3303 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
3304 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
3306 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
3307 FSYNC, which compiles an option called disable_fsync that allows for
3308 bypassing fsync(). The documentation is heavily laced with warnings.
3310 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
3312 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
3313 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
3314 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
3315 including adding "make clean"; (3) Added -fPIC when compiling the test
3316 dynamically loaded module, to get rid of a warning.
3318 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
3319 message fails, move_frozen_messages = true and ignore_bounce_errors_after
3320 = 0s. The bug is otherwise harmless.
3322 PH/07 There was a bug in the dovecot authenticator such that the value of
3323 $auth1 could be overwritten, and so not correctly preserved, after a
3324 successful authentication. This usually meant that the value preserved by
3325 the server_setid option was incorrect.
3327 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
3329 PH/09 Installed PCRE release 7.0.
3331 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
3332 run for batched SMTP input. It is now run at the start of every message
3333 in the batch. While fixing this I discovered that the process information
3334 (output by running exiwhat) was not always getting set for -bs and -bS
3335 input. This is fixed, and it now also says "batched" for BSMTP.
3337 PH/11 Added control=no_pipelining.
3339 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
3340 patch, slightly modified), and move the expansion of helo_data till after
3341 the connection is made in the smtp transport (so it can use these
3344 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
3346 PH/14 Added log_selector = +pid.
3348 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
3350 PH/16 Add ${if forany and ${if forall.
3352 PH/17 Added dsn_from option to vary the From: line in DSNs.
3354 PH/18 Flush SMTP output before performing a callout, unless control =
3355 no_callout_flush is set.
3357 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
3358 was true (the default) a successful delivery failed to delete the retry
3359 item, thus causing premature timeout of the address. The bug is now
3362 PH/20 Added hosts_avoid_pipelining to the smtp transport.
3364 PH/21 Long custom messages for fakedefer and fakereject are now split up
3365 into multiline responses in the same way that messages for "deny" and
3366 other ACL rejections are.
3368 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
3369 with slight modification.
3371 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
3372 draft, changing the syntax and factoring some duplicate code".
3374 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
3375 for deliveries of the second and subsequent messages over the same SMTP
3378 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
3379 ${reduce, with only minor "tidies".
3381 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
3383 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
3384 expansion side effects.
3386 PH/27 When a message times out after an over-quota error from an Exim-imposed
3387 quota, the bounce message says "mailbox is full". This message was not
3388 being given when it was a system quota that was exceeded. It now should
3391 MH/03 Made $recipients available in local_scan(). local_scan() already has
3392 better access to the recipient list through recipients_list[], but
3393 $recipients can be useful in postmaster-provided expansion strings.
3395 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
3396 in the case of a MAIL command with additional options following the
3397 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
3398 were accidentally chopped off.
3400 PH/29 SMTP synchronization checks are implemented when a command is read -
3401 there is a check that no more input is waiting when there shouldn't be
3402 any. However, for some commands, a delay in an ACL can mean that it is
3403 some time before the response is written. In this time, more input might
3404 arrive, invalidly. So now there are extra checks after an ACL has run for
3405 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
3406 pipelining has not been advertised.
3408 PH/30 MH's patch to allow iscntrl() characters to be list separators.
3410 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
3411 returned in the SMTP response when smtp_return_error_details was false.
3412 This has been fixed.
3414 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
3415 instead of the C I/O that was originally supplied, because problems were
3416 reported on Solaris.
3418 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
3419 Exim which did not show up earlier: it was assuming that a call to
3420 SSL_CTX_set_info_callback() might give an error value. In fact, there is
3421 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
3422 was a macro that became an assignment, so it seemed to work. This has
3423 changed to a proper function call with a void return, hence the compile
3424 error. Exim's code has been fixed.
3426 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
3429 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
3431 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
3433 PH/37 If a message is not accepted after it has had an id assigned (e.g.
3434 because it turns out to be too big or there is a timeout) there is no
3435 "Completed" line in the log. When some messages of this type were
3436 selected by exigrep, they were listed as "not completed". Others were
3437 picked up by some special patterns. I have improved the selection
3438 criteria to be more general.
3440 PH/38 The host_find_failed option in the manualroute router can now be set
3441 to "ignore", to completely ignore a host whose IP address cannot be
3442 found. If all hosts are ignored, the behaviour is controlled by the new
3443 host_all_ignored option.
3445 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
3446 homing or because of multiple MX records with /mx) generated more than
3447 one IP address, and the following item turned out to be the local host,
3448 all the secondary addresses of the first item were incorrectly removed
3449 from the list, along with the local host and any following hosts (which
3450 is what is supposed to happen).
3452 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
3453 whoever called Exim into the -H file. In the case of the daemon it was
3454 behaving confusingly. When first started, it used values for whoever
3455 started the daemon, but after a SIGHUP it used the Exim user (because it
3456 calls itself on a restart). I have changed the code so that it now always
3459 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
3460 message are rejected with the same error (e.g. no authentication or bad
3461 sender address), and a DATA command is nevertheless sent (as can happen
3462 with PIPELINING or a stupid MUA), the error message that was given to the
3463 RCPT commands is included in the rejection of the DATA command. This is
3464 intended to be helpful for MUAs that show only the final error to their
3467 PH/42 Another patch from the Sieve maintainer.
3469 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
3470 Eximstats - Fixed some broken HTML links and added missing column headers
3472 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
3473 columns for Rejects, Temp Rejects, Ham, and Spam rows.
3475 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
3477 PH/43 Yet another patch from the Sieve maintainer.
3479 PH/44 I found a way to check for a TCP/IP connection going away before sending
3480 the response to the final '.' that terminates a message, but only in the
3481 case where the client has not sent further data following the '.'
3482 (unfortunately, this is allowed). However, in many cases there won't be
3483 any further data because there won't be any more messages to send. A call
3484 to select() can be used: if it shows that the input is "ready", there is
3485 either input waiting, or the socket has been closed. An attempt to read
3486 the next input character can distinguish the two cases. Previously, Exim
3487 would have sent an OK response which the client would never have see.
3488 This could lead to message repetition. This fix should cure that, at
3489 least in a lot of common cases.
3491 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
3492 advertised in response to EHLO.
3498 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
3499 fixed by 4.65/MH/01 (is this a record?) are fixed:
3501 (i) An empty string was always treated as zero by the numeric comparison
3502 operators. This behaviour has been restored.
3504 (ii) It is documented that the numeric comparison operators always treat
3505 their arguments as decimal numbers. This was broken in that numbers
3506 starting with 0 were being interpreted as octal.
3508 While fixing these problems I realized that there was another issue that
3509 hadn't been noticed. Values of message_size_limit (both the global option
3510 and the transport option) were treated as octal if they started with 0.
3511 The documentation was vague. These values are now always treated as
3512 decimal, and I will make that clear in the documentation.
3518 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
3519 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
3522 MH/01 Don't check that the operands of numeric comparison operators are
3523 integers when their expansion is in "skipping" mode (fixes bug
3524 introduced by 4.64-PH/07).
3526 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
3527 child addresses, Exim now panics and dies. Previously, because the count
3528 is held in a short int, deliveries were likely to be lost. As such a
3529 large number of recipients for a single message is ridiculous
3530 (performance will be very, very poor), I have chosen to impose a limit
3531 rather than extend the field.
3537 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
3538 leftover -K file (the existence of which was triggered by #402).
3539 While we were at it, introduced process PID as part of the -K
3540 filename. This should rule out race conditions when creating
3543 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
3544 processing considerably. Previous code took too long for large mails,
3545 triggering a timeout which in turn triggers #401.
3547 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
3548 in the DK code in transports.c. sendfile() is not really portable,
3549 hence the _LINUX specificness.
3551 TF/01 In the add_headers option to the mail command in an Exim filter,
3552 there was a bug that Exim would claim a syntax error in any
3553 header after the first one which had an odd number of characters
3556 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
3557 callout verification, Exim cached a "reject" for the entire domain. This
3558 is correct for most verifications, but it is not correct for a recipient
3559 verification with use_sender or use_postmaster set, because in that case
3560 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
3561 case of MAIL FROM:<> rejection from other early rejections (e.g.
3562 rejection of HELO). When verifying a recipient using a non-null MAIL
3563 address, the cache is ignored if it shows MAIL FROM:<> rejection.
3564 Whatever the result of the callout, the value of the domain cache is
3565 left unchanged (for any other kind of callout, getting as far as trying
3566 RCPT means that the domain itself is ok).
3568 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
3571 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
3572 manifest itself as EPIPE rather than ECONNECT. When tidying away a
3573 session, the daemon ignores ECONNECT errors and logs others; it now
3574 ignores EPIPE as well.
3576 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
3577 (quoted-printable decoding).
3579 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
3580 later the small subsequent patch to fix an introduced bug.
3582 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
3584 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
3586 PH/08 An error is now given if message_size_limit is specified negative.
3588 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
3589 to be given (somewhat) arbitrary names.
3591 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
3594 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
3595 miscellaneous code fixes
3597 PH/10 Added the log_reject_target ACL modifier to specify where to log
3600 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
3601 hostname. This is wrong, because it relates to the incoming message (and
3602 probably the interface on which it is arriving) and not to the outgoing
3603 callout (which could be using a different interface). This has been
3604 changed to use the value of the helo_data option from the smtp transport
3605 instead - this is what is used when a message is actually being sent. If
3606 there is no remote transport (possible with a router that sets up host
3607 addresses), $smtp_active_hostname is used.
3609 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
3610 tweaks were necessary in order to get it to work (see also 21 below):
3611 (a) The code assumed that strncpy() returns a negative number on buffer
3612 overflow, which isn't the case. Replaced with Exim's string_format()
3614 (b) There were several signed/unsigned issues. I just did the minimum
3615 hacking in of casts. There is scope for a larger refactoring.
3616 (c) The code used strcasecmp() which is not a standard C function.
3617 Replaced with Exim's strcmpic() function.
3618 (d) The code set only $1; it now sets $auth1 as well.
3619 (e) A simple test gave the error "authentication client didn't specify
3620 service in request". It would seem that Dovecot has changed its
3621 interface. Fortunately there's a specification; I followed it and
3622 changed what the client sends and it appears to be working now.
3624 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
3627 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
3628 address A is aliased to B and C, where B exists and C does not. Without
3629 -v the output is "A verified" because verification stops after a
3630 successful redirection if more than one address is generated. However,
3631 with -v the child addresses are also verified. Exim was outputting "A
3632 failed to verify" and then showing the successful verification for C,
3633 with its parentage. It now outputs "B failed to verify", showing B's
3634 parentage before showing the successful verification of C.
3636 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
3637 look up a TXT record in a specific list after matching in a combined
3640 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
3641 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
3642 they consult the DNS. I had assumed they would set it the way they
3643 wanted; and indeed my experiments on Linux seem to show that in some
3644 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
3645 To be on the safe side, however, I have now made the interface to
3646 host_find_byname() similar to host_find_bydns(), with an argument
3647 containing the DNS resolver options. The host_find_byname() function now
3648 sets these options at its start, just as host_find_bydns() does. The smtp
3649 transport options dns_qualify_single and dns_search_parents are passed to
3650 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
3651 of host_find_byname() use the default settings of RES_DEFNAMES
3652 (qualify_single) but not RES_DNSRCH (search_parents).
3654 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
3655 spool_read_header() do less string testing, by means of a preliminary
3656 switch on the second character of optional "-foo" lines. (This is
3657 overdue, caused by the large number of possibilities that now exist.
3658 Originally there were few.) While I was there, I also converted the
3659 str(n)cmp tests so they don't re-test the leading "-" and the first
3660 character, in the hope this might squeeze out yet more improvement.
3662 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
3663 flag allowing group syntax was set by the header_syntax check but not
3664 turned off, possible causing trouble later; (2) The flag was not being
3665 set at all for the header_verify test, causing "group"-style headers to
3666 be rejected. I have now set it in this case, and also caused header_
3667 verify to ignore an empty address taken from a group. While doing this, I
3668 came across some other cases where the code for allowing group syntax
3669 while scanning a header line wasn't quite right (mostly, not resetting
3670 the flag correctly in the right place). These bugs could have caused
3671 trouble for malformed header lines. I hope it is now all correct.
3673 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
3674 with the "reply" argument non-NULL. The code, however (which originally
3675 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
3676 but it didn't always do it. This confused somebody who was copying the
3677 code for some other use. I have removed all the tests.
3679 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
3680 feature that was used to support insecure browsers during the U.S. crypto
3681 embargo. It requires special client support, and Exim is probably the
3682 only MTA that supported it -- and would never use it because real RSA is
3683 always available. This code has been removed, because it had the bad
3684 effect of slowing Exim down by computing (never used) parameters for the
3685 RSA_EXPORT functionality.
3687 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
3688 authenticator to fail if there's a tab character in the incoming data
3689 (there should never be unless someone is messing about, as it's supposed
3690 to be base64-encoded). Also added, on Timo's advice, the "secured" option
3691 if the connection is using TLS or if the remote IP is the same as the
3692 local IP, and the "valid-client-cert option" if a client certificate has
3695 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
3696 authenticators. This can be used for authorization after authentication
3697 succeeds. (In the case of plaintext, it servers for both authentication
3700 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
3701 if any retry times were supplied.
3703 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
3704 connection, where there is no client IP address to check. In this
3705 situation, the verify now always succeeds.
3707 PH/25 Applied John Jetmore's -Mset patch.
3709 PH/26 Added -bem to be like -Mset, but loading a message from a file.
3711 PH/27 In a string expansion for a processed (not raw) header when multiple
3712 headers of the same name were present, leading whitespace was being
3713 removed from all of them, but trailing whitespace was being removed only
3714 from the last one. Now trailing whitespace is removed from each header
3715 before concatenation. Completely empty headers in a concatenation (as
3716 before) are ignored.
3718 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
3719 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
3721 PH/29 [Removed. This was a change that I later backed out, and forgot to
3722 correct the ChangeLog entry (that I had efficiently created) before
3723 committing the later change.]
3725 PH/30 Exim was sometimes attempting to deliver messages that had suffered
3726 address errors (4xx response to RCPT) over the same connection as other
3727 messages routed to the same hosts. Such deliveries are always "forced",
3728 so retry times are not inspected. This resulted in far too many retries
3729 for the affected addresses. The effect occurred only when there were more
3730 hosts than the hosts_max_try setting in the smtp transport when it had
3731 the 4xx errors. Those hosts that it had tried were not added to the list
3732 of hosts for which the message was waiting, so if all were tried, there
3733 was no problem. Two fixes have been applied:
3735 (i) If there are any address or message errors in an SMTP delivery, none
3736 of the hosts (tried or untried) are now added to the list of hosts
3737 for which the message is waiting, so the message should not be a
3738 candidate for sending over the same connection that was used for a
3739 successful delivery of some other message. This seems entirely
3740 reasonable: after all the message is NOT "waiting for some host".
3741 This is so "obvious" that I'm not sure why it wasn't done
3742 previously. Hope I haven't missed anything, but it can't do any
3743 harm, as the worst effect is to miss an optimization.
3745 (ii) If, despite (i), such a delivery is accidentally attempted, the
3746 routing retry time is respected, so at least it doesn't keep
3747 hammering the server.
3749 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
3750 in ${readsocket because some servers need this prod.
3752 PH/32 Added some extra debug output when updating a wait-xxx database.
3754 PH/33 The hint "could be header name not terminated by colon", which has been
3755 given for certain expansion errors for a long time, was not being given
3756 for the ${if def:h_colon_omitted{... case.
3758 PH/34 The spec says: "With one important exception, whenever a domain list is
3759 being scanned, $domain contains the subject domain." There was at least
3760 one case where this was not true.
3762 PH/35 The error "getsockname() failed: connection reset by peer" was being
3763 written to the panic log as well as the main log, but it isn't really
3764 panic-worthy as it just means the connection died rather early on. I have
3765 removed the panic log writing for the ECONNRESET error when getsockname()
3768 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
3769 runs only) independently of the message's sender address. This meant
3770 that, if the 4xx error was in fact related to the sender, a different
3771 message to the same recipient with a different sender could confuse
3772 things. In particular, this can happen when sending to a greylisting
3773 server, but other circumstances could also provoke similar problems.
3774 I have changed the default so that the retry time for these errors is now
3775 based a combination of the sender and recipient addresses. This change
3776 can be overridden by setting address_retry_include_sender=false in the
3779 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
3780 remote server are returned as part of bounce messages. This was not
3781 happening for LMTP over a pipe (the lmtp transport), but now it is the
3782 same for both kinds of LMTP.
3784 PH/38 Despite being documented as not happening, Exim was rewriting addresses
3785 in header lines that were in fact CNAMEs. This is no longer the case.
3787 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
3788 and queue runs started by the daemon processed all messages. This has
3789 been fixed so that -R and -S can now usefully be given with -q<time>.
3791 PH/40 Import PCRE release 6.7 (fixes some bugs).
3793 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
3795 PH/42 Give an error if -q is specified more than once.
3797 PH/43 Renamed the variables $interface_address and $interface_port as
3798 $received_ip_address and $received_port, to make it clear that these
3799 values apply to message reception, and not to the outgoing interface when
3800 a message is delivered. (The old names remain recognized, of course.)
3802 PH/44 There was no timeout on the connect() call when using a Unix domain
3803 socket in the ${readsocket expansion. There now is.
3805 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
3806 be meaningful with "accept".
3808 SC/01 Eximstats V1.43
3809 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
3811 SC/02 Eximstats V1.44
3812 Use a glob alias rather than an array ref in the generated
3813 parser. This improves both readability and performance.
3815 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
3816 Collect SpamAssassin and rejection statistics.
3817 Don't display local sender or destination tables unless
3818 there is data to show.
3819 Added average volumes into the top table text output.
3821 SC/04 Eximstats V1.46
3822 Collect data on the number of addresses (recipients)
3823 as well as the number of messages.
3825 SC/05 Eximstats V1.47
3826 Added 'Message too big' to the list of mail rejection
3827 reasons (thanks to Marco Gaiarin).
3829 SC/06 Eximstats V1.48
3830 Mainlog lines which have GMT offsets and are too short to
3831 have a flag are now skipped.
3833 SC/07 Eximstats V1.49 (Alain Williams)
3834 Added the -emptyok flag.
3836 SC/08 Eximstats V1.50
3837 Fixes for obtaining the IP address from reject messages.
3839 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
3840 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
3841 whitespace changes from 4.64-PH/27
3843 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
3846 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
3847 are found, allow negative numbers in numeric criteria)
3849 JJ/06 exipick.20061117.2, added new $message_body_missing variable
3851 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
3852 to match changes made in 4.64-PH/43
3854 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
3856 PH/47 Put in an explicit test for a DNS lookup of an address record where the
3857 "domain" is actually an IP address, and force a failure. This locks out
3858 those revolvers/nameservers that support "A-for-A" lookups, in
3859 contravention of the specifications.
3861 PH/48 When a host name was looked up from an IP address, and the subsequent
3862 forward lookup of the name timed out, the host name was left in
3863 $sender_host_name, contrary to the specification.
3865 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3866 restricted to single-key lookups, Exim was not diagnosing an error if
3867 * or *@ was used with a query-style lookup.
3869 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3871 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3872 long ago, but noone interested enough thought of it. Let's just say that
3873 the "1.1" means that there are some new functions that weren't there at
3874 some point in the past.
3876 PH/51 Error processing for expansion failure of helo_data from an smtp
3877 transport during callout processing was broken.
3879 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3880 tested/used via the -bh/-bhc/-bs options.
3882 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3883 bug, fixed in subsequent PCRE releases).
3885 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3886 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3888 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3894 SC/01 Use a glob alias rather than an array ref in eximstats generated
3895 parser. This improves both readability and performance.
3897 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3898 Don't display local sender or destination tables in eximstats unless
3899 there is data to show.
3900 Added average volumes into the eximstats top table text output.
3902 SC/03 Collect data on the number of addresses (recipients) as well
3903 as the number of messages in eximstats.
3905 TF/01 Correct an error in the documentation for the redirect router. Exim
3906 does (usually) call initgroups() when daemonizing.
3908 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3909 with consistent privilege compared to when running as a daemon.
3911 TF/03 Note in the spec that $authenticated_id is not set for local
3912 submissions from trusted users.
3914 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3915 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3917 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3918 by adding some example configuration directives to the default
3919 configuration file. A little bit of work is required to uncomment the
3920 directives and define how usernames and passwords are checked, but
3921 there is now a framework to start from.
3923 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3924 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3925 without this. I don't know how relevant this is to other LDAP libraries.
3927 PH/02 Add the verb name to the "unknown ACL verb" error.
3929 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3931 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3933 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3934 directory not expanded when it should be if an expanded home directory
3935 was set for the address (which is overridden by the transport).
3937 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3940 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3941 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3942 because it is too late at that time, and has no effect.
3944 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3945 security issue with \' (bugzilla #107). I could not use the
3946 PQescapeStringConn() function, because it needs a PGconn value as one of
3949 PH/08 When testing addresses using -bt, indicate those final addresses that
3950 are duplicates that would not cause an additional delivery. At least one
3951 person was confused, thinking that -bt output corresponded to deliveries.
3952 (Suppressing duplicates isn't a good idea as you lose the information
3953 about possibly different redirections that led to the duplicates.)
3955 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3956 systems where poll() doesn't work, in particular OS X.
3958 PH/10 Added more information to debugging output for retry time not reached.
3960 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3961 operations in malware.c.
3963 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3966 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3967 syslog_duplication set false, log lines that would normally be written
3968 both the the main log and to the reject log were not written to syslog at
3971 PH/14 In the default configuration, change the use of "message" in ACL warn
3972 statements to "add_header".
3974 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3975 not followed by a command (e.g. "seen endif").
3977 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3978 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3981 PH/17 Added extra conditions to the default value of delay_warning_condition
3985 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3986 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3987 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3990 The Auto-Submitted: and various List- headers are standardised, whereas I
3991 don't think Precedence: ever was.
3993 PH/18 Refactored debugging code in route_finduser() to show more information,
3994 in particular, the error code if getpwnam() issues one.
3996 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3997 This is apparently needed in addition to the PH/07 change above to avoid
3998 any possible encoding problems.
4000 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
4001 but not after initializing Perl.
4003 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
4004 output them only if debugging. By default they are written stderr,
4005 apparently, which is not desirable.
4007 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
4010 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
4013 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
4015 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
4016 authenticated or an ident call has been made. Suppress the default
4017 values for $authenticated_id and $authenticated_sender (but permit -oMai
4018 and -oMas) when testing with -bh.
4020 PH/24 Re-jigged the order of the tests in the default configuration so that the
4021 tests for valid domains and recipients precede the DNS black list and CSA
4022 tests, on the grounds that those ones are more expensive.
4024 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
4025 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
4026 This bug exists in every version of Exim that I still have, right back to
4029 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
4030 However, an attempt to turn on case-sensitivity in a regex key by
4031 including (?-i) didn't work because the subject string was already
4032 lowercased, and the effects were non-intuitive. It turns out that a
4033 one-line patch can be used to allow (?-i) to work as expected.
4039 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
4040 other effects) broke the use of negated acl sub-conditions.
4042 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
4045 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
4046 "Deny" causes Exim to reject the incoming connection with a 554 error.
4047 Unfortunately, if there is a major crisis, such as a disk failure,
4048 tcp-wrappers gives "deny", whereas what one would like would be some
4049 kind of temporary error. A kludge has been added to help with this.
4050 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
4051 554 error is used if errno is still zero or contains ENOENT (which occurs
4052 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
4055 PH/03 Add -lutil to the default FreeBSD LIBS setting.
4057 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
4058 errors. Otherwise a message that provokes a temporary error (when other
4059 messages do not) can cause a whole host to time out.
4061 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
4062 addresses were routed directly to files or pipes from a redirect router.
4063 File deliveries just didn't batch; pipe deliveries might have suffered
4066 PH/06 A failure to get a lock for a hints database would erroneously always say
4067 "Failed to get write lock", even when it was really a read lock.
4069 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
4070 of 0600. This has been changed to use the value of the lockfile_mode
4071 option (which defaults to 0600).
4073 PH/08 Applied small patch from the Sieve maintainer.
4075 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
4076 folder from quota calculations, a direct delivery into this folder messed
4077 up the contents of the maildirsize file. This was because the regex was
4078 used only to exclude .Trash (or whatever) when the size of the mailbox
4079 was calculated. There was no check that a delivery was happening into an
4080 excluded directory. This bug has been fixed by ignoring all quota
4081 processing for deliveries into excluded directories.
4083 PH/10 Added the maildirfolder_create_regex option to appendfile.
4089 PH/01 The code for finding all the local interface addresses on a FreeBSD
4090 system running IPv6 was broken. This may well have applied to all BSD
4091 systems, as well as to others that have similar system calls. The broken
4092 code found IPv4 interfaces correctly, but gave incorrect values for the
4093 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
4094 that it would not match correctly against @[] and not recognize the IPv6
4097 PH/02 The ipliteral router was not recognizing addresses of the form user@
4098 [ipv6:....] because it didn't know about the "ipv6:" prefix.
4100 PH/03 Added disable_ipv6.
4102 PH/04 Changed $reply_address to use the raw form of the headers instead of the
4103 decoded form, because it is most often used to construct To: headers
4104 lines in autoreplies, and the decoded form may well be syntactically
4105 invalid. However, $reply_address has leading white space removed, and all
4106 newlines turned into spaces so that the autoreply transport does not
4109 PH/05 If group was specified without a user on a router, and no group or user
4110 was specified on a transport, the group from the router was ignored.
4112 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
4113 for visible compile-time settings that can be used to change these
4114 numbers, for those that want even more. Backwards compatibility with old
4115 spool files has been maintained. However, going back to a previous Exim
4116 release will lost any variables that are in spool files.
4118 PH/07 Two small changes when running in the test harness: increase delay when
4119 passing a TCP/IP connection to a new process, in case the original
4120 process has to generate a bounce, and remove special handling of
4121 127.0.0.2 (sic), which is no longer necessary.
4123 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
4124 be the same on different OS.
4126 PH/09 Moved a debug statement in filter processing to avoid a race problem when
4129 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
4130 whether --show-vars was specified or not
4132 JJ/02 exipick: Added support for new ACL variable spool format introduced
4135 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
4136 syntactically invalid From: or Reply-to: line, and a filter used this to
4137 generate an autoreply, and therefore failed to obtain an address for the
4138 autoreply, Exim could try to deliver to a non-existent relative file
4139 name, causing unrelated and misleading errors. What now happens is that
4140 it logs this as a hard delivery error, but does not attempt to create a
4143 PH/11 The exinext utility has a -C option for testing purposes, but although
4144 the given file was scanned by exinext itself; it wasn't being passed on
4145 when Exim was called.
4147 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
4148 an end-of-file indication when reading a command response.
4150 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
4151 compiled. In many other places in Exim, IPv6 addresses are always
4152 recognized, so I have changed this. It also means that IPv4 domain
4153 literals of the form [IPV4:n.n.n.n] are now always recognized.
4155 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
4156 used if the router is not running as root, for example, when verifying at
4157 ACL time, or when using -bh. The debugging output from this situation was
4158 non-existent - all you got was a failure to exec. I have made two
4161 (a) Failures to set uid/gid, the current directory, or a process leader
4162 in a subprocess such as that created by queryprogram now generate
4163 suitable debugging output when -d is set.
4165 (b) The queryprogram router detects when it is not running as root,
4166 outputs suitable debugging information if -d is set, and then runs
4167 the subprocess without attempting to change uid/gid.
4169 PH/15 Minor change to Makefile for building test_host (undocumented testing
4172 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
4173 additional section of a DNS packet that returns MX or SRV records.
4174 Instead, it always explicitly searches for A/AAAA records. This avoids
4175 major problems that occur when a DNS server includes only records of one
4176 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
4177 fixed another bug: if SRV records were looked up and the corresponding
4178 address records were *not* found in the additional section, the port
4179 values from the SRV records were lost.
4181 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
4182 using the correct key (the original address) when searching the retry
4183 rules in order to find which one to use for generating the retry hint.
4185 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
4186 adding the default one. Similarly, if it contains a Reply-To: header, the
4187 errors_reply_to option, if set, is not used.
4189 PH/19 When calculating a retry time, Exim used to measure the "time since
4190 failure" by looking at the "first failed" field in the retry record. Now
4191 it does not use this if it is later than than the arrival time of the
4192 message. Instead it uses the arrival time. This makes for better
4193 behaviour in cases where some deliveries succeed, thus re-setting the
4194 "first failed" field. An example is a quota failure for a huge message
4195 when small messages continue to be delivered. Without this change, the
4196 "time since failure" will always be short, possible causing more frequent
4197 delivery attempts for the huge message than are intended.
4198 [Note: This change was subsequently modified - see PH/04 for 4.62.]
4200 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
4201 $1, $2, $3) because the numerical variables can be reset during some
4202 expansion items (e.g. "match"), thereby losing the authentication data.
4204 PH/21 Make -bV show the size of off_t variables so that the test suite can
4205 decide whether to run tests for quotas > 2G.
4207 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
4208 mailbox_size, and mailbox_filecount in the appendfile transport. If a
4209 filecount value is greater than 2G or if a quota value is greater than 2G
4210 on a system where the size of off_t is not greater than 4, a panic error
4213 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
4214 never match. The debug and -bh output now contains an explicit error
4215 message indicating a malformed IPv4 address or mask.
4217 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
4218 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
4219 PH/23 above applies.
4221 PH/25 Do not write to syslog when running in the test harness. The only
4222 occasion when this arises is a failure to open the main or panic logs
4223 (for which there is an explicit test).
4225 PH/26 Added the /no_tell option to "control=freeze".
4227 PH/27 If a host name lookup failed very early in a connection, for example, if
4228 the IP address matched host_lookup and the reverse lookup yielded a name
4229 that did not have a forward lookup, an error message of the form "no IP
4230 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
4231 could be logged. Now it outputs the IP address instead of "NULL".
4233 PH/28 An enabling patch from MH: add new function child_open_exim2() which
4234 allows the sender and the authenticated sender to be set when
4235 submitting a message from within Exim. Since child_open_exim() is
4236 documented for local_scan(), the new function should be too.
4238 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
4239 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
4240 results in an empty string is now treated as unset.
4242 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
4244 PH/31 Added sender_verify_fail logging option.
4246 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
4247 needed by RFC 822 but not by RFC 2822 was commented out. I have now
4248 tidied the source and removed it altogether.
4250 PH/33 When a queue run was abandoned because the load average was too high, a
4251 log line was always written; now it is written only if the queue_run log
4252 selector is set. In addition, the log line for abandonment now contains
4253 information about the queue run such as the pid. This is always present
4254 in "start" and "stop" lines but was omitted from the "abandon" line.
4256 PH/34 Omit spaces between a header name and the colon in the error message that
4257 is given when verify = headers_syntax fails (if there are lots of them,
4258 the message gets confusing).
4260 PH/35 Change the default for dns_check_names_pattern to allow slashes within
4261 names, as there are now some PTR records that contain slashes. This check
4262 is only to protect against broken name servers that fall over on strange
4263 characters, so the fact that it applies to all lookups doesn't matter.
4265 PH/36 Now that the new test suite is complete, we can remove some of the
4266 special code in Exim that was needed for the old test suite. For example,
4267 sorting DNS records because real resolvers return them in an arbitrary
4268 order. The new test suite's fake resolver always returns records in the
4271 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
4272 bounces) except when queue_only is set, to avoid logging races between
4273 the different processes.
4275 PH/38 Panic-die if .include specifies a non-absolute path.
4277 PH/39 A tweak to the "H" retry rule from its user.
4279 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
4280 a label. They prevented compilation on older perls.
4282 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
4283 a warning to be raised on newish perls.
4285 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
4286 on queue. Changes to match documented behaviour of showing count of
4287 messages matching specified criteria.
4289 PH/40 Changed the default ident timeout from 30s to 5s.
4291 PH/41 Added support for the use of login_cap features, on those BSD systems
4292 that have them, for controlling the resources used by pipe deliveries.
4294 PH/42 The content-scanning code uses fopen() to create files in which to put
4295 message data. Previously it was not paying any attention to the mode of
4296 the files. Exim runs with umask(0) because the rest of the code creates
4297 files with open(), and sets the required mode explicitly. Thus, these
4298 files were ending up world-writeable. This was not a big issue, because,
4299 being within the spool directory, they were not world-accessible. I have
4300 created a function called modefopen, which takes an additional mode
4301 argument. It sets umask(777), creates the file, chmods it to the required
4302 mode, then resets the umask. All the relevant calls to fopen() in the
4303 content scanning code have been changed to use this function.
4305 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
4306 to 24 hours. This avoids potential overflow problems when processing G
4307 and H retry rules. I suspect nobody ever tinkers with this value.
4309 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
4311 PH/45 When the plaintext authenticator is running as a client, the server's
4312 challenges are checked to ensure they are valid base64 strings. By
4313 default, the authentication attempt is cancelled if an invalid string is
4314 received. Setting client_ignore_invalid_base64 true ignores these errors.
4315 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
4316 they are received. Thus, the responses can be made to depend on the
4317 challenges. If an invalid string is ignored, an empty string is placed in
4320 PH/46 Messages that are created by the autoreply transport now contains a
4321 References: header, in accordance with RFCs 2822 and 3834.
4323 PH/47 Added authenticated_sender_force to the smtp transport.
4325 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
4327 PH/49 Installed latest patch from the Sieve maintainer.
4329 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
4330 was also set, the appendfile transport was unnecessarily scanning a
4331 directory of message files (e.g. for maildir delivery) to find the count
4332 of files (along with the size), even though it did not need this
4333 information. It now does the scan only if it needs to find either the
4334 size of the count of files.
4336 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
4338 PH/52 Two bugs concerned with error handling when the smtp transport is
4341 (i) Exim was not creating retry information for temporary errors given
4342 for individual recipients after the DATA command when the smtp transport
4343 was used in LMTP mode. This meant that they could be retried too
4344 frequently, and not timed out correctly.
4346 (ii) Exim was setting the flag that allows error details to be returned
4347 for LMTP errors on RCPT commands, but not for LMTP errors for individual
4348 recipients that were returned after the DATA command.
4350 PH/53 This is related to PH/52, but is more general: for any failing address,
4351 when detailed error information was permitted to be returned to the
4352 sender, but the error was temporary, then after the final timeout, only
4353 "retry timeout exceeded" was returned. Now it returns the full error as
4354 well as "retry timeout exceeded".
4356 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
4357 do this, and (what is worse) MTAs that accept it.
4359 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
4360 will now be deprecated.
4362 PH/56 New os.c-cygwin from the Cygwin maintainer.
4364 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
4365 formats (previously only available in exim formats via -bpr, -bpru,
4366 and -bpra. Now also available in native and exiqgrep formats)
4368 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
4369 with very large, slow to parse queues
4371 JJ/08 exipick: added ! as generic prefix to negate any criteria format
4373 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
4375 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
4376 responses to authentication challenges, though it was showing the
4377 challenges; (ii) I've removed the CR characters from the debug output for
4380 PH/58 Allow for the insertion of a newline as well as a space when a string
4381 is turned into more than one encoded-word during RFC 2047 encoding. The
4382 Sieve code now uses this.
4384 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
4385 data_4xx, lost_connection, tls_required.
4387 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
4388 message was being sent as an SMTP response.
4390 PH/61 Add -l and -k options to exicyclog.
4392 PH/62 When verifying, if an address was redirected to one new address, so that
4393 verification continued, and the new address failed or deferred after
4394 having set something in $address_data, the value of $address_data was not
4395 passed back to the ACL. This was different to the case when no
4396 redirection occurred. The value is now passed back in both cases.
4398 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
4399 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
4400 use login_cap.h, so on its own it isn't the distinguishing feature. The
4401 new name refers directly to the setclassresources() function.
4403 PH/65 Added configuration files for NetBSD3.
4405 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
4407 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
4408 is preferred over IPv4.
4410 PH/68 The bounce_return_message and bounce_return_body options were not being
4411 honoured for bounces generated during the reception of non-SMTP messages.
4412 In particular, this applied to messages rejected by the ACL. This bug has
4413 been fixed. However, if bounce_return_message is true and bounce_return_
4414 body is false, the headers that are returned for a non-SMTP message
4415 include only those that have been read before the error was detected.
4416 (In the case of an ACL rejection, they have all been read.)
4418 PH/69 The HTML version of the specification is now built in a directory called
4419 spec_html instead of spec.html, because the latter looks like a path with
4420 a MIME-type, and this confuses some software.
4422 PH/70 Catch two compiler warnings in sieve.c.
4424 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
4425 function verify_get_ident() calls ip_connect() to connect a socket, but
4426 if the "connect()" function timed out, ip_connect() used to close the
4427 socket. However, verify_get_ident() also closes the socket later, and in
4428 between Exim writes to the log, which may get opened at this point. When
4429 the socket was closed in ip_connect(), the log could get the same file
4430 descriptor number as the socket. This naturally causes chaos. The fix is
4431 not to close the socket in ip_connect(); the socket should be closed by
4432 the function that creates it. There was only one place in the code where
4433 this was missing, in the iplookup router, which I don't think anybody now
4434 uses, but I've fixed it anyway.
4436 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
4437 well as to direct DNS lookups. Otherwise the handling of names in host
4438 lists is inconsistent and therefore confusing.
4444 PH/01 Two changes to the default runtime configuration:
4446 (1) Move the checks for relay_from_hosts and authenticated clients from
4447 after to before the (commented out) DNS black list checks.
4449 (2) Add control=submission to the relay_from_hosts and authenticated
4450 clients checks, on the grounds that messages accepted by these
4451 statements are most likely to be submissions.
4453 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
4455 (1) Generate an error if the third argument for the ${prvs expansion is
4458 (2) Treat a missing third argument of ${prvscheck as if it were an empty
4461 (3) Reset the variables that are obtained from the first argument of
4462 ${prvscheck and used in the second argument before leaving the code,
4463 because their memory is reclaimed, so using them afterwards may do
4466 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
4467 one (it's much easier than Tom thought :-).
4469 (5) Because of (4), we can now allow for the use of $prvscheck_result
4470 inside the third argument.
4472 PH/03 For some reason, the default setting of PATH when running a command from
4473 a pipe transport was just "/usr/bin". I have changed it to
4476 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
4477 anything to be listed in the output from -bV.
4479 PH/05 When a filter generated an autoreply, the entire To: header line was
4480 quoted in the delivery log line, like this:
4482 => >A.N.Other <ano@some.domain> <original@ddress> ...
4484 This has been changed so that it extracts the operative address. There
4485 may be more than one such address. If so, they are comma-separated, like
4488 => >ano@some.domain,ona@other.domain <original@ddress> ...
4490 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
4491 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
4492 not being looked up in the rDNS to get a host name, Exim was showing the
4493 IP address twice in Received: lines, even though the IP addresses were
4494 identical. For example:
4496 Received: from [1.2.3.4] (helo=[1.2.3.4])
4498 However, if the real host name was known, it was omitting the HELO data
4499 if it matched the actual IP address. This has been tidied up so that it
4500 doesn't show the same IP address twice.
4502 PH/07 When both +timestamp and +memory debugging was on, the value given by
4503 $tod_xxx expansions could be wrong, because the tod_stamp() function was
4504 called by the debug printing, thereby overwriting the timestamp buffer.
4505 Debugging no longer uses the tod_stamp() function when +timestamp is set.
4507 PH/08 When the original message was included in an autoreply transport, it
4508 always said "this is a copy of the message, including all the headers",
4509 even if body_only or headers_only was set. It now gives an appropriate
4512 PH/09 Applied a patch from the Sieve maintainer which:
4514 o fixes some comments
4515 o adds the (disabled) notify extension core
4516 o adds some debug output for the result of if/elsif tests
4517 o points to the current vacation draft in the documentation
4518 and documents the missing references header update
4522 o fixes a bug in processing the envelope test (when testing
4523 multiple envelope elements, the last element determined the
4526 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
4527 Electronic Mail") by including:
4529 Auto-submitted: auto-generated
4531 in the messages that it generates (bounce messages and others, such as
4532 warnings). In the case of bounce messages for non-SMTP messages, there was
4533 also a typo: it was using "Auto_submitted" (underscore instead of
4534 hyphen). Since every message generated by Exim is necessarily in response
4535 to another message, thes have all been changed to:
4537 Auto-Submitted: auto-replied
4539 in accordance with these statements in the RFC:
4541 The auto-replied keyword:
4543 - SHOULD be used on messages sent in direct response to another
4544 message by an automatic process,
4546 - MUST NOT be used on manually-generated messages,
4548 - MAY be used on Delivery Status Notifications (DSNs) and Message
4549 Disposition Notifications (MDNs),
4551 - MUST NOT be used on messages generated by automatic or periodic
4552 processes, except for messages which are automatic responses to
4555 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
4556 to the default Received: header definition.
4558 PH/12 Added log selector acl_warn_skipped (default on).
4560 PH/13 After a successful wildlsearch lookup, discard the values of numeric
4561 variables because (a) they are in the wrong storage pool and (b) even if
4562 they were copied, it wouldn't work properly because of the caching.
4564 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
4565 checking when decoding. Apparently there are clients that generate
4566 overlong encoded strings. Why am I not surprised?
4568 PH/15 If the first argument of "${if match_address" was not empty, but did not
4569 contain an "@" character, Exim crashed. Now it writes a panic log message
4570 and treats the condition as false.
4572 PH/16 In autoreply, treat an empty string for "once" the same as unset.
4574 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
4575 extension "envelope-auth". The code is finished and in agreement with
4576 other implementations, but there is no documentation so far and in fact,
4577 nobody wrote the draft yet. This extension is currently #undef'ed, thus
4578 not changing the active code.
4580 Print executed "if" and "elsif" statements when debugging is used. This
4581 helps a great deal to understand what a filter does.
4583 Document more things not specified clearly in RFC3028. I had all this
4584 sorted out, when out of a sudden new issues came to my mind. Oops."
4586 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
4589 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
4590 canonical form (as documented). However, after a host name lookup from
4591 the IP address, check_host() was doing a simple string comparison with
4592 addresses acquired from the DNS when checking that the found name did
4593 have the original IP as one of its addresses. Since any found IPv6
4594 addresses are likely to be in abbreviated form, the comparison could
4595 fail. Luckily, there already exists a function for doing the comparison
4596 by converting both addresses to binary, so now that is used instead of
4597 the text comparison.
4599 PH/20 There was another similar case to PH/19, when a complete host name was
4600 given in a host list; looking up its IP address could give an abbreviated
4601 form, whereas the current host's name might or might not be abbreviated.
4602 The same fix has been applied.
4608 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
4609 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
4612 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
4613 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
4615 PH/03 Typo: missing ".o" in src/pcre/Makefile.
4617 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
4618 header line, restrict the check to what is listed in RFCs 2369 and 2929.
4619 Also, for "Auto-Submitted", treat anything other than "no" as
4620 non-personal, in accordance with RFC 3834. (Previously it treated
4621 anything starting "auto-" as non-personal.)
4623 TF/01 The control=submission/name=... option had a problem with syntax
4624 errors if the name included a slash character. The /name= option
4625 now slurps the rest of the string, so it can include any characters
4626 but it must come last in the list of options (after /sender_retain
4629 PH/05 Some modifications to the interface to the fake nameserver for the new
4637 TK/01 Added the "success_on_redirect" address verification option. See
4638 NewStuff for rationale and an example.
4640 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
4642 PH/02 Patch to exigrep to allow it to work on syslog lines.
4644 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
4645 fread() to skip over the body file's header line, because in Cygwin the
4646 header line is locked and is inaccessible.
4648 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
4649 co-exist for some time) to make it clear that it is the Exim ID that is
4650 referenced, not the Message-ID: header line.
4652 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
4653 string_format() function, because snprintf() does not exist on all
4656 PH/06 The use of forbid_filter_existstest now also locks out the use of the
4657 ${stat: expansion item.
4659 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
4660 protocol synchronization error", to keep the pedants happy.
4662 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
4663 well as for IRIX systems, when gcc is being used. See the host.c source
4666 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
4668 PH/10 Named domain lists were not working if used in a queue_smtp_domains
4671 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
4672 transport and to the smtp transport in LMTP mode.
4674 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
4676 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
4677 run a filter in a subprocess. This could lead to confusion in subsequent
4678 lookups in the parent process. There should also be a search_tidyup() at
4679 the end of the subprocess.
4681 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
4682 only if the host matched helo_try_verify_hosts, which caused the
4683 verification to occur when the EHLO/HELO command was issued. The ACL just
4684 tested the remembered result. Now, if a previous verification attempt has
4685 not happened, "verify = helo" does it there and then.
4687 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
4689 TK/03 Fix log output including CR from clamd.
4691 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
4692 exist provoked a memory error which could cause a segfault.
4694 PH/15 Installed PCRE 6.2
4696 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
4698 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
4699 of the problem. Specifically, suggested +O2 rather than +O1 for the
4702 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
4704 PH/20 If a delivery was routed to a non-standard port by means of an SRV
4705 record, the port was not correctly logged when the outgoing_port log
4706 selector was set (it logged the transort's default port).
4708 PH/21 Added support for host-specific ports to manualroute, queryprogram,
4709 fallback_hosts, and "hosts" in the smtp transport.
4711 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
4712 host errors such as "Connection refused".
4714 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
4715 authentication with radiusclient 0.4.9:
4717 - Error returned from rc_read_config was caught wrongly
4718 - Username/password not passed on to radius server due to wrong length.
4720 The presumption is that some radiusclient API changes for 4.51/PH/17
4721 were not taken care of correctly. The code is still untested by me (my
4722 Linux distribution still has 0.3.2 of radiusclient), but it was
4723 contributed by a Radius user.
4725 PH/24 When doing a callout, the value of $domain wasn't set correctly when
4726 expanding the "port" option of the smtp transport.
4728 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
4729 while reading a MIME header. Thanks to Tom Hughes for a patch.
4731 PH/24 Include config.h inside local_scan.h so that configuration settings are
4734 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
4735 that in an ACL for RCPT (for example), you can examine exactly what was
4738 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
4739 commands, but it was not correctly comparing the address with the actual
4740 client host address. Thus, it would show the EHLO address in Received:
4741 header lines when this was not necessary.
4743 PH/27 Added the % operator to ${eval:}.
4745 PH/28 Exim tries to create and chdir to its spool directory when it starts;
4746 it should be ignoring failures (because with -C, for example, it has lost
4747 privilege). It wasn't ignoring creation failures other than "already
4750 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
4753 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
4754 because an input line was too long, either on its own, or by virtue of
4755 too many continuations, the temporary file was not being removed, and the
4756 return code was incorrect.
4758 PH/31 Missing "BOOL" in function definition in filtertest.c.
4760 PH/32 Applied Sieve patches from the maintainer.
4762 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
4764 PH/33 Added "verify = not_blind".
4766 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
4767 Local/Makefile (with some defaults set). These are used in built scripts
4768 such as exicyclog, but they have never been used in the exim_install
4769 script (though there are many overriding facilities there). I have
4770 arranged that the exim_install script now takes note of these two
4773 PH/35 Installed configuration files for Dragonfly.
4775 PH/36 When a locally submitted message by a trusted user did not contain a
4776 From: header, and the sender address was obtained from -f or from an SMTP
4777 MAIL command, and the trusted user did not use -F to supply a sender
4778 name, $originator_name was incorrectly used when constructing a From:
4779 header. Furthermore, $originator_name was used for submission mode
4780 messages from external hosts without From: headers in a similar way,
4781 which is clearly wrong.
4783 PH/37 Added control=suppress_local_fixups.
4785 PH/38 When log_selector = +received_sender was set, and the addition of the
4786 sender made the log line's construction buffer exactly full, or one byte
4787 less than full, an overflow happened when the terminating "\n" was
4790 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
4791 when the result of a list match is failure because a DNS lookup failed.
4793 PH/40 RM_COMMAND is now used in the building process.
4795 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
4796 the "build-* directories that it finds.
4798 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
4799 address in a domain literal was a prefix of an interface address.
4801 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
4802 when verifying a sender address, unless rewrite_headers is false.
4804 PH/44 Wrote a long comment about why errors_to addresses are verified as
4805 recipients, not senders.
4807 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
4808 the ratelimit ACL was added.
4810 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
4812 PH/46 Added extra information about PostgreSQL errors to the error string.
4814 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
4815 avoiding the need to install special zones in a real server. This is
4816 backwards compatible; if it can't find the fake resolver, it drops back.
4817 Thus, both old and new test suites can be run.
4819 TF/02 Added util/ratelimit.pl
4821 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
4822 clock is set back in time.
4824 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
4825 Candler <B.Candler@pobox.com>.
4827 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
4828 OK for addresses that are the result of redirections.
4830 PH/48 A number of further additions for the benefit of the new test suite,
4831 including a fake gethostbyname() that interfaces to the fake DNS resolver
4834 TF/06 The fix for widen_domains has also been applied to qualify_single and
4835 search_parents which are the other dnslookup options that can cause
4838 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
4841 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
4843 TF/07 Exim produced the error message "an SRV record indicated no SMTP
4844 service" if it encountered an MX record with an empty target hostname.
4845 The message is now "an MX or SRV record indicated no SMTP service".
4847 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
4848 if the DNS of the sending site is misconfigured. This is quite a
4849 common situation. This change restores the behaviour of treating a
4850 helo verification defer as a failure.
4852 PH/51 If self=fail was set on a router, the bounce message did not include the
4853 actual error message.
4859 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
4861 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4862 command itself ended in error, the underlying message about the transport
4863 filter timeout was being overwritten with the pipe command error. Now the
4864 underlying error message should be appended to the second error message.
4866 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4868 PH/02 Reduce the amount of output that "make" produces by default. Full output
4869 can still be requested.
4871 PH/03 The warning log line about a condition test deferring for a "warn" verb
4872 was being output only once per connection, rather than after each
4873 occurrence (because it was using the same function as for successful
4874 "warn" verbs). This seems wrong, so I have changed it.
4876 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4877 it should not have, which might have caused a crash in the right
4878 circumstances, but probably never did.
4880 PH/04 Installed a modified version of Tony Finch's patch to make submission
4881 mode fix the return path as well as the Sender: header line, and to
4882 add a /name= option so that you can make the user's friendly name appear
4885 TF/03 Added the control = fakedefer ACL modifier.
4887 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4888 Mark Lowes for thorough testing.
4890 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4892 TK/03 Merged latest SRS patch from Miles Wilton.
4894 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4895 with the definition in sysexits.h (which is #included earlier).
4896 Fortunately, Exim does not actually use EX_OK. The code used to try to
4897 preserve the sysexits.h value, by assuming that macro definitions were
4898 scanned for macro replacements. I have been disabused of this notion,
4899 so now the code just undefines EX_OK before #including unistd.h.
4901 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4902 in the smtp transport. When a block could not be written in a single
4903 write() function, the timeout was being re-applied to each part-write.
4904 This seems wrong - if the receiver was accepting one byte at a time it
4905 would take for ever. The timeout is now adjusted when this happens. It
4906 doesn't have to be particularly precise.
4908 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4909 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4911 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4912 without a domain if the check to <postmaster@domain> fails.
4914 SC/01 Eximstats: added -xls and the ability to specify output files
4915 (patch written by Frank Heydlauf).
4917 SC/02 Eximstats: use FileHandles for outputting results.
4919 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4921 SC/04 Eximstats: fixed display of large numbers with -nvr option
4923 SC/05 Eximstats: fixed merging of reports with empty tables.
4925 SC/06 Eximstats: added the -include_original_destination flag
4927 SC/07 Eximstats: removed tabs and trailing whitespace.
4929 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4931 TK/06 MBOX spool code: Add real "From " MBOX separator line
4932 so the .eml file is really in mbox format (even though
4933 most programs do not really care). Patch from Alex Miller.
4935 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4936 The latter is generated from $received_to and is only set if the
4937 message has one envelope recipient. SA can use these headers,
4938 obviously out-of-the-box. Patch from Alex Miller.
4940 PH/08 The ${def test on a variable was returning false if the variable's
4941 value was "0", contrary to what the specification has always said!
4942 The result should be true unless the variable is empty.
4944 PH/09 The syntax error of a character other than { following "${if
4945 def:variable_name" (after optional whitespace) was not being diagnosed.
4946 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4947 accidental colon was present, for example, could give incorrect results.
4949 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4950 result is used (not including appendfile, where other changes are about
4953 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4954 This involved changing a lot of size variables from int to off_t. It
4955 should work with maildirs and everything.
4957 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4958 spamd dying while we are connected to it.
4960 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4963 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4964 function for BDB 4.3.
4966 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4968 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4969 the sources for PCRE have been re-arranged and more files are now
4972 PH/15 The code I had for printing potentially long long variables in PH/11
4973 above was not the best (it lost precision). The length of off_t variables
4974 is now inspected at build time, and an appropriate printing format (%ld
4975 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4976 to be "long long int" or "long int". This is needed for the internal
4977 formatting function string_vformat().
4979 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4980 the configuration file to be ":syslog", then the script "guesses" where
4981 the logs files are, rather than using the compiled in default. In our
4982 case the guess is not the same as the compiled default, so the script
4983 suddenly stopped working when I started to use syslog. The patch checks
4984 to see if log_file_path is "". If so, it attempts to read it from exim
4985 with no configuration file to get the compiled in version, before it
4986 falls back to the previous guessing code."
4988 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4989 implementing BATV in an Exim configuration. See NewStuff for the gory
4992 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4993 Makefile that are specific to HP-UX.
4995 PH/18 If the "use_postmaster" option was set for a recipient callout together
4996 with the "random" option, the postmaster address was used as the MAIL
4997 FROM address for the random test, but not for the subsequent recipient
4998 test. It is now used for both.
5000 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
5001 patch removes a few documentation additions to RFC 3028, because the
5002 latest draft now contains them. It adds the new en;ascii-case comparator
5003 and a new error check for 8bit text in MIME parts. Comparator and
5004 require names are now matched exactly. I enabled the subaddress
5005 extension, but it is not well tested yet (read: it works for me)."
5007 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
5008 rework some of the code of TK/09 above to avoid the hardwired use of
5009 "%lld" and "long long". Replaced the call to snprintf() with a call to
5012 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
5013 records point to non-existent hosts", "retry timeout exceeded", and
5014 "retry time not reached for any host after a long failure period".
5016 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
5017 experimental DomainKeys support:
5019 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
5020 (2) On an error such as an illegally used "control", the wrong name for
5021 the control was given.
5023 These problems did NOT occur unless DomainKeys support was compiled.
5025 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
5027 PH/24 Added ${if match_ip condition.
5029 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
5030 databases so that it will be absolutely obvious if a crash occurs in the
5031 DB library. This is a regular occurrence (often caused by mis-matched
5034 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
5035 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
5036 picked up on a user's system that detects such things. There doesn't seem
5037 to be a gcc warning option for this - only an attribute that has to be
5038 put on the function's prototype. It seems that in Fedora Core 4 they have
5039 set this on a number of new functions. No doubt there will be more in due
5042 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
5043 specify a transport. However, if an address that was verified by such a
5044 router was the subject of a callout, Exim crashed because it tried to
5045 read the rcpt_include_affixes from the non-existent transport. Now it
5046 just assumes that the setting of that option is false. This bug was
5047 introduced by 4.51/PH/31.
5049 PH/28 Changed -d+all to exclude +memory, because that information is very
5050 rarely of interest, but it makes the output a lot bigger. People tend to
5051 do -d+all out of habit.
5053 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
5054 code in os-type was giving problems when libc.so lives in lib64, like on
5057 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
5058 aren't the modern standard, and it seems that some systems' include files
5059 don't always have them. Exim was already checking for some of the newer
5060 ones like T_AAAA, and defining it itself. I've added checks for all the
5061 record types that Exim uses.
5063 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
5064 not automatically generating a new one, as it is supposed to. This
5065 prevented TLS from working. If the file did exist, but contained invalid
5066 data, a new version was generated, as expected. It was only the case of a
5067 non-existent file that was broken.
5069 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
5070 with a change in libdomainkeys > 0.64.
5072 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
5073 from DNS. If the selector record carries the flag, it now has
5074 precedence over the domain-wide flag.
5076 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
5078 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
5079 the use of a port name that isn't defined in /etc/services) occurred, the
5080 message was deferred as in a normal delivery, and thus remained on the
5081 spool, instead of being failed because of the mua_wrapper setting. This
5082 is now fixed, and I tidied up some of the mua_wrapper messages at the
5085 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
5086 the messages in a hash of arrays rather than using individual hashes.
5087 This is a bit cleaner and results in dramatic memory savings, albeit
5088 at a slight CPU cost.
5090 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
5091 as requested by Marc Sherman.
5093 SC/10 Eximstats: added histograms for user specified patterns as requested
5096 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
5098 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
5099 fopen() in the content-scanning modules that did not already have it.
5105 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
5106 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
5108 TK/02 Fix ACL "control" statement not being available in MIME ACL.
5110 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
5112 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
5113 to test Sieve filters that use "vacation".
5115 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
5116 that changes the way the GnuTLS parameters are stored in the cache file.
5117 The new format can be generated externally. For backward compatibility,
5118 if the data in the cache doesn't make sense, Exim assumes it has read an
5119 old-format file, and it generates new data and writes a new file. This
5120 means that you can't go back to an older release without removing the
5123 PH/03 A redirect router that has both "unseen" and "one_time" set does not
5124 work if there are any delivery delays because "one_time" forces the
5125 parent to be marked "delivered", so its unseen clone is never tried
5126 again. For this reason, Exim now forbids the simultaneous setting of
5129 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
5130 redirected to themselves ("homonym" addresses). Read the long ChangeLog
5131 entry if you want to know the details. The fix, however, neglected to
5132 consider the case when local delivery batching is involved. The test for
5133 "previously delivered" was not happening when checking to see if an
5134 address could be batched with a previous (undelivered) one; under
5135 certain circumstances this could lead to multiple deliveries to the same
5138 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
5139 in its include files, and this causes problems building Exim.
5141 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
5142 header_syntax) but Exim was just ignoring anything given after a slash.
5143 In particular, this caused confusion with an attempt to use "verify =
5144 reverse_host_lookup/defer_ok". An error is now given when options are
5145 supplied for verify items that do not have them. (Maybe reverse_host_
5146 lookup should have a defer_ok option, but that's a different point.)
5148 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
5149 defined by RFC 821) to 2048, because there were problems with some AUTH
5150 commands, and RFC 1869 says the size should be increased for extended
5151 SMTP commands that take arguments.
5153 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
5156 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
5157 "unknown" error; now it says that the functionality isn't in the binary.
5159 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
5160 an address' error message when a string expansion fails (syntax or
5161 whatever). Otherwise the password may appear in the log. Following change
5162 PH/42 below, there is no longer a chance of it appearing in a bounce
5165 PH/11 Installed exipick version 20050225.0 from John Jetmore.
5167 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
5168 of its addresses was ever tried. (Bugzilla bug #2.)
5170 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
5171 the result incorrectly in the debug output. (It correctly added a newline
5172 to what was transported.)
5174 TF/01 Added $received_time.
5176 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
5177 commented out examples of how to interface to a virus scanner and to
5178 SpamAssassin. Also added commented examples of av_scanner and
5179 spamd_address settings.
5181 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
5182 and controls are allowed in which ACLs. There were a couple of minor
5183 errors. Some of the entries in the conditions table (which is a table of
5184 where they are NOT allowed) were getting very unwieldy; rewrote them as a
5185 negation of where the condition IS allowed.
5187 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
5189 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
5190 header file does not have a version number, so I've had to invent a new
5191 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
5192 API. The code is untested by me (my Linux distribution still has 0.3.2 of
5193 radiusclient), but it was contributed by a Radius user.
5195 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
5196 files or directories, for OpenSSL.
5198 PH/19 When an Exim process that is running as root has to create an Exim log
5199 file, it does so in a subprocess that runs as exim:exim so as to get the
5200 ownership right at creation (otherwise, other Exim processes might see
5201 the file with the wrong ownership). There was no test for failure of this
5202 fork() call, which would lead to the process getting stuck as it waited
5203 for a non-existent subprocess. Forks do occasionally fail when resources
5204 run out. I reviewed all the other calls to fork(); they all seem to check
5207 PH/20 When checking for unexpected SMTP input at connect time (before writing
5208 the banner), Exim was not dealing correctly with a non-positive return
5209 from the read() function. If the client had disconnected by this time,
5210 the result was a log entry for a synchronization error with an empty
5211 string after "input=" when read() returned zero. If read() returned -1
5212 (an event I could not check), uninitialized data bytes were printed.
5213 There were reports of junk text (parts of files, etc) appearing after
5216 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
5218 PH/22 Added support for macro redefinition, and (re)definition in between
5219 driver and ACL definitions.
5221 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
5222 forgetting to use the resulting value; it was using the unexpanded value.
5224 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
5225 hadn't been configured. The fix is from Juergen Kreileder, who
5226 understands it better than I do:
5228 "Here's what I see happening with three configured cyrus_sasl
5229 authenticators configured (plain, login, cram-md5):
5231 On startup auth_cyrus_sasl_init() gets called for each of these.
5232 This means three calls to sasl_listmech() without a specified mech_list.
5233 => SASL tests which mechs of all available mechs actually work
5234 => three warnings about OTP not working
5235 => the returned list contains: plain, login, cram-md5, digest-md5, ...
5237 With the patch, sasl_listmech() also gets called three times. But now
5238 SASL's mech_list option is set to the server_mech specified in the the
5239 authenticator. Or in other words, the answer from sasl_listmech()
5240 gets limited to just the mech you're testing for (which is different
5242 => the return list contains just 'plain' or 'login', 'cram-md5' or
5243 nothing depending on the value of ob->server_mech.
5245 I've just tested the patch: Authentication still works fine,
5246 unavailable mechs specified in the exim configuration are still
5247 caught, and the auth.log warnings about OTP are gone."
5249 PH/25 When debugging is enabled, the contents of the command line are added
5250 to the debugging output, even when log_selector=+arguments is not
5253 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
5254 answer is "GNU", and only if the return is "GNU/something" is the answer
5257 PH/27 $acl_verify_message is now set immediately after the failure of a
5258 verification in an ACL, and so is available in subsequent modifiers. In
5259 particular, the message can be preserved by coding like this:
5261 warn !verify = sender
5262 set acl_m0 = $acl_verify_message
5264 Previously, $acl_verify_message was set only while expanding "message"
5265 and "log_message" when a very denied access.
5267 PH/28 Modified OS/os.c-Linux with
5269 -#ifndef OS_LOAD_AVERAGE
5270 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
5272 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
5273 nomenclature these days.)
5275 PH/29 Installed patch from the Sieve maintainer that adds the options
5276 sieve_useraddress and sieve_subaddress to the redirect router.
5278 PH/30 In these circumstances:
5279 . Two addresses routed to the same list of hosts;
5280 . First host does not offer TLS;
5281 . First host accepts first address;
5282 . First host gives temporary error to second address;
5283 . Second host offers TLS and a TLS session is established;
5284 . Second host accepts second address.
5285 Exim incorrectly logged both deliveries with the TLS parameters (cipher
5286 and peerdn, if requested) that were in fact used only for the second
5289 PH/31 When doing a callout as part of verifying an address, Exim was not paying
5290 attention to any local part prefix or suffix that was matched by the
5291 router that accepted the address. It now behaves in the same way as it
5292 does for delivery: the affixes are removed from the local part unless
5293 rcpt_include_affixes is set on the transport.
5295 PH/32 Add the sender address, as F=<...>, to the log line when logging a
5296 timeout during the DATA phase of an incoming message.
5298 PH/33 Sieve envelope tests were broken for match types other than :is. I have
5299 applied a patch sanctioned by the Sieve maintainer.
5301 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
5302 the uid or gid is negative. A case of a negative gid caused this to be
5303 noticed. The fix allows for either to be negative.
5305 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
5306 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
5309 PH/36 The change PH/12 above was broken. Fixed it.
5311 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
5312 the grounds that routing the same address twice would always produce the
5313 same answer. This might have been true once, but it is certainly no
5314 longer true now. Routing a child address may depend on the previous
5315 routing that produced that child. Some complicated redirection strategies
5316 went wrong when messages had multiple recipients, and made Exim's
5317 behaviour dependent on the order in which the addresses were given.
5319 I have moved the duplicate checking until after the routing is complete.
5320 Exim scans the addresses that are assigned to local and remote
5321 transports, and removes any duplicates. This means that more work will be
5322 done, as duplicates will always all be routed, but duplicates are
5323 presumably rare, so I don't expect this is of any significance.
5325 For deliveries to pipes, files, and autoreplies, the duplicate checking
5326 still happens during the routing process, since they are not going to be
5329 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
5330 It corrects a timeout issue with spamd. This is Ian's comment: "The
5331 background is that sometimes spamd either never reads data from a
5332 connection it has accepted, or it never writes response data. The exiscan
5333 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
5334 blindly assumes that writes won't block so it may never time out."
5336 PH/39 Allow G after quota size as well as K and M.
5338 PH/40 The value set for $authenticated_id in an authenticator may not contain
5339 binary zeroes or newlines because the value is written to log lines and
5340 to spool files. There was no check on this. Now the value is run through
5341 the string_printing() function so that such characters are converted to
5342 printable escape sequences.
5344 PH/41 $message_linecount is a new variable that contains the total number of
5345 lines in the message. Compare $body_linecount, which is the count for the
5348 PH/42 Exim no longer gives details of delivery errors for specific addresses in
5349 bounce and delay warning messages, except in certain special cases, which
5352 (a) An SMTP error message from a remote host;
5353 (b) A message specified in a :fail: redirection;
5354 (c) A message specified in a "fail" command in a system filter;
5355 (d) A message specified in a FAIL return from the queryprogram router;
5356 (e) A message specified by the cannot_route_message router option.
5358 In these cases only, Exim does include the error details in bounce and
5359 warning messages. There are also a few cases where bland messages such
5360 as "unrouteable address" or "local delivery error" are given.
5362 PH/43 $value is now also set for the "else" part of a ${run expansion.
5364 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
5365 being worked on, but at least Exim now implements the latest version to
5368 PH/45 In a pipe transport, although a timeout while waiting for the pipe
5369 process to complete was treated as a delivery failure, a timeout while
5370 writing the message to the pipe was logged, but erroneously treated as a
5371 successful delivery. Such timeouts include transport filter timeouts. For
5372 consistency with the overall process timeout, these timeouts are now
5373 treated as errors, giving rise to delivery failures by default. However,
5374 there is now a new Boolean option for the pipe transport called
5375 timeout_defer, which, if set TRUE, converts the failures into defers for
5376 both kinds of timeout. A transport filter timeout is now identified in
5379 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
5380 systems where "make" and "gmake" are different, calling "gmake" at top
5381 level broke things. I've arranged for the value of $(MAKE) to be passed
5382 from the Makefile to this script so that it can call the same version of
5386 A note about Exim versions 4.44 and 4.50
5387 ----------------------------------------
5389 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
5390 changes of various kinds. As a consequence, a big documentation update was
5391 needed. This delayed the release for rather longer than seemed good, especially
5392 in the light of a couple of (minor) security issues. Therefore, the changes
5393 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
5394 release. So 4.44 and 4.50 are in effect two different branches that both start
5397 I have left the 4.50 change log unchanged; it contains all the changes since
5398 4.43. The change log for 4.44 is below; many of its items are identical to
5399 those for 4.50. This seems to be the most sensible way to preserve the
5400 historical information.
5406 1. Minor wording change to the doc/README.SIEVE file.
5408 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
5409 computation of the current number of files was incorrect.
5411 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5412 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5415 4. Give more explanation in the error message when the command for a transport
5416 filter fails to execute.
5418 5. There are several places where Exim runs a non-Exim command in a
5419 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5420 was being done only for the command run by the queryprogram router. It is
5421 now done for all such subprocesses. The other cases are: ${run, transport
5422 filters, and the commands run by the lmtp and pipe transports.
5424 6. Added CONFIGURE_GROUP build-time option.
5426 7. Some older OS have a limit of 256 on the maximum number of file
5427 descriptors. Exim was using setrlimit() to set 1000 as a large value
5428 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5429 systems. I've change it so that if it can't get 1000, it tries for 256.
5431 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5432 was an oversight, and furthermore, ever since the addition of extra
5433 controls (e.g. 4.43/32), the checks on when to allow different forms of
5434 "control" were broken. There should now be diagnostics for all cases when a
5435 control that does not make sense is encountered.
5437 9. Added the /retain_sender option to "control=submission".
5439 10. $recipients is now available in the predata ACL (oversight).
5441 11. Tidy the search cache before the fork to do a delivery from a message
5442 received from the command line. Otherwise the child will trigger a lookup
5443 failure and thereby defer the delivery if it tries to use (for example) a
5444 cached ldap connection that the parent has called unbind on.
5446 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5447 of $address_data from the recipient verification was clobbered by the
5448 sender verification.
5450 13. The value of address_data from a sender verification is now available in
5451 $sender_address_data in subsequent conditions in the ACL statement.
5453 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
5455 15. Added a new option "connect=<time>" to callout options, to set a different
5458 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5459 was its contents. (It was OK if the option was not defined at all.)
5461 17. A "Completed" log line is now written for messages that are removed from
5462 the spool by the -Mrm option.
5464 18. New variables $sender_verify_failure and $recipient_verify_failure contain
5465 information about exactly what failed.
5467 19. Added -dd to debug only the daemon process.
5469 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
5470 handles timeouts, both on the server side and network timeouts. Renamed the
5471 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
5473 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
5474 It is now set to "smtps".
5476 22. $host_address is now set to the target address during the checking of
5477 ignore_target_hosts.
5479 23. When checking ignore_target_hosts for an ipliteral router, no host name was
5480 being passed; this would have caused $sender_host_name to have been used if
5481 matching the list had actually called for a host name (not very likely,
5482 since this list is usually IP addresses). A host name is now passed as
5485 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5486 code that specifies a non-restarting handler (typically sigaction() in
5487 modern systems) in an attempt to fix a rare and obscure crash bug.
5489 25. Narrowed the window for a race in the daemon that could cause it to ignore
5490 SIGCHLD signals. This is not a major problem, because they are used only to
5491 wake it up if nothing else does.
5493 26. A malformed maildirsize file could cause Exim to calculate negative values
5494 for the mailbox size or file count. Odd effects could occur as a result.
5495 The maildirsize information is now recalculated if the size or filecount
5498 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5499 support for a long time. Removed HAVE_SYS_VFS_H.
5501 28. Installed the latest version of exipick from John Jetmore.
5503 29. In an address list, if the pattern was not a regular expression, an empty
5504 subject address (from a bounce message) matched only if the pattern was an
5505 empty string. Non-empty patterns were not even tested. This was the wrong
5506 because it is perfectly reasonable to use an empty address as part of a
5507 database query. An empty address is now tested by patterns that are
5508 lookups. However, all the other forms of pattern expect the subject to
5509 contain a local part and a domain, and therefore, for them, an empty
5510 address still always fails if the pattern is not itself empty.
5512 30. Exim went into a mad DNS loop when attempting to do a callout where the
5513 host was specified on an smtp transport, and looking it up yielded more
5514 than one IP address.
5516 31. Re-factored the code for checking spool and log partition space into a
5517 function that finds that data and another that does the check. The former
5518 is then used to implement four new variables: $spool_space, $log_space,
5519 $spool_inodes, and $log_inodes.
5521 32. The RFC2047 encoding function was originally intended for short strings
5522 such as real names; it was not keeping to the 75-character limit for
5523 encoded words that the RFC imposes. It now respects the limit, and
5524 generates multiple encoded words if necessary. To be on the safe side, I
5525 have increased the buffer size for the ${rfc2047: expansion operator from
5528 33. It is now permitted to omit both strings after an "if" condition; if the
5529 condition is true, the result is "true". As before, when the second string
5530 is omitted, a false condition yields an empty string. This makes it less
5531 cumbersome to write custom ACL and router conditions.
5533 34. Failure to deliver a bounce message always caused it to be frozen, even if
5534 there was an errors_to setting on the router. The errors_to setting is now
5537 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5538 canonical form (fully expanded) before being placed in
5539 $sender_host_address.
5541 36. The table in the code that translates DNS record types into text (T_A to
5542 "A" for instance) was missing entries for NS and CNAME. It is just possible
5543 that this could have caused confusion if both these types were looked up
5544 for the same domain, because the text type is used as part of Exim's
5545 per-process caching. But the chance of anyone hitting this buglet seems
5548 37. The dnsdb lookup has been extended in a number of ways.
5550 (1) There is a new type, "zns", which walks up the domain tree until it
5551 finds some nameserver records. It should be used with care.
5553 (2) There is a new type, "mxh", which is like "mx" except that it returns
5554 just the host names, not the priorities.
5556 (3) It is now possible to give a list of domains (or IP addresses) to be
5557 looked up. The behaviour when one of the lookups defers can be
5558 controlled by a keyword.
5560 (4) It is now possible to specify the separator character for use when
5561 multiple records are returned.
5563 38. The dnslists ACL condition has been extended: it is now possible to supply
5564 a list of IP addresses and/or domains to be looked up in a particular DNS
5567 39. Added log_selector=+queue_time_overall.
5569 40. When running the queue in the test harness, wait just a tad after forking a
5570 delivery process, to get repeatability of debugging output.
5572 41. Include certificate and key file names in error message when GnuTLS fails
5573 to set them up, because the GnuTLS error message doesn't include the name
5574 of the failing file when there is a problem reading it.
5576 42. Allow both -bf and -bF in the same test run.
5578 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
5580 44. The "Exiscan patch" is now merged into the mainline Exim source.
5582 45. Sometimes the final signoff response after QUIT could fail to get
5583 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5584 < 0 before doing a fflush(). This bug looks as though it goes back to the
5585 introduction of TLS in release 3.20, but "sometimes" must have been rare
5586 because the tests only now provoked it.
5588 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
5589 (this can affect the format of dates).
5591 47. exim_tidydb, when checking for the continued existence of a message for
5592 which it has found a message-specific retry record, was not finding
5593 messages that were in split spool directories. Consequently, it was
5594 deleting retry records that should have stayed in existence.
5596 48. Steve fixed some bugs in eximstats.
5598 49. The SPA authentication driver was not abandoning authentication and moving
5599 on to the next authenticator when an expansion was forced to fail,
5600 contradicting the general specification for all authenticators. Instead it
5601 was generating a temporary error. It now behaves as specified.
5603 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
5604 (the order specifies the preference for clients). The order is now AES256,
5605 AES128, 3DES, ARCFOUR128.
5607 51. Small patch to Sieve code - explicitly set From: when generating an
5610 52. Exim crashed if a remote delivery caused a very long error message to be
5611 recorded - for instance if somebody sent an entire SpamAssassin report back
5612 as a large number of 550 error lines. This bug was coincidentally fixed by
5613 increasing the size of one of Exim's internal buffers (big_buffer) that
5614 happened as part of the Exiscan merge. However, to be on the safe side, I
5615 have made the code more robust (and fixed the comments that describe what
5618 53. Now that there can be additional text after "Completed" in log lines (if
5619 the queue_time_overall log selector is set), a one-byte patch to exigrep
5620 was needed to allow it to recognize "Completed" as not the last thing in
5623 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
5624 patch that reportedly fixes this has been added. I am not expert enough to
5625 create a test for it. This is what the patch creator wrote:
5627 "I found a little strange behaviour of ldap code when working with
5628 Windows 2003 AD Domain, where users was placed in more than one
5629 Organization Units. When I tried to give exim partial DN, the exit code
5630 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
5631 But simultaneously result of request was absolutely normal ldap result,
5632 so I produce this patch..."
5634 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
5635 REFERENCE, so I have modified the code to exclude the patch when that macro
5638 55. Some experimental protocols are using DNS PTR records for new purposes. The
5639 keys for these records are domain names, not reversed IP addresses. The
5640 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5641 leaves it alone. Component reversal etc. now happens only for IP addresses.
5644 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
5646 57. Double the size of the debug message buffer (to 2048) so that more of very
5647 long debug lines gets shown.
5649 58. The exicyclog utility now does better if the number of log files to keep
5650 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
5652 59. Two changes related to the smtp_active_hostname option:
5654 (1) $smtp_active_hostname is now available as a variable.
5655 (2) The default for smtp_banner uses $smtp_active_hostname instead
5656 of $primary_hostname.
5658 60. The host_aton() function is supposed to be passed a string that is known
5659 to be a valid IP address. However, in the case of IPv6 addresses, it was
5660 not checking this. This is a hostage to fortune. Exim now panics and dies
5661 if the condition is not met. A case was found where this could be provoked
5662 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5663 components; fortuitously, this particular loophole had already been fixed
5664 by change 4.50/55 above.
5666 If there are any other similar loopholes, the new check in host_aton()
5667 itself should stop them being exploited. The report I received stated that
5668 data on the command line could provoke the exploit when Exim was running as
5669 exim, but did not say which command line option was involved. All I could
5670 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5671 running as the user.
5674 61. There was a buffer overflow vulnerability in the SPA authentication code
5675 (which came originally from the Samba project). I have added a test to the
5676 spa_base64_to_bits() function which I hope fixes it.
5679 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
5680 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
5682 63. The daemon start-up calls getloadavg() while still root for those OS that
5683 need the first call to be done as root, but it missed one case: when
5684 deliver_queue_load_max is set with deliver_drop_privilege. This is
5685 necessary for the benefit of the queue runner, because there is no re-exec
5686 when deliver_drop_privilege is set.
5688 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
5689 This has been fixed.
5691 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5692 was in use, was not putting the data itself into the right store pool;
5693 consequently, it could be overwritten for a subsequent message in the same
5694 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5697 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
5699 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
5700 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
5701 the function were treating the return as a boolean value, which happened to
5702 work because 0=false and not-0=true, but is not correct code.
5704 68. The host_aton() function was not handling scoped IPv6 addresses (those
5705 with, for example, "%eth0" on the end) correctly.
5707 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
5708 negated items (that is, ~something) in unsigned ints. Some compilers
5709 apparently mutter when there is no cast.
5711 70. If an address verification called from an ACL failed, and did not produce a
5712 user-specific message (i.e. there was only a "system" message), nothing was
5713 put in $acl_verify_message. In this situation, it now puts the system
5716 71. Change 4.23/11 added synchronization checking at the start of an SMTP
5717 session; change 4.31/43 added the unwanted input to the log line - except
5718 that it did not do this in the start of session case. It now does.
5720 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
5721 This is wrong and can cause the other end to generate a synchronization
5722 error if it is another Exim or anything else that does the synchronization
5723 check. A QUIT command is no longer sent after a timeout.
5725 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
5726 during host lookups.
5728 74. The defer_ok option of callout verification was not working if it was used
5729 when verifying addresses in header lines, that is, for this case:
5731 verify = header_sender/callout=defer_ok
5733 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
5734 those file descriptors could be used for SMTP connections. If anything
5735 wrote to stderr (the example that came up was "warn" in embedded Perl), it
5736 could be sent to the SMTP client, causing chaos. The daemon now opens
5737 stdin, stdout, and stderr to /dev/null when it puts itself into the
5740 76. Arrange for output from Perl's "warn" command to be written to Exim's main
5741 log by default. The user can override this with suitable Perl magic.
5743 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
5744 the log message when discard triggers, was not working for the DATA ACL or
5745 for the non-SMTP ACL.
5747 78. Error message wording change in sieve.c.
5749 79. If smtp_accept_max_per_host was set, the number of connections could be
5750 restricted to fewer than expected, because the daemon was trying to set up
5751 a new connection before checking whether the processes handling previous
5752 connections had finished. The check for completed processes is now done
5753 earlier. On busy systems, this bug wouldn't be noticed because something
5754 else would have woken the daemon, and it would have reaped the completed
5757 80. If a message was submitted locally by a user whose login name contained one
5758 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
5759 It caused a spool format error. I have fixed the spool reading code. A
5760 related problem was that the "from" clause in the Received: line became
5761 illegal because of the space(s). It is now covered by ${quote_local_part.
5763 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
5766 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
5768 83. Patch from Sieve maintainer to fix unterminated string problem in
5769 "vacation" handling.
5771 84. Some minor changes to the Linux configuration files to help with other
5772 OS variants using glibc.
5774 85. One more patch for Sieve to update vacation handling to latest spec.
5777 ----------------------------------------------------
5778 See the note above about the 4.44 and 4.50 releases.
5779 ----------------------------------------------------
5785 1. Change 4.43/35 introduced a bug that caused file counts to be
5786 incorrectly computed when quota_filecount was set in an appendfile
5789 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5790 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5793 3. Give more explanation in the error message when the command for a transport
5794 filter fails to execute.
5796 4. There are several places where Exim runs a non-Exim command in a
5797 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5798 was being done only for the command run by the queryprogram router. It is
5799 now done for all such subprocesses. The other cases are: ${run, transport
5800 filters, and the commands run by the lmtp and pipe transports.
5802 5. Some older OS have a limit of 256 on the maximum number of file
5803 descriptors. Exim was using setrlimit() to set 1000 as a large value
5804 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5805 systems. I've change it so that if it can't get 1000, it tries for 256.
5807 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5808 was an oversight, and furthermore, ever since the addition of extra
5809 controls (e.g. 4.43/32), the checks on when to allow different forms of
5810 "control" were broken. There should now be diagnostics for all cases when a
5811 control that does not make sense is encountered.
5813 7. $recipients is now available in the predata ACL (oversight).
5815 8. Tidy the search cache before the fork to do a delivery from a message
5816 received from the command line. Otherwise the child will trigger a lookup
5817 failure and thereby defer the delivery if it tries to use (for example) a
5818 cached ldap connection that the parent has called unbind on.
5820 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5821 of $address_data from the recipient verification was clobbered by the
5822 sender verification.
5824 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5825 was its contents. (It was OK if the option was not defined at all.)
5827 11. A "Completed" log line is now written for messages that are removed from
5828 the spool by the -Mrm option.
5830 12. $host_address is now set to the target address during the checking of
5831 ignore_target_hosts.
5833 13. When checking ignore_target_hosts for an ipliteral router, no host name was
5834 being passed; this would have caused $sender_host_name to have been used if
5835 matching the list had actually called for a host name (not very likely,
5836 since this list is usually IP addresses). A host name is now passed as
5839 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5840 code that specifies a non-restarting handler (typically sigaction() in
5841 modern systems) in an attempt to fix a rare and obscure crash bug.
5843 15. Narrowed the window for a race in the daemon that could cause it to ignore
5844 SIGCHLD signals. This is not a major problem, because they are used only to
5845 wake it up if nothing else does.
5847 16. A malformed maildirsize file could cause Exim to calculate negative values
5848 for the mailbox size or file count. Odd effects could occur as a result.
5849 The maildirsize information is now recalculated if the size or filecount
5852 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5853 support for a long time. Removed HAVE_SYS_VFS_H.
5855 18. Updated exipick to current release from John Jetmore.
5857 19. Allow an empty sender to be matched against a lookup in an address list.
5858 Previously the only cases considered were a regular expression, or an
5861 20. Exim went into a mad DNS lookup loop when doing a callout where the
5862 host was specified on the transport, if the DNS lookup yielded more than
5865 21. The RFC2047 encoding function was originally intended for short strings
5866 such as real names; it was not keeping to the 75-character limit for
5867 encoded words that the RFC imposes. It now respects the limit, and
5868 generates multiple encoded words if necessary. To be on the safe side, I
5869 have increased the buffer size for the ${rfc2047: expansion operator from
5872 22. Failure to deliver a bounce message always caused it to be frozen, even if
5873 there was an errors_to setting on the router. The errors_to setting is now
5876 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5877 canonical form (fully expanded) before being placed in
5878 $sender_host_address.
5880 24. Updated eximstats to version 1.33
5882 25. Include certificate and key file names in error message when GnuTLS fails
5883 to set them up, because the GnuTLS error message doesn't include the name
5884 of the failing file when there is a problem reading it.
5886 26. Expand error message when OpenSSL has problems setting up cert/key files.
5889 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5890 (this can affect the format of dates).
5892 28. exim_tidydb, when checking for the continued existence of a message for
5893 which it has found a message-specific retry record, was not finding
5894 messages that were in split spool directories. Consequently, it was
5895 deleting retry records that should have stayed in existence.
5897 29. eximstats updated to version 1.35
5898 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5899 1.35 - bugfix such that pie charts by volume are generated correctly
5901 30. The SPA authentication driver was not abandoning authentication and moving
5902 on to the next authenticator when an expansion was forced to fail,
5903 contradicting the general specification for all authenticators. Instead it
5904 was generating a temporary error. It now behaves as specified.
5906 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5907 (the order specifies the preference for clients). The order is now AES256,
5908 AES128, 3DES, ARCFOUR128.
5910 31. Small patch to Sieve code - explicitly set From: when generating an
5913 32. Exim crashed if a remote delivery caused a very long error message to be
5914 recorded - for instance if somebody sent an entire SpamAssassin report back
5915 as a large number of 550 error lines. This bug was coincidentally fixed by
5916 increasing the size of one of Exim's internal buffers (big_buffer) that
5917 happened as part of the Exiscan merge. However, to be on the safe side, I
5918 have made the code more robust (and fixed the comments that describe what
5921 33. Some experimental protocols are using DNS PTR records for new purposes. The
5922 keys for these records are domain names, not reversed IP addresses. The
5923 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5924 leaves it alone. Component reversal etc. now happens only for IP addresses.
5927 34. The host_aton() function is supposed to be passed a string that is known
5928 to be a valid IP address. However, in the case of IPv6 addresses, it was
5929 not checking this. This is a hostage to fortune. Exim now panics and dies
5930 if the condition is not met. A case was found where this could be provoked
5931 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5932 components; fortuitously, this particular loophole had already been fixed
5933 by change 4.50/55 or 4.44/33 above.
5935 If there are any other similar loopholes, the new check in host_aton()
5936 itself should stop them being exploited. The report I received stated that
5937 data on the command line could provoke the exploit when Exim was running as
5938 exim, but did not say which command line option was involved. All I could
5939 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5940 running as the user.
5943 35. There was a buffer overflow vulnerability in the SPA authentication code
5944 (which came originally from the Samba project). I have added a test to the
5945 spa_base64_to_bits() function which I hope fixes it.
5948 36. The daemon start-up calls getloadavg() while still root for those OS that
5949 need the first call to be done as root, but it missed one case: when
5950 deliver_queue_load_max is set with deliver_drop_privilege. This is
5951 necessary for the benefit of the queue runner, because there is no re-exec
5952 when deliver_drop_privilege is set.
5954 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5955 was in use, was not putting the data itself into the right store pool;
5956 consequently, it could be overwritten for a subsequent message in the same
5957 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5960 38. Sometimes the final signoff response after QUIT could fail to get
5961 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5962 < 0 before doing a fflush(). This bug looks as though it goes back to the
5963 introduction of TLS in release 3.20, but "sometimes" must have been rare
5964 because the tests only now provoked it.
5970 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5971 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5972 Now it returns an integer. A number of calls were still expecting a T/F
5973 return. Fortuitously, in all cases, the tests worked in OK situations,
5974 which is the norm. However, things would have gone wrong on any write
5975 failures on the smtp file descriptor. This function is used when sending
5976 messages over SMTP and also when doing verify callouts.
5978 2. When Exim is called to do synchronous delivery of a locally submitted
5979 message (the -odf or -odi options), it no longer closes stderr before doing
5982 3. Implemented the mua_wrapper option.
5984 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5986 5. Implemented the functions header_remove(), header_testname(),
5987 header_add_at_position(), and receive_remove_recipient(), and exported them
5990 6. If an ACL "warn" statement specified the addition of headers, Exim already
5991 inserted X-ACL-Warn: at the start if there was no header name. However, it
5992 was not making this test for the second and subsequent header lines if
5993 there were newlines in the string. This meant that an invalid header could
5994 be inserted if Exim was badly configured.
5996 7. Allow an ACL "warn" statement to add header lines at the start or after all
5997 the Received: headers, as well as at the end.
5999 8. Added the rcpt_4xx retry error code.
6001 9. Added postmaster_mailfrom=xxx to callout verification option.
6003 10. Added mailfrom=xxxx to the callout verification option, for verify=
6006 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
6007 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
6009 12. Inserted some casts to stop certain compilers warning when using pointer
6010 differences as field lengths or precisions in printf-type calls (mostly
6011 affecting debugging statements).
6013 13. Added optional readline() support for -be (dynamically loaded).
6015 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
6016 same clock tick as a message's arrival, so that its received time was the
6017 same as the "first fail" time on the retry record, and that message
6018 remained on the queue past the ultimate address timeout, every queue runner
6019 would try a delivery (because it was past the ultimate address timeout) but
6020 after another failure, the ultimate address timeout, which should have then
6021 bounced the address, did not kick in. This was a "< instead of <=" error;
6022 in most cases the first failure would have been in the next clock tick
6023 after the received time, and all would be well.
6025 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
6026 being recognized when the domain list was tested by the match_domain
6027 condition in an expansion string.
6029 16. Added the ${str2b64: operator.
6031 17. Exim was always calling setrlimit() to set a large limit for the number of
6032 processes, without checking whether the existing limit was already
6033 adequate. (It did check for the limit on file descriptors.) Furthermore,
6034 errors from getrlimit() and setrlimit() were being ignored. Now they are
6035 logged to the main and panic logs, but Exim does carry on, to try to do its
6036 job under whatever limits there are.
6038 18. Imported PCRE 5.0.
6040 19. Trivial typo in log message " temporarily refused connection" (the leading
6043 20. If the log selector return_path_on_delivery was set and an address was
6044 redirected to /dev/null, the delivery process crashed because it assumed
6045 that a return path would always be set for a "successful" delivery. In this
6046 case, the whole delivery is bypassed as an optimization, and therefore no
6049 21. Internal re-arrangement: the function for sending a challenge and reading
6050 a response while authentication was assuming a zero-terminated challenge
6051 string. It's now changed to take a pointer and a length, to allow for
6052 binary data in such strings.
6054 22. Added the cyrus_sasl authenticator (code supplied by MBM).
6056 23. Exim was not respecting finduser_retries when seeking the login of the
6057 uid under which it was called; it was always trying 10 times. (The default
6058 setting of finduser_retries is zero.) Also, it was sleeping after the final
6059 failure, which is pointless.
6061 24. Implemented tls_on_connect_ports.
6063 25. Implemented acl_smtp_predata.
6065 26. If the domain in control=submission is set empty, Exim assumes that the
6066 authenticated id is a complete email address when it generates From: or
6067 Sender: header lines.
6069 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
6070 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
6071 chown and chgrp in /bin and hostname in /usr/bin.
6073 28. Exim was keeping the "process log" file open after each use, just as it
6074 does for the main log. This opens the possibility of it remaining open for
6075 long periods when the USR1 signal hits a daemon. Occasional processlog
6076 errors were reported, that could have been caused by this. Anyway, it seems
6077 much more sensible not to leave this file open at all, so that is what now
6080 29. The long-running daemon process does not normally write to the log once it
6081 has entered its main loop, and it closes the log before doing so. This is
6082 so that log files can straightforwardly be renamed and moved. However,
6083 there are a couple of unusual error situations where the daemon does write
6084 log entries, and I had neglected to close the log afterwards.
6086 30. The text of an SMTP error response that was received during a remote
6087 delivery was being truncated at 512 bytes. This is too short for some of
6088 the long messages that one sometimes sees. I've increased the limit to
6091 31. It is now possible to make retry rules that apply only when a message has a
6092 specific sender, in particular, an empty sender.
6094 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
6095 it possible to be selective about when SMTP synchronization is enforced.
6097 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
6099 32. Implemented hosts_connection_nolog.
6101 33. Added an ACL for QUIT.
6103 34. Setting "delay_warning=" to disable warnings was not working; it gave a
6106 35. Added mailbox_size and mailbox_filecount to appendfile.
6108 36. Added control = no_multiline_responses to ACLs.
6110 37. There was a bug in the logic of the code that waits for the clock to tick
6111 in the case where the clock went backwards by a substantial amount such
6112 that the microsecond fraction of "now" was more than the microsecond
6113 fraction of "then" (but the whole seconds number was less).
6115 38. Added support for the libradius Radius client library this is found on
6116 FreeBSD (previously only the radiusclient library was supported).
6122 1. When certain lookups returned multiple values in the form name=value, the
6123 quoting of the values was not always being done properly. Specifically:
6124 (a) If the value started with a double quote, but contained no whitespace,
6126 (b) If the value contained whitespace other than a space character (i.e.
6127 tabs or newlines or carriage returns) it was not quoted.
6128 This fix has been applied to the mysql and pgsql lookups by writing a
6129 separate quoting function and calling it from the lookup code. The fix
6130 should probably also be applied to nisplus, ibase and oracle lookups, but
6131 since I cannot test any of those, I have not disturbed their existing code.
6133 2. A hit in the callout cache for a specific address caused a log line with no
6134 reason for rejecting RCPT. Now it says "Previous (cached) callout
6135 verification failure".
6137 3. There was an off-by-one bug in the queryprogram router. An over-long
6138 return line was truncated at 256 instead of 255 characters, thereby
6139 overflowing its buffer with the terminating zero. As well as fixing this, I
6140 have increased the buffer size to 1024 (and made a note to document this).
6142 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
6143 when Exim is waiting for an SMTP response from a remote server, Exim
6144 restarts its select() call on the socket, thereby resetting its timeout.
6145 This is not a problem when such interrupts are rare. Somebody set up a cron
6146 job to run exiwhat every 2 minutes, which is less than the normal select()
6147 timeout (5 or 10 minutes). This meant that the select() timeout never
6148 kicked in because it was always reset. I have fixed this by comparing the
6149 time when an interrupt arrives with the time at the start of the first call
6150 to select(). If more time than the timeout has elapsed, the interrupt is
6151 treated as a timeout.
6153 5. Some internal re-factoring in preparation for the addition of Sieve
6154 extensions (by MH). In particular, the "personal" test is moved to a
6155 separate function, and given an option for scanning Cc: and Bcc: (which is
6156 not set for Exim filters).
6158 6. When Exim created an email address using the login of the caller as the
6159 local part (e.g. when creating a From: or Sender: header line), it was not
6160 quoting the local part when it contained special characters such as @.
6162 7. Installed new OpenBSD configuration files.
6164 8. Reworded some messages for syntax errors in "and" and "or" conditions to
6165 try to make them clearer.
6167 9. Callout options, other than the timeout value, were being ignored when
6168 verifying sender addresses in header lines. For example, when using
6170 verify = header_sender/callout=no_cache
6172 the cache was (incorrectly) being used.
6174 10. Added a missing instance of ${EXE} to the exim_install script; this affects
6175 only the Cygwin environment.
6177 11. When return_path_on_delivery was set as a log selector, if different remote
6178 addresses in the same message used different return paths and parallel
6179 remote delivery occurred, the wrong values would sometimes be logged.
6180 (Whenever a remote delivery process finished, the return path value from
6181 the most recently started remote delivery process was logged.)
6183 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
6184 lines when AUTH and/or TLS are in use. This is the "received protocol"
6185 field. Exim used to use "asmtp" for authenticated SMTP, without any
6186 indication (in the protocol name) for TLS use. Now it follows the RFC and
6187 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
6188 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
6189 names appear in log lines as well as in Received: header lines.
6191 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
6192 extensions, and comparison tests, and to fix some bugs.
6194 14. Changes to the "personal" filter test:
6196 (1) The test was buggy in that it was just doing the equivalent of
6197 "contains" tests on header lines. For example, if a user's address was
6198 anne@some.where, the "personal" test would incorrectly be true for
6200 To: susanne@some.where
6202 This test is now done by extracting each address from the header in turn,
6203 and checking the entire address. Other tests that are part of "personal"
6204 are now done using regular expressions (for example, to check local parts
6205 of addresses in From: header lines).
6207 (2) The list of non-personal local parts in From: addresses has been
6208 extended to include "listserv", "majordomo", "*-request", and "owner-*",
6209 taken from the Sieve specification recommendations.
6211 (3) If the message contains any header line starting with "List-" it is
6212 treated as non-personal.
6214 (4) The test for "circular" in the Subject: header line has been removed
6215 because it now seems ill-conceived.
6217 15. Minor typos in src/EDITME comments corrected.
6219 16. Installed latest exipick from John Jetmore.
6221 17. If headers_add on a router specified a text string that was too long for
6222 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
6223 of string_sprintf() is now avoided.
6225 18. $message_body_size was not set (it was always zero) when running the DATA
6226 ACL and the local_scan() function.
6228 19. For the "mail" command in an Exim filter, no default was being set for
6229 the once_repeat time, causing a random time value to be used if "once" was
6230 specified. (If the value happened to be <= 0, no repeat happened.) The
6231 default is now 0s, meaning "never repeat". The "vacation" command was OK
6232 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
6233 (I found it when inspecting the code).
6235 20. There is now an overall timeout for performing a callout verification. It
6236 defaults to 4 times the callout timeout, which applies to individual SMTP
6237 commands during the callout. The overall timeout applies when there is more
6238 than one host that can be tried. The timeout is checked before trying the
6239 next host. This prevents very long delays if there are a large number of
6240 hosts and all are timing out (e.g. when the network connections are timing
6241 out). The value of the overall timeout can be changed by specifying an
6242 additional sub-option for "callout", called "maxwait". For example:
6244 verify = sender/callout=5s,maxwait=20s
6246 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
6247 to the end before writing, but this should make it even safer).
6249 22. Exim was forgetting that it had advertised PIPELINING for the second and
6250 subsequent messages on an SMTP connection. It was also not resetting its
6251 memory on STARTTLS and an internal HELO.
6253 23. When Exim logs an SMTP synchronization error within a session, it now
6254 records whether PIPELINING has been advertised or not.
6256 24. Added 3 instances of "(long int)" casts to time_t variables that were being
6257 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
6258 rather than long int.
6260 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
6262 26. Added the never_mail option to autoreply.
6268 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
6269 crash if the getsockname() call failed; this can happen if a connection is
6270 closed very soon after it is established. The problem was simply in the
6271 order in which certain operations were done, causing Exim to try to write
6272 to the SMTP stream before it had set up the file descriptor. The bug has
6273 been fixed by making things happen in the correct order.
6279 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
6280 before the connection was closed, thus losing the rejection response.
6282 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
6283 some early Solaris releases, but causes trouble in current releases where
6284 socklen_t is defined.
6286 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
6289 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
6292 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
6293 that releasing the top of it at the end releases what was used for sub-
6294 expansions (unless the block got too big). However, discard this block if
6295 the first thing is a variable or header, so that we can use its block when
6296 it is dynamic (useful for very large $message_headers, for example).
6298 6. Lookups now cache *every* query, not just the most recent. A new, separate
6299 store pool is used for this. It can be recovered when all lookup caches are
6300 flushed. Lookups now release memory at the end of their result strings.
6301 This has involved some general refactoring of the lookup sources.
6303 7. Some code has been added to the store_xxx() functions to reduce the amount
6304 of flapping under certain conditions.
6306 8. log_incoming_interface used to affect only the <= reception log lines. Now
6307 it causes the local interface and port to be added to several more SMTP log
6308 lines, for example "SMTP connection from", and rejection lines.
6310 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
6312 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
6314 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
6315 could be overwritten at the end of the current message (or the start of a
6316 new message if it was set in a HELO ACL). The value is now preserved for
6317 the duration of the SMTP connection.
6319 12. If a transport had a headers_rewrite setting, and a matching header line
6320 contained an unqualified address, that address was qualified, even if it
6321 did not match any rewriting rules. The underlying bug was that the values
6322 of the flags that permit the existence of unqualified sender and recipient
6323 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
6324 non-local messages, and by -bnq for local messages) were not being
6325 preserved with the message after it was received.
6327 13. When Exim was logging an SMTP synchronization error, it could sometimes log
6328 "next input=" as part of the text comprising the host identity instead of
6329 the correct text. The code was using the same buffer for two different
6330 strings. However, depending on which order the printing function evaluated
6331 its arguments, the bug did not always show up. Under Linux, for example, my
6332 test suite worked just fine.
6334 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
6335 doesn't work with some older versions of Perl. It has been changed to "my",
6336 which in any case is probably the better facility to use.
6338 15. A really picky compiler found some instances of statements for creating
6339 error messages that either had too many or two few arguments for the format
6342 16. The size of the buffer for calls to the DNS resolver has been increased
6343 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
6344 for addresses that have a lot of PTR records. This alleviates a problem; it
6345 does not fully solve it.
6347 17. A dnsdb lookup for PTR records that receives more data than will fit in the
6348 buffer now truncates the list and logs the incident, which is the same
6349 action as happens when Exim is looking up a host name and its aliases.
6350 Previously in this situation something unpredictable would happen;
6351 sometimes it was "internal error: store_reset failed".
6353 18. If a server dropped the connection unexpectedly when an Exim client was
6354 using GnuTLS and trying to read a response, the client delivery process
6355 crashed while trying to generate an error log message.
6357 19. If a "warn" verb in an ACL added multiple headers to a message in a single
6358 string, for example:
6360 warn message = H1: something\nH2: something
6362 the text was added as a single header line from Exim's point of view
6363 though it ended up OK in the delivered message. However, searching for the
6364 second and subsequent header lines using $h_h2: did not work. This has been
6365 fixed. Similarly, if a system filter added multiple headers in this way,
6366 the routers could not see them.
6368 20. Expanded the error message when iplsearch is called with an invalid key to
6369 suggest using net-iplsearch in a host list.
6371 21. When running tests using -bh, any delays imposed by "delay" modifiers in
6372 ACLs are no longer actually imposed (and a message to that effect is
6375 22. If a "gecos" field in a passwd entry contained escaped characters, in
6376 particular, if it contained a \" sequence, Exim got it wrong when building
6377 a From: or a Sender: header from that name. A second bug also caused
6378 incorrect handling when an unquoted " was present following a character
6379 that needed quoting.
6381 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
6382 was not being matched caselessly.
6384 24. Arranged for all hyphens in the exim.8 source to be escaped with
6387 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
6388 properly. Recipient callout cache records were still being keyed to include
6389 the sender, even when use_sender was set false. This led to far more
6390 callouts that were necessary. The sender is no longer included in the key
6391 when use_sender is false.
6393 26. Added "control = submission" modifier to ACLs.
6395 27. Added the ${base62d: operator to decode base 62 numbers.
6397 28. dnsdb lookups can now access SRV records.
6399 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
6400 the configuration file.
6402 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
6403 (-v) mode. This makes the output for a verbose queue run more intelligible.
6405 31. Added a use_postmaster feature to recipient callouts.
6407 32. Added the $body_zerocount variable, containing the number of binary zero
6408 bytes in the message body.
6410 33. The time of last modification of the "new" subdirectory is now used as the
6411 "mailbox time last read" when there is a quota error for a maildir
6414 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
6416 35. Added +ignore_unknown as a special item in host lists.
6418 36. Code for decoding IPv6 addresses in host lists is now included, even if
6419 IPv6 support is not being compiled. This fixes a bug in which an IPv6
6420 address was recognized as an IP address, but was then not correctly decoded
6421 into binary, causing unexpected and incorrect effects when compared with
6428 1. Very minor rewording of debugging text in manualroute to say "list of
6429 hosts" instead of "hostlist".
6431 2. If verify=header_syntax was set, and a header line with an unqualified
6432 address (no domain) and a large number of spaces between the end of the
6433 name and the colon was received, the reception process suffered a buffer
6434 overflow, and (when I tested it) crashed. This was caused by some obsolete
6435 code that should have been removed. The fix is to remove it!
6437 3. When running in the test harness, delay a bit after writing a bounce
6438 message to get a bit more predictability in the log output.
6440 4. Added a call to search_tidyup() just before forking a reception process. In
6441 theory, someone could use a lookup in the expansion of smtp_accept_max_
6442 per_host which, without the tidyup, could leave open a database connection.
6444 5. Added the variables $recipient_data and $sender_data which get set from a
6445 lookup success in an ACL "recipients" or "senders" condition, or a router
6446 "senders" option, similar to $domain_data and $local_part_data.
6448 6. Moved the writing of debug_print from before to after the "senders" test
6451 7. Change 4.31/66 (moving the time when the Received: is generated) caused
6452 problems for message scanning, either using a data ACL, or using
6453 local_scan() because the Received: header was not generated till after they
6454 were called (in order to set the time as the time of reception completion).
6455 I have revised the way this works. The header is now generated after the
6456 body is received, but before the ACL or local_scan() are called. After they
6457 are run, the timestamp in the header is updated.
6463 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
6464 before starting a queue runner without re-exec. This happened only when
6465 deliver_drop_privilege was set or when the Exim user was set to root. The
6466 effect of the bug was that timeouts during subsequent deliveries caused
6467 crashes instead of being properly handled. The handler is now left at its
6468 default (and expected) setting.
6470 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
6471 message, again when deliver_drop_privilege is set or Exim is run as root.
6472 The bug described in (1) was not present in this case, but the tidying up
6473 of the other signals was missing. I have made the two cases consistent.
6475 3. The ignore_target_hosts setting on a manualroute router was being ignored
6476 for hosts that were looked up using the /MX notation.
6478 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
6481 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
6482 operated on the sender address. After changing the $sender_address to <>
6483 for the sender address verify, Exim was re-instated it as the original
6484 (before rewriting) address, but remembering that it had rewritten it, so it
6485 wasn't rewriting it again. This bug also had the effect of breaking the
6486 sender address verification caching when the sender address was rewritten.
6488 6. The ignore_target_hosts option was being ignored by the ipliteral router.
6489 This has been changed so that if the ip literal address matches
6490 ignore_target_hosts, the router declines.
6492 7. Added expansion conditions match_domain, match_address, and match_local_
6493 part (NOT match_host).
6495 8. The placeholder for the Received: header didn't have a length field set.
6497 9. Added code to Exim itself and to exim_lock to test for a specific race
6498 condition that could lead to file corruption when using MBX delivery. The
6499 issue is with the lockfile that is created in /tmp. If this file is removed
6500 after a process has opened it but before that process has acquired a lock,
6501 there is the potential for a second process to recreate the file and also
6502 acquire a lock. This could lead to two Exim processes writing to the file
6503 at the same time. The added code performs the same test as UW imapd; it
6504 checks after acquiring the lock that its file descriptor still refers to
6505 the same named file.
6507 10. The buffer for building added header lines was of fixed size, 8192 bytes.
6508 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
6511 11. Added the smtp_active_hostname option. If used, this will typically be made
6512 to depend on the incoming interface address. Because $interface_address is
6513 not set up until the daemon has forked a reception process, error responses
6514 that can happen earlier (such as "too many connections") no longer contain
6517 12. If an expansion in a condition on a "warn" statement fails because a lookup
6518 defers, the "warn" statement is abandoned, and the next ACL statement is
6519 processed. Previously this caused the whole ACL to be aborted.
6521 13. Added the iplsearch lookup type.
6523 14. Added ident_timeout as a log selector.
6525 15. Added tls_certificate_verified as a log selector.
6527 16. Added a global option tls_require_ciphers (compare the smtp transport
6528 option of the same name). This controls incoming TLS connections.
6530 17. I finally figured out how to make tls_require_ciphers do a similar thing
6531 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
6532 before starting the TLS session.
6534 18. Tabs are now shown as \t in -bP output.
6536 19. If the log selector return_path_on_delivery was set, Exim crashed when
6537 bouncing a message because it had too many Received: header lines.
6539 20. If two routers both had headers_remove settings, and the first one included
6540 a superfluous trailing colon, the final name in the first list and the
6541 first name in the second list were incorrectly joined into one item (with a
6542 colon in the middle).
6548 1. Added -C and -D options to the exinext utility, mainly to make it easier
6549 to include in the automated testing, but these could be helpful when
6550 multiple configurations are in use.
6552 2. The exinext utility was not formatting the output nicely when there was
6553 an alternate port involved in the retry record key, nor when there was a
6554 message id as well (for retries that were specific to a specific message
6555 and a specific host). It was also confused by IPv6 addresses, because of
6556 the additional colons they contain. I have fixed the IPv4 problem, and
6557 patched it up to do a reasonable job for IPv6.
6559 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
6560 delivery, the log line now contains "pipelined" if PIPELINING was used.
6562 4. An SMTP transport process used to panic and die if the bind() call to set
6563 an explicit outgoing interface failed. This has been changed; it is now
6564 treated in the same way as a connect() failure.
6566 5. A reference to $sender_host_name in the part of a conditional expansion
6567 that was being skipped was still causing a DNS lookup. This no longer
6570 6. The def: expansion condition was not recognizing references to header lines
6571 that used bh_ and bheader_.
6573 7. Added the _cache feature to named lists.
6575 8. The code for checking quota_filecount in the appendfile transport was
6576 allowing one more file than it should have been.
6578 9. For compatibility with Sendmail, the command line option
6586 and sets the incoming protocol and host name (for trusted callers). The
6587 host name and its colon can be omitted when only the protocol is to be set.
6588 Note the Exim already has two private options, -pd and -ps, that refer to
6589 embedded Perl. It is therefore impossible to set a protocol value of "d" or
6590 "s", but I don't think that's a major issue.
6592 10. A number of refactoring changes to the code, none of which should affect
6595 (a) The number of logging options was getting close to filling up the
6596 32-bit word that was used as a bit map. I have split them into two classes:
6597 those that are passed in the argument to log_write(), and those that are
6598 only ever tested independently outside of that function. These are now in
6599 separate 32-bit words, so there is plenty of room for expansion again.
6600 There is no change in the user interface or the logging behaviour.
6602 (b) When building, for example, log lines, the code previously used a
6603 macro that called string_cat() twice, in order to add two strings. This is
6604 not really sufficiently general. Furthermore, there was one instance where
6605 it was actually wrong because one of the argument was used twice, and in
6606 one call a function was used. (As it happened, calling the function twice
6607 did not affect the overall behaviour.) The macro has been replaced by a
6608 function that can join an arbitrary number of extra strings onto a growing
6611 (c) The code for expansion conditions now uses a table and a binary chop
6612 instead of a serial search (which was left over from when there were very
6613 few conditions). Also, it now recognizes conditions like "pam" even when
6614 the relevant support is not compiled in: a suitably worded error message is
6615 given if an attempt is made to use such a condition.
6617 11. Added ${time_interval:xxxxx}.
6619 12. A bug was causing one of the ddress fields not to be passed back correctly
6620 from remote delivery subprocesses. The field in question was not being
6621 subsequently used, so this caused to problems in practice.
6623 13. Added new log selectors queue_time and deliver_time.
6625 14. Might have fixed a bug in maildirsizefile handling that threw up
6626 "unexpected character" debug warnings, and recalculated the data
6627 unnecessarily. In any case, I expanded the warning message to give more
6630 15. Added the message "Restricted characters in address" to the statements in
6631 the default ACL that block characters like @ and % in local parts.
6633 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
6634 Three changes have been made:
6636 (a) There was a serious bug; a negative response to MAIL caused the whole
6637 recipient domain to be cached as invalid, thereby blocking all messages
6638 to all local parts at the same domain, from all senders. This bug has
6639 been fixed. The domain is no longer cached after a negative response to
6640 MAIL if the sender used is not empty.
6642 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
6645 (c) A new callout option, "use_sender" has been added for people who want
6646 the modified behaviour.
6652 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
6655 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
6656 indeed breaks things for older releases.
6658 3. Added additional logging to the case where there is a problem reading data
6659 from a filter that is running in a subprocess using a pipe, in order to
6660 try to track down a specific problem.
6662 4. Testing facility fudge: when running in the test harness and attempting
6663 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
6664 getting "No route to host". Convert this to a timeout.
6666 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
6669 6. Some OS don't have socklen_t but use size_t instead. This affects the
6670 fifth argument of getsockopt() amongst other things. This is now
6671 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
6672 can be set for individual OS. I have set it for SunOS5, OSF1, and
6673 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
6674 some earlier ones do not.
6676 7. Change 4.30/15 was not doing the test caselessly.
6678 8. The standard form for an IPv6 address literal was being rejected by address
6679 parsing in, for example, MAIL and RCPT commands. An example of this kind of
6680 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6681 this, as well as the form without the "IPv6" on the front (but only when
6682 address literals are enabled, of course).
6684 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
6686 10. Exim crashed if a message with an empty sender address specified by -f
6687 encountered a router with an errors_to setting. This could be provoked only
6688 by a command such as
6692 where an empty string was supplied; "<>" did not hit this bug.
6694 11. Installed PCRE release 4.5.
6696 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
6697 remained set. It is now erased.
6699 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
6700 times from message ids (which are base 36 rather than the normal 62).
6702 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
6703 were being counted as actual protocol errors, and logged if the log
6704 selector +smtp_protocol_error was set. One cannot be perfect in this test,
6705 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
6706 and DATA following a set of rejected RCPTs do not count as protocol errors.
6707 In other words, Exim assumes they were pipelined, though this may not
6708 actually be the case. Of course, in all cases the client gets an
6709 appropriate error code.
6711 15. If a lookup fails in an ACL condition, a message about the failure may
6712 be available; it is used if testing the ACL cannot continue, because most
6713 such messages specify what the cause of the deferral is. However, some
6714 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
6715 that caused an old message to be retained and used if a later statement
6716 caused a defer, replacing the real cause of the deferral.
6718 16. If an IP address had so many PTR records that the DNS lookup buffer
6719 was not large enough to hold them, Exim could crash while trying to process
6720 the truncated data. It now detects and logs this case.
6722 17. Further to 4.21/58, another change has been made: if (and only if) the
6723 first line of a message (the first header line) ends with CRLF, a bare LF
6724 in a subsequent header line has a space inserted after it, so as not to
6725 terminate the header.
6727 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
6728 unnecessarily, used atoi() instead of strtol(), and didn't check the
6729 termination when getting file sizes from file names by regex.
6731 19. Completely re-implemented the support for maildirsize files, in the light
6732 of a number of problems with the previous contributed implementation
6733 (4.30/29). In particular:
6735 . If the quota is zero, the maildirsize file is maintained, but no quota is
6738 . If the maildir directory does not exist, it is created before any attempt
6739 to write a maildirsize file.
6741 . The quota value in the file is just a cache; if the quota is changed in
6742 the transport, the new value overrides.
6744 . A regular expression is available for excluding directories from the
6747 20. The autoreply transport checks the characters in options that define the
6748 message's headers; it allows continued headers, but it was checking with
6749 isspace() after an embedded newline instead of explicitly looking for a
6752 21. If all the "regular" hosts to which an address was routed had passed their
6753 expiry times, and had not reached their retry times, the address was
6754 bounced, even if fallback hosts were defined. Now Exim should go on to try
6757 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
6758 equivalent code in the SMTP transport. Some hosts send humungous responses
6759 to HELO/EHLO, more than 1024 it seems.
6761 23. Refactoring: code in filter.c used (void *) for "any old type" but this
6762 gives compiler warnings in some environments. I've now done it "properly",
6765 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
6766 (because of problems with the built-in one) was declared to return uschar *
6767 instead of char *, causing compiler failure.
6769 25. Fixed a file descriptor leak when processing alias/forward files.
6771 26. Fixed a minor format string issue in dbfn.c.
6773 27. Typo in exim.c: ("dmbnz" for "dbmnz").
6775 28. If a filter file refered to $h_xxx or $message_headers, and the headers
6776 contained RFC 2047 "words", Exim's memory could, under certain conditions,
6779 29. When a sender address is verified, it is cached, to save repeating the test
6780 when there is more than one recipient in a message. However, when the
6781 verification involves a callout, it is possible for different callout
6782 options to be set for different recipients. It is too complicated to keep
6783 track of this in the cache, so now Exim always runs a verification when a
6784 callout is required, relying on the callout cache for the optimization.
6785 The overhead is duplication of the address routing, but this should not be
6788 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
6789 to be verified with callout=postmaster, and the main callout worked but the
6790 postmaster check failed, the verification correctly failed. However, if a
6791 subsequent RCPT command asked for sender verification *without* the
6792 postmaster check, incorrect caching caused this verification also to fail,
6795 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
6796 it was not caching the DNS options (qualify_single, search_parents) that
6797 were used when the lookup failed. A subsequent lookup with different
6798 options therefore always gave the same answer, though there were cases
6799 where it should not have. (Example: a "domains = !$mx_any" option on a
6800 dnslookup router: the "domains" option is always processed without any
6801 widening, but the router might have qualify_single set.) Now Exim uses the
6802 cached value only when the same options are set.
6804 32. Added John Jetmore's "exipick" utility to the distribution.
6806 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
6807 than a timeout (e.g. a certificate is required, and is not provided), an
6808 Exim server now closes the connection immediately. Previously it waited for
6809 the client to close - but if the client is SSL, it seems that they each
6810 wait for each other, leading to a delay before one of them times out.
6812 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
6813 maintained 0.8.x compatibility because I don't think many are using it, and
6814 it is clearly obsolete.
6816 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
6819 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
6820 client certificate was expired. A simple patch fixes this, though I don't
6821 understand the full logic of why the verify callback is called multiple
6824 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
6825 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
6826 which causes problems with some clients (such as the Certicom SSL Plus
6827 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
6828 disables the coutermeasure allowing Eudora to connect."
6830 38. Exim was not checking that a write() to a log file succeeded. This could
6831 lead to Bad Things if a log got too big, in particular if it hit a file
6832 size limit. Exim now panics and dies if it cannot write to a log file, just
6833 as it does if it cannot open a log file.
6835 39. Modified OS/Makefile-Linux so that it now contains
6837 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
6839 The two -D definitions ensure that Exim is compiled with large file
6840 support, which makes it possible to handle log files that are bigger than
6843 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
6844 instance) a domain was checked against a named list that involved a lookup,
6845 causing $domain_data to be set, then another domain was checked against the
6846 same list, then the first domain was re-checked, the value of $domain_data
6847 after the final check could be wrong. In particular, if the second check
6848 failed, it could be set empty. This bug probably also applied to
6851 41. The strip_trailing_dot option was not being applied to the address given
6852 with the -f command-line option.
6854 42. The code for reading a message's header from the spool was incrementing
6855 $received_count, but never initializing it. This meant that the value was
6856 incorrect (doubled) while delivering a message in the same process in which
6857 it was received. In the most common configuration of Exim, this never
6858 happens - a fresh exec is done - but it can happen when
6859 deliver_drop_privilege is set.
6861 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6862 it now includes up to 150 characters of the unexpected data in the log
6865 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6866 and building data strings. The size of both of these buffers was 10 000
6867 bytes - far larger than anybody would *ever* want, thought I. Needless to
6868 say, somebody hit the limit. I have increased the maximum line length to
6869 20 000 and the maximum data length of concatenated lines to 100 000. I have
6870 also fixed two bugs, because there was no checking on these buffers. Tsk,
6871 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6872 buffer is too small.
6874 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6875 lsearch lookups. Now it does.
6877 46. When parsing a route_list item in a manualroute router, a fixed-length
6878 buffer was used for the list of hosts. I made this 1024 bytes long,
6879 thinking that nobody would ever have a list of hosts that long. Wrong.
6880 Somebody had a whole pile of complicated expansion conditions, and the
6881 string was silently truncated, leading to an expansion error. It turns out
6882 that it is easier to change to an unlimited length (owing to other changes
6883 that have happened since this code was originally written) than to build
6884 structure for giving a limitation error. The length of the item that
6885 expands into the list of hosts is now unlimited.
6887 47. The lsearch lookup could not handle data where the length of text line was
6888 more than 4095 characters. Such lines were truncated, leading to shortened
6889 data being returned. It should now handle lines of any length.
6891 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6892 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6895 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6896 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6897 no longer happens in comment lines. A list of replacements is now placed
6898 at the head of all of the source files, except those whose only change is
6899 to replace PERL_COMMAND in the very first #! line.
6901 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6902 messages on the queue, with a bottom-up merge sort, using code contributed
6903 by Michael Haardt. This should make operations like -bp somewhat faster on
6904 large queues. It won't affect queue runners, except when queue_run_in_order
6907 51. Installed eximstats 1.31 in the distribution.
6909 52. Added support for SRV lookups to the dnslookup router.
6911 53. If an ACL referred to $message_body or $message_body_end, the value was not
6912 reset for any messages that followed in the same SMTP session.
6914 54. The store-handling optimization for building very long strings was not
6915 differentiating between the different store pools. I don't think this
6916 actually made any difference in practice, but I've tidied it.
6918 55. While running the routers to verify a sender address, $sender_address
6919 was still set to the sender address. This is wrong, because when routing to
6920 send a bounce to the sender, it would be empty. Therefore, I have changed
6921 it so that, while verifying a sender address, $sender_address is set to <>.
6922 (There is no change to what happens when verifying a recipient address.)
6924 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6925 target A or AAAA records (if not already returned) without resetting the
6926 qualify_single or search_parents options of the DNS resolver. These are
6927 inappropriate in this case because the targets of MX and SRV records must
6928 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6929 target that, when qualified, matched something in the local domain. These
6930 two options are now turned off when doing these lookups.
6932 57. It seems that at least some releases of Reiserfs (which does not have the
6933 concept of a fixed number of inodes) returns zero and not -1 for the
6934 number of available inodes. This interacted badly with check_spool_inodes,
6935 which assumed that -1 was the "no such thing" setting. What I have done is
6936 to check that the total number of inodes is greater than zero before doing
6937 the test of how many are available.
6939 58. When a "warn" ACL statement has a log_message modifier, the message is
6940 remembered, and not repeated. This is to avoid a lot of repetition when a
6941 message has many recipients that cause the same warning to be written.
6942 However, Exim was preserving the list of already written lines for an
6943 entire SMTP session, which doesn't seem right. The memory is now reset if a
6944 new message is started.
6946 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6947 debugging output unless log_rewrite was also set.
6949 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6950 of (int)(handle) when we know that handle contains (void *)(-1).
6952 61. The Exim daemon panic-logs an error return when it closes the incoming
6953 connection. However "connection reset by peer" seems to be common, and
6954 isn't really an error worthy of noting specially, so that particular error
6957 62. When Exim is trying to find all the local interfaces, it used to panic and
6958 die if the ioctl to get the interface flags failed. However, it seems that
6959 on at least one OS (Solaris 9) it is possible to have an interface that is
6960 included in the list of interfaces, but for which you get a failure error
6961 for this call. This happens when the interface is not "plumbed" into a
6962 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6963 failure of the "get flags" call assumes that the interface is down.
6965 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6966 makes life easier for people who are doing arithmetic on fields extracted
6967 from dates, where you often get leading zeros that should not be
6968 interpreted as octal.
6970 64. Added qualify_domain to the redirect router, to override the global
6973 65. If a pathologically long header line contained very many addresses (the
6974 report of this problem mentioned 10 000) and each of them was rewritten,
6975 Exim could use up a very large amount of memory. (It kept on making new
6976 copies of the header line as it rewrote, and never released the old ones.)
6977 At the expense of a bit more processing, the header rewriting function has
6978 been changed so that it no longer eats memory in this way.
6980 66. The generation of the Received: header has been moved from the time that a
6981 message starts to be received, to the time that it finishes. The timestamp
6982 in the Received: header should now be very close to that of the <= log
6983 line. There are two side-effects of this change:
6985 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6986 logged header lines no longer include the local Received: line, because
6987 it has not yet been created. The same applies to a copy of the message
6988 that is returned to a non-SMTP sender when a message is rejected.
6990 (b) When a filter file is tested using -bf, no additional Received: header
6991 is added to the test message. After some thought, I decided that this
6994 This change does not affect the value of $received_for. It is still set
6995 after address rewriting, but before local_scan() is called.
6997 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6999 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
7000 gave an unhelpful panic error message, and a defer error. I have managed to
7001 change this behaviour so that it now rejects any supplied certificate,
7002 which seems right, as the list of acceptable certificates is empty.
7004 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
7005 gave an unhelpful defer error. I have not managed to make this reject any
7006 supplied certificates, but the error message it gives is "no certificate
7007 supplied", which is not helpful.
7009 70. exigrep's output now also includes lines that are not associated with any
7010 message, but which match the given pattern. Implemented by a patch from
7011 Martin Sluka, which also tidied up the Perl a bit.
7013 71. Recipient callout verification, like sender verification, was using <> in
7014 the MAIL FROM command. This isn't really the right thing, since the actual
7015 sender may affect whether the remote host accepts the recipient or not. I
7016 have changed it to use the actual sender in the callout; this means that
7017 the cache record is now keyed on a recipient/sender pair, not just the
7018 recipient address. There doesn't seem to be a real danger of callout loops,
7019 since a callout by the remote host to check the sender would use <>.
7020 [SEE ABOVE: changed after hitting problems.]
7022 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
7023 temporary errors. However, in the case of such a code being given after
7024 the end of a data transmission (i.e. after ".") Exim was failing to write
7025 a retry record for the message. (Yes, there was some broken host that was
7026 actually sending 8xx at this point.)
7028 73. An unknown lookup type in a host list could cause Exim to panic-die when
7029 the list was checked. (An example that provoked this was putting <; in the
7030 middle of a list instead of at the start.) If this happened during a DATA
7031 ACL check, a -D file could be left lying around. This kind of configuration
7032 error no longer causes Exim to die; instead it causes a defer error. The
7033 incident is still logged to the main and panic logs.
7035 74. Buglet left over from Exim 3 conversion. The message "too many messages
7036 in one connection" was written to the rejectlog but not the mainlog, except
7037 when address rewriting (yes!) was being logged.
7039 75. Added write_rejectlog option.
7041 76. When a system filter was run not as root (that is, when system_filter_user
7042 was set), the values of the $n variables were not being returned to the
7043 main process; thus, they were not subsequently available in the $sn
7046 77. Added +return_path_on_delivery log selector.
7048 78. A connection timeout was being treated differently from recipients deferred
7049 when testing hosts_max_try with a message that was older than the host's
7050 retry timeout. (The host should not be counted, thus allowing all hosts to
7051 be tried at least once before bouncing.) This may have been the cause of an
7052 occasionally reported bug whereby a message would remain on the queue
7053 longer than the retry timeout, but would be bounced if a delivery was
7054 forced. I say "may" because I never totally pinned down the problem;
7055 setting up timeout/retry tests is difficult. See also the next item.
7057 79. The ultimate address timeout was not being applied to errors that involved
7058 a combination of host plus message (for example, a timeout on a MAIL
7059 command). When an address resolved to a number of possible hosts, and they
7060 were not all tried for each delivery (e.g. because of hosts_max_try), a
7061 message could remain on the queue longer than the retry timeout.
7063 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
7066 81. Fixed an obscure SMTP outgoing bug which required at least the following
7067 conditions: (a) there was another message waiting for the same server;
7068 (b) the server returned 5xx to all RCPT commands in the first message so
7069 that the message was not completed; (c) the server dropped the connection
7070 or gave a negative response to the RSET that Exim sends to abort the
7071 transaction. The observed case was a dropped connection after DATA that had
7072 been sent in pipelining mode. That is, the server had advertised PIPELINING
7073 but was not implementing it correctly. The effect of the bug was incorrect
7074 behaviour, such as trying another host, and this could lead to a crash.
7080 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
7081 and daemon.c were passed as pointers to ints; they should have been
7082 pointers to socklen_t variables (which are typically unsigned ints).
7084 2. Some signed/unsigned type warnings in the os.c file for Linux have been
7087 3. Fixed a really odd bug that affected only the testing scheme; patching a
7088 certain fixed string in the binary changed the value of another string that
7089 happened to be identical to the end of the original first string.
7091 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
7092 name", it returns that address as the IP address. On some operating
7093 systems (e.g. Solaris), it also passes back the IP address string as the
7094 "host name". However, on others (e.g. Linux), it passes back an empty
7095 string. Exim wasn't checking for this, and was changing the host name to an
7096 empty string, assuming it had been canonicalized.
7098 5. Although rare, it is permitted to have more than one PTR record for a given
7099 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
7100 all the names associated with an address, because they do in Solaris.
7101 However, it seems that they do not in Linux for data that comes from the
7102 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
7103 I found this out when I moved to a new Linux workstation and tried to run
7104 the Exim test suite.
7106 To get round this problem I have changed the code so that it now does its
7107 own call to the DNS to look up PTR records when searching for a host name.
7108 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
7109 addresses that are only in /etc/hosts are still found.
7111 This behaviour is, however, controlled by an option called host_lookup_
7112 order, which defaults to "bydns:byaddr". If people want to use the other
7113 order, or indeed, just use one or the other means of lookup, they can
7114 specify it in this variable.
7116 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
7117 some operating systems, this comes back from gethostbyaddr() as an empty
7118 string, and this is what Exim used to test for. However, it seems that in
7119 other systems, "." is yielded. Exim now tests for this case too.
7121 7. The values of check_spool_space and check_log_space are now held internally
7122 as a number of kilobytes instead of an absolute number of bytes. If a
7123 numbers is specified without 'K' or 'M', it is rounded up to the nearest
7124 kilobyte. This means that much larger values can be stored.
7126 8. Exim monitor: an attempt to get the action menu when not actually pointing
7127 at a message produces an empty menu entitled "No message selected". This
7128 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
7129 no entries in it ("Shell widget menu has zero width and/or height"). So I
7130 have added a single, blank menu entry in this case.
7132 9. Added ${quote_local_part.
7134 10. MIME decoding is now applied to the contents of Subject: header lines when
7137 11. Now that a reference to $sender_host_address automatically causes a reverse
7138 lookup to occur if necessary (4.13/18), there is no need to arrange for a
7139 host lookup before query-style lookups in lists that might use this
7140 variable. This has therefore been abolished, and the "net-" prefix is no
7141 longer necessary for query-style lookups.
7143 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
7144 have been a typo for LFLAGS, so it has been changed.
7146 13. The install script calls Exim with "-C /dev/null" in order to find the
7147 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
7148 to be output. However, since Exim outputs its version number before the
7149 error, it didn't break the script. It just looked ugly. I fixed this by
7150 always allowing "-C /dev/null" if the caller is root.
7152 14. Ignore overlarge ACL variable number when reading spool file - insurance
7153 against a later release with more variables having written the file.
7155 15. The standard form for an IPv6 address literal was being rejected by EHLO.
7156 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
7157 this, as well as the form without the "IPv6" on the front.
7159 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
7160 OS/Makefile-Darwin file.
7162 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
7163 only to LDAP libraries that do not have LDAP_OPT_DEREF.
7165 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
7166 variable. However, some OS use ints for inodes. I've added cast to long int
7167 to get rid of the compiler warning.
7169 19. I had forgotten to lock out "/../" in configuration file names when
7170 ALT_CONFIG_PREFIX was set.
7172 20. Routers used for verification do not need to specify transports. However,
7173 if such a router generated a host list, and callout was configured, Exim
7174 crashed, because it could not find a port number from the (non-existent)
7175 transport. It now assumes port 25 in this circumstance.
7177 21. Added the -t option to exigrep.
7179 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
7180 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
7181 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
7182 them is set, LOOKUP_LSEARCH is forced.
7184 23. "exim -bV" now outputs a list of lookups that are included in the binary.
7186 24. Added sender and host information to the "rejected by local_scan()" log
7187 line; previously there was no indication of these.
7189 25. Added .include_if_exists.
7191 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
7192 Linux. It turns out that not all file systems support this. Apparently some
7193 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
7194 do it.) To cope with this, the error EINVAL, which means that sync-ing is
7195 not supported on the file descriptor, is now ignored when Exim is trying to
7196 sync a directory. This applies only to Linux.
7198 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
7200 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
7201 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
7204 29. Added support for maildirsize files from supplied patch (modified a bit).
7206 30. The use of :fail: followed by an empty string could lead Exim to respond to
7207 sender verification failures with (e.g.):
7209 550 Verification failed for <xxx>
7210 550 Sender verify failed
7212 where the first response line was missing the '-' that indicates it is not
7213 the final line of the response.
7215 31. The loop for finding the name of the user that called Exim had a hardwired
7216 limit of 10; it now uses the value of finduser_retries, which is used for
7217 all other user lookups.
7219 32. Added $received_count variable, available in data and not_smtp ACLs, and at
7222 33. Exim was neglecting to zero errno before one call of strtol() when
7223 expanding a string and expecting an integer value. On some systems this
7224 resulted in spurious "integer overflow" errors. Also, it was casting the
7225 result into an int without checking.
7227 34. Testing for a connection timeout using "timeout_connect" in the retry rules
7228 did not work. The code looks as if it has *never* worked, though it appears
7229 to have been documented since at least release 1.62. I have made it work.
7231 35. The "timeout_DNS" error in retry rules, also documented since at least
7232 1.62, also never worked. As it isn't clear exactly what this means, and
7233 clearly it isn't a major issue, I have abolished the feature by treating it
7234 as "timeout", and writing a warning to the main and panic logs.
7236 36. The display of retry rules for -brt wasn't always showing the error code
7239 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
7240 timeout_connect_A, timeout_connect_MX.
7242 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
7243 to the empty sender.
7245 39. The daemon was not analysing the content of -oX till after it had closed
7246 stderr and disconnected from the controlling terminal. This meant that any
7247 syntax errors were only noted on the panic log, and the return code from
7248 the command was 0. By re-arranging the code a little, I've made the
7249 decoding happen first, so such errors now appear on stderr, and the return
7250 code is 1. However, the actual setting up of the sockets still happens in
7251 the disconnected process, so errors there are still only recorded on the
7254 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
7255 connections (as happens on some IP stacks) was logged at start up time as
7256 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
7257 it from "IPv6 and IPv4", which means that two separate sockets are being
7260 41. The debug output for gethostbyname2() or getipnodebyname() failures now
7261 says whether AF_INET or AF_INET6 was passed as an argument.
7263 42. Exiwhat output was messed up when time zones were included in log
7266 43. Exiwhat now gives more information about the daemon's listening ports,
7267 and whether -tls-on-connect was used.
7269 44. The "port" option of the smtp transport is now expanded.
7271 45. A "message" modifier in a "warn" statement in a non-message ACL was being
7272 silently ignored. Now an error message is written to the main and panic
7275 46. There's a new ACL modifier called "logwrite" which writes to a log file
7276 as soon as it is encountered.
7278 47. Added $local_user_uid and $local_user_gid at routing time.
7280 48. Exim crashed when trying to verify a sender address that was being
7283 49. Exim was recognizing only a space character after ".include". It now also
7284 recognizes a tab character.
7286 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
7287 extracting the relevant information from the specification. The man page no
7288 longer contains scrambled data for the -d option, and I've added a section
7289 at the front about calling Exim under different names.
7291 51. Added "extra_headers" argument to the "mail" command in filter files.
7293 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
7296 53. Installed eximstats 1.29.
7298 54. Added transport_filter_timeout as a generic transport option.
7300 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
7301 Cc: header lines. This was required by RFC 822, but it not required by RFC
7304 56. Exim used to add From:, Date:, and Message-Id: header lines to any
7305 incoming messages that did not have them. Now it does so only if the
7306 message originates locally, that is, if there is no associated remote host
7307 address. When Resent- header lines are present, this applies to the Resent-
7308 lines rather than the non-Resent- lines.
7310 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
7311 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
7313 58. Messages for configuration errors now include the name of the main
7314 configuration file - useful now that there may be more than one file in a
7315 list (.included file names were always shown).
7317 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
7318 for those rare installations that do not start the daemon as root or run it
7319 setuid root. I've cut out the call to initgroups() if the daemon is not
7322 60. The Exim user and group can now be bound into the binary as text strings
7323 that are looked up at the start of Exim's processing.
7325 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
7327 62. Added $mailstore_basename variable.
7329 63. Installed patch to sieve.c from Michael Haardt.
7331 64. When Exim failed to open the panic log after failing to open the main log,
7332 the original message it was trying to log was written to stderr and debug
7333 output, but if they were not available (the usual case in production), it
7334 was lost. Now it is written to syslog before the two lines that record the
7335 failures to open the logs.
7337 65. Users' Exim filters run in subprocesses under the user's uid. It is
7338 possible for a "deliver" command or an alias in a "personal" command to
7339 provoke an address rewrite. If logging of address rewriting is configured,
7340 this fails because the process is not running as root or exim. There may be
7341 a better way of dealing with this, but for the moment (because 4.30 needs
7342 to be released), I have disabled address rewrite logging when running a
7343 filter in a non-root, non-exim process.
7349 1. The buildconfig auxiliary program wasn't quoting the value set for
7350 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
7351 not defined. This bug was masked in 4.22 by the effect that was fixed in
7354 2. Some messages that were rejected after a message id was allocated were
7355 shown as "incomplete" by exigrep. It no longer does this for messages that
7356 are rejected by local_scan() or the DATA or non-SMTP ACLs.
7358 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
7359 have allow_domain_literals set, the ID did not get logged in the <= line.
7360 Domain literals are now always recognized in Message-ID: header lines.
7362 4. The first argument for a ${extract expansion item is the key name or field
7363 number. Leading and trailing spaces in this item were not being ignored,
7364 causing some misleading effects.
7366 5. When deliver_drop_privilege was set, single queue runner processes started
7367 manually (i.e. by the command "exim -q") or by the daemon (which uses the
7368 same command in the process it spins off) were not dropping privilege.
7370 6. When the daemon running as "exim" started a queue runner, it always
7371 re-executed Exim in the spun-off process. This is a waste of effort when
7372 deliver_drop_privilege is set. The new process now just calls the
7373 queue-runner function directly.
7379 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
7382 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
7383 ignored. Though the use of -oP was forcing the writing of a pid file, it
7384 was always written to the default place.
7386 3. If the message "no IP address found for host xxxx" is generated during
7387 incoming verification, it is now followed by identification of the incoming
7388 connection (so you can more easily find what provoked it).
7390 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
7392 5. Added some features to "harden" Exim a bit more against certain attacks:
7394 (a) There is now a build-time option called FIXED_NEVER_USERS that can
7395 be put in Local/Makefile. This is like the never_users runtime option,
7396 but it cannot be overridden. The default setting is "root".
7398 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
7399 prefix string with which any file named in a -C command line option
7402 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
7403 is retained for -C and -D only if the caller of Exim is root. Without
7404 it, the exim user may also use -C and -D and retain privilege.
7406 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
7407 command line option is disabled.
7409 6. Macro names set by the -D option must start with an upper case letter, just
7410 like macro names defined in the configuration file.
7412 7. Added "dereference=" facility to LDAP.
7414 8. Two instances of the typo "uknown" in the source files are fixed.
7416 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
7417 the Configure-Makefile script screwed up while processing it.
7419 10. Incorporated PCRE 4.4.
7421 11. The SMTP synchronization check was not operating right at the start of an
7422 SMTP session. For example, it could not catch a HELO sent before the client
7423 waited for the greeting. There is now a check for outstanding input at the
7424 point when the greeting is written. Because of the duplex, asynchronous
7425 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
7426 way, but not yet received, when the check is performed.
7428 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
7429 on TCP/IP sockets, because this apparently causes some broken clients to
7432 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
7433 unchanged) from the Cygwin maintainer.
7435 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
7436 was supported instead of testing for mbx. Effectively a typo.
7438 15. The spa authenticator server code was not checking that the input it
7439 received was valid base64.
7441 16. The debug output line for the "set" modifier in ACLs was not showing the
7442 name of the variable that was being set.
7444 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
7446 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
7447 lookup on its own. Something else was needed to trigger the lookup. For
7448 example, a match in host_lookup or the need for a host name in a host list.
7449 Now, if $sender_host_name is referenced and the host name has not yet been
7450 looked up, a lookup is performed. If the lookup fails, the variable remains
7451 empty, and $host_lookup_failed is set to "1".
7453 19. Added "eqi" as a case-independent comparison operator.
7455 20. The saslauthd authentication condition could segfault if neither service
7456 nor realm was specified.
7458 21. If an overflowing value such as "2048M" was set for message_size_limit, the
7459 error message that was logged was misleading, and incoming SMTP
7460 connections were dropped. The message is now more accurate, and temporary
7461 errors are given to SMTP connections.
7463 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
7464 (except RSET) with a 421 error, until QUIT is received. However, it was
7465 failing to send a response to QUIT.
7467 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
7468 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
7469 ACL is now run after the helo_try_verify_hosts code.
7471 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
7472 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
7473 case-independent, so other case variants are also recognized). Apparently
7474 some systems use these upper case variants.
7476 25. If more than two messages were waiting for the same host, and a transport
7477 filter was specified for the transport, Exim sent two messages over the
7478 same TCP/IP connection, and then failed with "socket operation on non-
7479 socket" when it tried to send the third.
7481 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
7483 27. The extern definition of crypt16() in expand.c was not being excluded when
7484 the OS had its own crypt16() function.
7486 28. Added bounce_return_body as a new option, and bounce_return_size_limit
7487 as a preferred synonym for return_size_limit, both as an option and as an
7490 29. Added LIBS=-liconv to OS/Makefile-OSF1.
7492 30. Changed the default configuration ACL to relax the local part checking rule
7493 for addresses that are not in any local domains. For these addresses,
7494 slashes and pipe symbols are allowed within local parts, but the sequence
7495 /../ is explicitly forbidden.
7497 31. SPA server authentication was not clearing the challenge buffer before
7500 32. log_message in a "warn" ACL statement was writing to the reject log as
7501 well as to the main log, which contradicts the documentation and doesn't
7502 seem right (because no rejection is happening). So I have stopped it.
7504 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
7505 However, I am unable to do any testing of this.
7507 34. Fixed an infelicity in the appendfile transport. When checking directories
7508 for a mailbox, to see if any needed to be created, it was accidentally
7509 using path names with one or more superfluous leading slashes; tracing
7510 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
7512 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
7513 contents are added to the log line that is written for every discarded
7514 recipient. (Previously a log_message setting was ignored.)
7516 36. The ${quote: operator now quotes the string if it is empty.
7518 37. The install script runs exim in order to find its version number. If for
7519 some reason other than non-existence or emptiness, which it checks, it
7520 could not run './exim', it was installing it with an empty version number,
7521 i.e. as "exim-". This error state is now caught, and the installation is
7524 38. An argument was missing from the function that creates an error message
7525 when Exim fails to connect to the socket for saslauthd authentication.
7526 This could cause Exim to crash, or give a corrupted message.
7528 39. Added isip, isip4, and isip6 to ${if conditions.
7530 40. The ACL variables $acl_xx are now saved with the message, and can be
7531 accessed later in routers, transports, and filters.
7533 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
7534 strings in the file are not string-expanded.
7536 42. If a MAIL command specified a SIZE value that was too large to fit into an
7537 int variable, the check against message_size_limit failed. Such values are
7538 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
7539 day this will have to be increased, but I don't think I want to be around
7540 when emails are that large.
7547 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
7548 iconv() is not standard in FreeBSD.
7550 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
7551 IPv6 enabled. The observed symptom was a segmentation fault on return from
7552 the function os_common_find_running_interfaces() in src/os.c.
7554 3. In the check_special_case() function in daemon.c I had used "errno" as an
7555 argument name, which causes warnings on some systems. This was basically a
7556 typo, since it was named "eno" in the comments!
7558 4. The code that waits for the clock to tick (at a resolution of some fraction
7559 of a second) so as to ensure message-id uniqueness was always waiting for
7560 at least one whole tick, when it could have waited for less. [This is
7561 almost certainly not relevant at current processor speeds, where it is
7562 unlikely to ever wait at all. But we try to future-proof.]
7564 5. The function that sleeps for a time interval that includes fractions of a
7565 second contained a race. It did not block SIGALRM between setting the
7566 timer, and suspending (a couple of lines later). If the interval was short
7567 and the sigsuspend() was delayed until after it had expired, the suspension
7568 never ended. On busy systems this could lead to processes getting stuck for
7571 6. Some uncommon configurations may cause a lookup to happen in a queue runner
7572 process, before it forks any delivery processes. The open lookup caching
7573 mechanism meant that the open file or database connection was passed into
7574 the delivery process. The problem was that delivery processes always tidy
7575 up cached lookup data. This could cause a problem for the next delivery
7576 process started by the queue runner, because the external queue runner
7577 process does not know about the closure. So the next delivery process
7578 still has data in the lookup cache. In the case of a file lookup, there was
7579 no problem because closing a file descriptor in a subprocess doesn't affect
7580 the parent. However, if the lookup was caching a connection to a database,
7581 the connection was closed, and the second delivery process was likely to
7582 see errors such as "PGSQL: query failed: server closed the connection
7583 unexpectedly". The problem has been fixed by closing all cached lookups
7584 in a queue runner before running a delivery process.
7586 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
7587 seem to have the "const" qualifier which it has on other OS. I've
7590 8. Change 4.21/2 was too strict. It is only if there are two authenticators
7591 *of the same type* (client or server) with the same public name that an
7592 error should be diagnosed.
7594 9. When Exim looked up a host name for an IP address, but failed to find the
7595 original IP address when looking up the host name (a safety check), it
7596 output the message "<ip address> does not match any IP for NULL", which was
7597 confusing, to say the least. The bug was that the host name should have
7598 appeared instead of "NULL".
7600 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
7601 user that is built into the binary, and the -C or -D options is used, root
7602 privilege is dropped before the configuration file is read. In addition,
7603 logging is switched to stderr instead of the normal log files. If the
7604 configuration then re-defines the Exim user, the unprivileged environment
7605 is probably not what is expected, so Exim logs a panic warning message (but
7608 However, if deliver_drop_privilege is set, the unprivileged state may well
7609 be exactly what is intended, so the warning has been cut out in that case,
7610 and Exim is allowed to try to write to its normal log files.
7616 1. smtp_return_error_details was not giving details for temporary sender
7617 or receiver verification errors.
7619 2. Diagnose a configuration error if two authenticators have the same public
7622 3. Exim used not to create the message log file for a message until the first
7623 delivery attempt. This could be confusing when incoming messages were held
7624 for policy or load reasons. The message log file is now created at the time
7625 the message is received, and an initial "Received" line is written to it.
7627 4. The automatically generated man page for command line options had a minor
7628 bug that caused no ill effects; however, a more serious problem was that
7629 the procedure for building the man page automatically didn't always
7630 operate. Consequently, release 4.20 contains an out-of-date version. This
7631 shouldn't happen again.
7633 5. When building Exim with embedded Perl support, the script that builds the
7634 Makefile was calling 'perl' to find its compile-time parameters, ignoring
7635 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
7637 6. The freeze_tell option was not being used for messages that were frozen on
7638 arrival, either by an ACL or by local_scan().
7640 7. Added the smtp_incomplete_transaction log selector.
7642 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
7643 was accepting AUTH without a new EHLO.
7645 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
7646 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
7649 10. Logging of TCP/IP connections (when configured) now happens in the main
7650 daemon process instead of the child process, so that the TCP/IP connection
7651 count is more accurate (but it can never be perfect).
7653 11. The use of "drop" in a nested ACL was not being handled correctly in the
7654 outer ACL. Now, if condition failure induced by the nested "drop" causes
7655 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
7656 or "require"), the connection is dropped.
7658 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
7659 that yield "discard" can now be used with an "accept" or a "discard" verb,
7660 but an error is generated for any others (because I can't see a useful way
7661 to define what should happen).
7663 13. When an ACL is read dynamically from a file (or anywhere else), the lines
7664 are now processed in the same way as lines in the Exim configuration file.
7665 In particular, continuation lines are supported.
7667 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
7669 15. Added -ti meaning -t -i.
7671 16. Check for letters, digits, hyphens, and dots in the names of dnslist
7672 domains, and warn by logging if others are found.
7674 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
7675 returned from GIFCONF when Exim is trying to find the list of interfaces on
7676 a host. The code in os.c has been modified to copy each ifreq to an aligned
7677 structure in all cases.
7679 Also, in some cases, the returned ifreq's were being copied to a 'struct
7680 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
7681 means the last couple of bytes of an IPv6 address could be chopped if the
7682 ifreq contained only a normal sockaddr (14 bytes storage).
7684 18. Named domain lists were not supported in the hosts_treat_as_local option.
7685 An entry such as +xxxx was not recognized, and was treated as a literal
7688 19. Ensure that header lines added by a DATA ACL are included in the reject log
7689 if the ACL subsequently rejects the message.
7691 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
7692 MD5 (which is deprecated).
7694 21. When testing a filter file using -bf, Exim was writing a message when it
7695 took the sender from a "From " line in the message, but it was not doing so
7696 when it took $return_path from a Return-Path: header line. It now does.
7698 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
7699 with a valid header line field name (a series of printing characters
7700 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
7702 23. Changed "disc" in the source to "disk" to conform to the documentation and
7703 the book and for uniformity.
7705 24. Ignore Sendmail's -Ooption=value command line item.
7707 25. When execve() failed while trying to run a command in a pipe transport,
7708 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
7709 could be confused with a return value of 69 from the command itself. This
7710 has been changed to 127, the value the shell returns if it is asked to run
7711 a non-existent command. The wording for the related log line suggests a
7712 non-existent command as the problem.
7714 26. If received_header_text expands to an empty string, do not add a Received:
7715 header line to the message. (Well, it adds a token one on the spool, but
7716 marks it "old" so that it doesn't get used or transmitted.)
7718 27. Installed eximstats 1.28 (addition of -nt option).
7720 28. There was no check for failure on the call to getsockname() in the daemon
7721 code. This can fail if there is a shortage of resources on the system, with
7722 ENOMEM, for example. A temporary error is now given on failure.
7724 29. Contrary to the C standard, it seems that in some environments, the
7725 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
7726 program. Exim now does this explicitly; it affects the formatting of
7727 timestamps using strftime().
7729 30. If exiqsumm was given junk data, it threw up some uninitialized variable
7730 complaints. I've now initialized all the variables, to avoid this.
7732 32. Header lines added by a system filter were not being "seen" during
7733 transport-time rewrites.
7735 33. The info_callback() function passed to OpenSSL is set up with type void
7736 (*)(SSL *, int, int), as described somewhere. However, when calling the
7737 function (actually a macro) that sets it up, the type void(*)() is
7738 expected. I've put in a cast to prevent warnings from picky compilers.
7740 34. If a DNS black list lookup found a CNAME record, but there were no A
7741 records associated with the domain it pointed at, Exim crashed.
7743 35. If a DNS black list lookup returned more than one A record, Exim ignored
7744 all but the first. It now scans all returned addresses if a particular IP
7745 value is being sought. In this situation, the contents of the
7746 $dnslist_value variable are a list of all the addresses, separated by a
7749 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
7750 to accept a host name and all its aliases if the forward lookup for any of
7751 them yielded the IP address of the incoming connection. Now it accepts only
7752 those names whose forward lookup yields the correct IP address. Any other
7753 names are discarded. This closes a loophole whereby a rogue DNS
7754 administrator could create reverse DNS records to break through a
7755 wildcarded host restriction in an ACL.
7757 37. If a user filter or a system filter that ran in a subprocess used any of
7758 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
7759 the wrong values were passed to the pipe command ($thisaddress had the
7760 value of $0, $0 had the value of $1, etc). This bug was introduced by
7761 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
7763 38. Improved the line breaking for long SMTP error messages from ACLs.
7764 Previously, if there was no break point between 40 and 75 characters, Exim
7765 left the rest of the message alone. Two changes have been made: (a) I've
7766 reduced the minimum length to 35 characters; (b) if it can't find a break
7767 point between 35 and 75 characters, it looks ahead and uses the first one
7768 that it finds. This may give the occasional overlong line, but at least the
7769 remaining text gets split now.
7771 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
7772 file descriptors might be low, and that setting 1000 would always raise it.
7773 It turns out that in some environments, the limit is already over 1000 and
7774 that lowering it causes trouble. So now Exim takes care not to decrease it.
7776 40. When delivering a message, the value of $return_path is set to $sender_
7777 address at the start of routing (routers may change the value). By an
7778 oversight, this default was not being set up when an address was tested by
7779 -bt or -bv, which affected the outcome if any router or filter referred to
7782 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
7783 log only when "message" or "log_message" are set. However, if one of the
7784 conditions was an address verification, or a call to a nested ACL, the
7785 messages generated by the underlying test were being passed through. This
7786 no longer happens. The underlying message is available in $acl_verify_
7787 message for both "message" and "log_message" expansions, so it can be
7788 passed through if needed.
7790 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
7791 new expansion $bh_ to give the encoded byte string without charset
7792 translation. Translation happens only if iconv() is available; HAVE_ICONV
7793 indicates this at build time. HEADERS_CHARSET gives the charset to
7794 translate to; headers_charset can change it in the configuration, and
7795 "headers charset" can change it in an individual filter file.
7797 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
7798 that creates RFC 2047 encoded "words" labels them as that charset instead
7799 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
7800 expansion operator; (ii) when Exim creates a From: line for a local
7801 message; (iii) when a header line is rewritten to include a "phrase" part.
7803 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
7804 buggy, causing it to skip the first lines of messages whose message ID
7805 ended in 'D'. This would not have bitten before Exim release 4.14, because
7806 message IDs were unlikely to end in 'D' before then. The effect was to have
7807 incorrect size information for certain domains.
7809 45. #include "config.h" was missing at the start of the crypt16.c module. This
7810 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
7813 46. If there was a timeout during a "random" callout check, Exim treated it as
7814 a failure of the random address, and carried on sending RSET and the real
7815 address. If the delay was just some slowness somewhere, the response to the
7816 original RCPT would be taken as a response to RSET and so on, causing
7817 mayhem of various kinds.
7819 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
7820 when I implemented it. It didn't allow for the fact that some option values
7821 may legitimately be negative (e.g. size_addition), and it didn't even do
7822 the right test for positive values.
7824 48. Domain names in DNS records are case-independent. Exim always looks them up
7825 in lower case. Some resolvers return domain names in exactly the case they
7826 appear in the zone file, that is, they may contain uppercase letters. Not
7827 all resolvers do this - some return always lower case. Exim was treating a
7828 change of case by a resolver as a change of domain, similar to a widening
7829 of a domain abbreviation. This triggered its re-routing code and so it was
7830 trying to route what was effectively the same domain again. This normally
7831 caused routing to fail (because the router wouldn't handle the domain
7832 twice). Now Exim checks for this case specially, and just changes the
7833 casing of the domain that it ultimately uses when it transmits the message
7836 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
7839 50. If a filter generated a file delivery with a non-absolute name (possible if
7840 no home directory exists for the router), the forbid_file option was not
7843 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
7844 the existing equality matching.
7846 52. Exim was using ints instead of ino_t variables in some places where it was
7847 dealing with inode numbers.
7849 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
7850 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
7851 called TMPDIR, and if it finds it is different, it changes its value.
7853 54. The smtp_printf() function is now made available to local_scan() so
7854 additional output lines can be written before returning. There is also an
7855 smtp_fflush() function to enable the detection of a dropped connection.
7856 The variables smtp_input and smtp_batched_input are exported to
7859 55. Changed the default runtime configuration: the message "Unknown user"
7860 has been removed from the ACL, and instead placed on the localuser router,
7861 using the cannot_route_message feature. This means that any verification
7862 failures that generate their own messages won't get overridden. Similarly,
7863 the "Unrouteable address" message that was in the ACL for unverifiable
7864 relay addresses has also been removed.
7866 56. Added hosts_avoid_esmtp to the smtp transport.
7868 57. The exicyclog script was not checking for the esoteric option
7869 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7870 will work only if exicyclog is run under the appropriate euid.
7872 58. Following a discussion on the list, the rules by which Exim recognises line
7873 endings on incoming messages have been changed. The -dropcr and drop_cr
7874 options are now no-ops, retained only for backwards compatibility. The
7875 following line terminators are recognized: LF CRLF CR. However, special
7876 processing applies to CR:
7878 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7879 nor a local message in the state where . is a terminator.
7881 (ii) If a bare CR is encountered in a header line, an extra space is added
7882 after the line terminator so as not to end the header. The reasoning
7883 behind this is that bare CRs in header lines are most likely either
7884 to be mistakes, or people trying to play silly games.
7886 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7887 was being incorrectly given as 18 bytes larger than it should have been.
7888 This is a VOB (very old bug).
7890 60. This may never have affected anything current, but just in case it has:
7891 When the local host is found other than at the start of a list of hosts,
7892 the local host, those with the same MX, and any that follow, are discarded.
7893 When the list in question was part of a longer list of hosts, the following
7894 hosts (not currently being processed) were also being discarded. This no
7895 longer happens. I'm not sure if this situation could ever has previously
7898 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7901 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7902 header. This is something that is recommended in a new Internet Draft, and
7903 is something that is documented as being done by Sendmail. There are two
7904 possible values. For messages generated by the autoreply transport, Exim
7907 Auto-Submitted: auto-replied
7909 whereas for all other generated messages (e.g. bounces) it adds
7911 Auto-Submitted: auto-generated
7913 63. The "personal" condition in filters now includes a test for the
7914 Auto-Submitted: header. If it contains the string "auto-" the message it
7915 not considered personal.
7917 64. Added rcpt_include_affixes as a generic transport option.
7919 65. Added queue_only_override (default true).
7921 66. Added the syslog_duplication option.
7923 67. If what should have been the first header line of a message consisted of
7924 a space followed by a colon, Exim was mis-interpreting it as a header line.
7925 It isn't of course - it is syntactically invalid and should therefore be
7926 treated as the start of the message body. The misbehaviour could have
7927 caused a number of strange effects, including loss of data in subsequent
7928 header lines, and spool format errors.
7930 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7931 client host had authenticated. This control can now be exercised by an ACL
7932 for more flexibility.
7934 69. By default, callouts do not happen when testing with -bh. There is now a
7935 variant, -bhc, which does actually run the callout code, including
7936 consulting and updating the callout cache.
7938 70. Added support for saslauthd authentication, courtesy of Alexander
7941 71. If statvfs() failed on the spool or log directories while checking their
7942 size for availability, Exim confusingly gave the error "space shortage".
7943 Furthermore, in debugging mode it crashed with a floating point exception.
7944 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7945 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7946 serious problem, Exim now writes to the main and panic logs when this
7947 happens, with details of the failure. It then refuses to accept the
7948 incoming message, giving the message "spool directory problem" or "log
7949 directory problem" with a 421 code for SMTP messages.
7951 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7952 0, 1, and 2 exist, because some OS complain for execs without them (see
7953 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7954 descriptors. However, the code omitted to check that the open succeeded,
7955 causing mysterious errors if for some reason the permissions on /dev/null
7956 got screwed. Now Exim writes a message to the main and panic logs, and
7957 bombs out if it can't open /dev/null.
7959 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7960 interact so that it is all more flexible. It is supposed to remain
7961 backwards compatible. Also added extra_local_interfaces.
7963 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7964 to bomb out with an assertion failure - to the client this appears as a
7965 connection drop. This problem occurs in the part of the code that was taken
7966 from the Samba project. Fortunately, the assertion is in a very simple
7967 function, so I have fixed this by reproducing the function inline in the
7968 one place where it is called, and arranging for authentication to fail
7969 instead of killing the process with assert().
7971 75. The SPA client code was not working when the server requested OEM rather
7972 than Unicode encoding.
7974 76. Added code to make require_files with a specific uid setting more usable in
7975 the case where statting the file as root fails - usually a non-root-mounted
7976 NFS file system. When this happens and the failure is EACCES, Exim now
7977 forks a subprocess and does the per-uid checking as the relevant uid.
7979 77. Added process_log_path.
7981 78. If log_file_path was not explicitly set, a setting of check_log_space or
7982 check_log_inodes was ignored.
7984 79. If a space check for the spool or log partitions fails, the incident is now
7985 logged. Of course, in the latter case the data may get lost...
7987 80. Added the %p formatting code to string_format() so that it can be used to
7988 print addresses in debug_print(). Adjusted all the address printing in the
7989 debugging in store.c to use %p rather than %d.
7991 81. There was a concern that a line of code in smtp_in.c could overflow a
7992 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7993 initially expressed, the concern was not well-founded, because trailing
7994 spaces are removed early. However, if the trailing spaces were followed by
7995 a NULL, they did not get removed, so the overflow was possible. Two fixes
7998 (a) I re-wrote the offending code in a cleaner fashion.
7999 (b) If an incoming SMTP command contains a NULL character, it is rejected
8002 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
8003 runs initgroups(), so that if the Exim user is in any additional groups,
8004 they will be used during message reception.
8010 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob