1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.403 2006/10/03 15:11:22 ph10 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 TF/01 In the add_headers option to the mail command in an Exim filter,
10 there was a bug that Exim would claim a syntax error in any
11 header after the first one which had an odd number of characters
14 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
15 callout verification, Exim cached a "reject" for the entire domain. This
16 is correct for most verifications, but it is not correct for a recipient
17 verification with use_sender or use_postmaster set, because in that case
18 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
19 case of MAIL FROM:<> rejection from other early rejections (e.g.
20 rejection of HELO). When verifying a recipient using a non-null MAIL
21 address, the cache is ignored if it shows MAIL FROM:<> rejection.
22 Whatever the result of the callout, the value of the domain cache is
23 left unchanged (for any other kind of callout, getting as far as trying
24 RCPT means that the domain itself is ok).
26 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
29 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
30 manifest itself as EPIPE rather than ECONNECT. When tidying away a
31 session, the daemon ignores ECONNECT errors and logs others; it now
32 ignores EPIPE as well.
34 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
35 (quoted-printable decoding).
37 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
38 later the small subsequent patch to fix an introduced bug.
40 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
42 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
44 PH/08 An error is now given if message_size_limit is specified negative.
46 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
47 to be given (somewhat) arbitrary names.
49 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
52 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
53 miscellaneous code fixes
55 PH/10 Added the log_reject_target ACL modifier to specify where to log
58 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
59 hostname. This is wrong, because it relates to the incoming message (and
60 probably the interface on which it is arriving) and not to the outgoing
61 callout (which could be using a different interface). This has been
62 changed to use the value of the helo_data option from the smtp transport
63 instead - this is what is used when a message is actually being sent. If
64 there is no remote transport (possible with a router that sets up host
65 addresses), $smtp_active_hostname is used.
67 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
68 tweaks were necessary in order to get it to work:
69 (a) The code assumed that strncpy() returns a negative number on buffer
70 overflow, which isn't the case. Replaced with Exim's string_format()
72 (b) There were several signed/unsigned issues. I just did the minimum
73 hacking in of casts. There is scope for a larger refactoring.
74 (c) The code used strcasecmp() which is not a standard C function.
75 Replaced with Exim's strcmpic() function.
76 (d) The code set only $1; it now sets $auth1 as well.
77 (e) A simple test gave the error "authentication client didn't specify
78 service in request". It would seem that Dovecot has changed its
79 interface. Fortunately there's a specification; I followed it and
80 changed what the client sends and it appears to be working now.
82 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
85 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
86 address A is aliased to B and C, where B exists and C does not. Without
87 -v the output is "A verified" because verification stops after a
88 successful redirection if more than one address is generated. However,
89 with -v the child addresses are also verified. Exim was outputting "A
90 failed to verify" and then showing the successful verification for C,
91 with its parentage. It now outputs "B failed to verify", showing B's
92 parentage before showing the successful verification of C.
94 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
95 look up a TXT record in a specific list after matching in a combined
102 SC/01 Use a glob alias rather than an array ref in eximstats generated
103 parser. This improves both readability and performance.
105 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
106 Don't display local sender or destination tables in eximstats unless
107 there is data to show.
108 Added average volumes into the eximstats top table text output.
110 SC/03 Collect data on the number of addresses (recipients) as well
111 as the number of messages in eximstats.
113 TF/01 Correct an error in the documentation for the redirect router. Exim
114 does (usually) call initgroups() when daemonizing.
116 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
117 with consistent privilege compared to when running as a daemon.
119 TF/03 Note in the spec that $authenticated_id is not set for local
120 submissions from trusted users.
122 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
123 Thanks to Dean Brooks <dean@iglou.com> for the patch.
125 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
126 by adding some example configuration directives to the default
127 configuration file. A little bit of work is required to uncomment the
128 directives and define how usernames and passwords are checked, but
129 there is now a framework to start from.
131 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
132 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
133 without this. I don't know how relevant this is to other LDAP libraries.
135 PH/02 Add the verb name to the "unknown ACL verb" error.
137 PH/03 Magnus Holmgren's patch for filter_prepend_home.
139 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
141 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
142 directory not expanded when it should be if an expanded home directory
143 was set for the address (which is overridden by the transport).
145 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
148 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
149 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
150 because it is too late at that time, and has no effect.
152 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
153 security issue with \' (bugzilla #107). I could not use the
154 PQescapeStringConn() function, because it needs a PGconn value as one of
157 PH/08 When testing addresses using -bt, indicate those final addresses that
158 are duplicates that would not cause an additional delivery. At least one
159 person was confused, thinking that -bt output corresponded to deliveries.
160 (Suppressing duplicates isn't a good idea as you lose the information
161 about possibly different redirections that led to the duplicates.)
163 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
164 systems where poll() doesn't work, in particular OS X.
166 PH/10 Added more information to debugging output for retry time not reached.
168 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
169 operations in malware.c.
171 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
174 PH/13 If write_rejectlog was set false when logging was sent to syslog with
175 syslog_duplication set false, log lines that would normally be written
176 both the the main log and to the reject log were not written to syslog at
179 PH/14 In the default configuration, change the use of "message" in ACL warn
180 statements to "add_header".
182 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
183 not followed by a command (e.g. "seen endif").
185 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
186 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
189 PH/17 Added extra conditions to the default value of delay_warning_condition
193 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
194 { match{$h_precedence:}{(?i)bulk|list|junk} } \
195 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
198 The Auto-Submitted: and various List- headers are standardised, whereas I
199 don't think Precedence: ever was.
201 PH/18 Refactored debugging code in route_finduser() to show more information,
202 in particular, the error code if getpwnam() issues one.
204 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
205 This is apparently needed in addition to the PH/07 change above to avoid
206 any possible encoding problems.
208 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
209 but not after initializing Perl.
211 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
212 output them only if debugging. By default they are written stderr,
213 apparently, which is not desirable.
215 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
218 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
221 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
223 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
224 authenticated or an ident call has been made. Suppress the default
225 values for $authenticated_id and $authenticated_sender (but permit -oMai
226 and -oMas) when testing with -bh.
228 PH/24 Re-jigged the order of the tests in the default configuration so that the
229 tests for valid domains and recipients precede the DNS black list and CSA
230 tests, on the grounds that those ones are more expensive.
232 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
233 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
234 This bug exists in every version of Exim that I still have, right back to
237 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
238 However, an attempt to turn on case-sensitivity in a regex key by
239 including (?-i) didn't work because the subject string was already
240 lowercased, and the effects were non-intuitive. It turns out that a
241 one-line patch can be used to allow (?-i) to work as expected.
247 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
248 other effects) broke the use of negated acl sub-conditions.
250 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
253 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
254 "Deny" causes Exim to reject the incoming connection with a 554 error.
255 Unfortunately, if there is a major crisis, such as a disk failure,
256 tcp-wrappers gives "deny", whereas what one would like would be some
257 kind of temporary error. A kludge has been added to help with this.
258 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
259 554 error is used if errno is still zero or contains ENOENT (which occurs
260 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
263 PH/03 Add -lutil to the default FreeBSD LIBS setting.
265 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
266 errors. Otherwise a message that provokes a temporary error (when other
267 messages do not) can cause a whole host to time out.
269 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
270 addresses were routed directly to files or pipes from a redirect router.
271 File deliveries just didn't batch; pipe deliveries might have suffered
274 PH/06 A failure to get a lock for a hints database would erroneously always say
275 "Failed to get write lock", even when it was really a read lock.
277 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
278 of 0600. This has been changed to use the value of the lockfile_mode
279 option (which defaults to 0600).
281 PH/08 Applied small patch from the Sieve maintainer.
283 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
284 folder from quota calculations, a direct delivery into this folder messed
285 up the contents of the maildirsize file. This was because the regex was
286 used only to exclude .Trash (or whatever) when the size of the mailbox
287 was calculated. There was no check that a delivery was happening into an
288 excluded directory. This bug has been fixed by ignoring all quota
289 processing for deliveries into excluded directories.
291 PH/10 Added the maildirfolder_create_regex option to appendfile.
297 PH/01 The code for finding all the local interface addresses on a FreeBSD
298 system running IPv6 was broken. This may well have applied to all BSD
299 systems, as well as to others that have similar system calls. The broken
300 code found IPv4 interfaces correctly, but gave incorrect values for the
301 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
302 that it would not match correctly against @[] and not recognize the IPv6
305 PH/02 The ipliteral router was not recognizing addresses of the form user@
306 [ipv6:....] because it didn't know about the "ipv6:" prefix.
308 PH/03 Added disable_ipv6.
310 PH/04 Changed $reply_address to use the raw form of the headers instead of the
311 decoded form, because it is most often used to construct To: headers
312 lines in autoreplies, and the decoded form may well be syntactically
313 invalid. However, $reply_address has leading white space removed, and all
314 newlines turned into spaces so that the autoreply transport does not
317 PH/05 If group was specified without a user on a router, and no group or user
318 was specified on a transport, the group from the router was ignored.
320 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
321 for visible compile-time settings that can be used to change these
322 numbers, for those that want even more. Backwards compatibility with old
323 spool files has been maintained. However, going back to a previous Exim
324 release will lost any variables that are in spool files.
326 PH/07 Two small changes when running in the test harness: increase delay when
327 passing a TCP/IP connection to a new process, in case the original
328 process has to generate a bounce, and remove special handling of
329 127.0.0.2 (sic), which is no longer necessary.
331 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
332 be the same on different OS.
334 PH/09 Moved a debug statement in filter processing to avoid a race problem when
337 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
338 whether --show-vars was specified or not
340 JJ/02 exipick: Added support for new ACL variable spool format introduced
343 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
344 syntactically invalid From: or Reply-to: line, and a filter used this to
345 generate an autoreply, and therefore failed to obtain an address for the
346 autoreply, Exim could try to deliver to a non-existent relative file
347 name, causing unrelated and misleading errors. What now happens is that
348 it logs this as a hard delivery error, but does not attempt to create a
351 PH/11 The exinext utility has a -C option for testing purposes, but although
352 the given file was scanned by exinext itself; it wasn't being passed on
353 when Exim was called.
355 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
356 an end-of-file indication when reading a command response.
358 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
359 compiled. In many other places in Exim, IPv6 addresses are always
360 recognized, so I have changed this. It also means that IPv4 domain
361 literals of the form [IPV4:n.n.n.n] are now always recognized.
363 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
364 used if the router is not running as root, for example, when verifying at
365 ACL time, or when using -bh. The debugging output from this situation was
366 non-existent - all you got was a failure to exec. I have made two
369 (a) Failures to set uid/gid, the current directory, or a process leader
370 in a subprocess such as that created by queryprogram now generate
371 suitable debugging ouput when -d is set.
373 (b) The queryprogram router detects when it is not running as root,
374 outputs suitable debugging information if -d is set, and then runs
375 the subprocess without attempting to change uid/gid.
377 PH/15 Minor change to Makefile for building test_host (undocumented testing
380 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
381 additional section of a DNS packet that returns MX or SRV records.
382 Instead, it always explicitly searches for A/AAAA records. This avoids
383 major problems that occur when a DNS server includes only records of one
384 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
385 fixed another bug: if SRV records were looked up and the corresponding
386 address records were *not* found in the additional section, the port
387 values from the SRV records were lost.
389 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
390 using the correct key (the original address) when searching the retry
391 rules in order to find which one to use for generating the retry hint.
393 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
394 adding the default one. Similarly, if it contains a Reply-To: header, the
395 errors_reply_to option, if set, is not used.
397 PH/19 When calculating a retry time, Exim used to measure the "time since
398 failure" by looking at the "first failed" field in the retry record. Now
399 it does not use this if it is later than than the arrival time of the
400 message. Instead it uses the arrival time. This makes for better
401 behaviour in cases where some deliveries succeed, thus re-setting the
402 "first failed" field. An example is a quota failure for a huge message
403 when small messages continue to be delivered. Without this change, the
404 "time since failure" will always be short, possible causing more frequent
405 delivery attempts for the huge message than are intended.
406 [Note: This change was subsequently modified - see PH/04 for 4.62.]
408 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
409 $1, $2, $3) because the numerical variables can be reset during some
410 expansion items (e.g. "match"), thereby losing the authentication data.
412 PH/21 Make -bV show the size of off_t variables so that the test suite can
413 decide whether to run tests for quotas > 2G.
415 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
416 mailbox_size, and mailbox_filecount in the appendfile transport. If a
417 filecount value is greater than 2G or if a quota value is greater than 2G
418 on a system where the size of off_t is not greater than 4, a panic error
421 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
422 never match. The debug and -bh output now contains an explicit error
423 message indicating a malformed IPv4 address or mask.
425 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
426 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
429 PH/25 Do not write to syslog when running in the test harness. The only
430 occasion when this arises is a failure to open the main or panic logs
431 (for which there is an explicit test).
433 PH/26 Added the /no_tell option to "control=freeze".
435 PH/27 If a host name lookup failed very early in a connection, for example, if
436 the IP address matched host_lookup and the reverse lookup yielded a name
437 that did not have a forward lookup, an error message of the form "no IP
438 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
439 could be logged. Now it outputs the IP address instead of "NULL".
441 PH/28 An enabling patch from MH: add new function child_open_exim2() which
442 allows the sender and the authenticated sender to be set when
443 submitting a message from within Exim. Since child_open_exim() is
444 documented for local_scan(), the new function should be too.
446 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
447 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
448 results in an empty string is now treated as unset.
450 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
452 PH/31 Added sender_verify_fail logging option.
454 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
455 needed by RFC 822 but not by RFC 2822 was commented out. I have now
456 tidied the source and removed it altogether.
458 PH/33 When a queue run was abandoned because the load average was too high, a
459 log line was always written; now it is written only if the queue_run log
460 selector is set. In addition, the log line for abandonment now contains
461 information about the queue run such as the pid. This is always present
462 in "start" and "stop" lines but was omitted from the "abandon" line.
464 PH/34 Omit spaces between a header name and the colon in the error message that
465 is given when verify = headers_syntax fails (if there are lots of them,
466 the message gets confusing).
468 PH/35 Change the default for dns_check_names_pattern to allow slashes within
469 names, as there are now some PTR records that contain slashes. This check
470 is only to protect against broken name servers that fall over on strange
471 characters, so the fact that it applies to all lookups doesn't matter.
473 PH/36 Now that the new test suite is complete, we can remove some of the
474 special code in Exim that was needed for the old test suite. For example,
475 sorting DNS records because real resolvers return them in an arbitrary
476 order. The new test suite's fake resolver always returns records in the
479 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
480 bounces) except when queue_only is set, to avoid logging races between
481 the different processes.
483 PH/38 Panic-die if .include specifies a non-absolute path.
485 PH/39 A tweak to the "H" retry rule from its user.
487 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
488 a label. They prevented compilation on older perls.
490 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
491 a warning to be raised on newish perls.
493 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
494 on queue. Changes to match documented behaviour of showing count of
495 messages matching specified criteria.
497 PH/40 Changed the default ident timeout from 30s to 5s.
499 PH/41 Added support for the use of login_cap features, on those BSD systems
500 that have them, for controlling the resources used by pipe deliveries.
502 PH/42 The content-scanning code uses fopen() to create files in which to put
503 message data. Previously it was not paying any attention to the mode of
504 the files. Exim runs with umask(0) because the rest of the code creates
505 files with open(), and sets the required mode explicitly. Thus, these
506 files were ending up world-writeable. This was not a big issue, because,
507 being within the spool directory, they were not world-accessible. I have
508 created a function called modefopen, which takes an additional mode
509 argument. It sets umask(777), creates the file, chmods it to the required
510 mode, then resets the umask. All the relevant calls to fopen() in the
511 content scanning code have been changed to use this function.
513 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
514 to 24 hours. This avoids potential overflow problems when processing G
515 and H retry rules. I suspect nobody ever tinkers with this value.
517 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
519 PH/45 When the plaintext authenticator is running as a client, the server's
520 challenges are checked to ensure they are valid base64 strings. By
521 default, the authentication attempt is cancelled if an invalid string is
522 received. Setting client_ignore_invalid_base64 true ignores these errors.
523 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
524 they are received. Thus, the responses can be made to depend on the
525 challenges. If an invalid string is ignored, an empty string is placed in
528 PH/46 Messages that are created by the autoreply transport now contains a
529 References: header, in accordance with RFCs 2822 and 3834.
531 PH/47 Added authenticated_sender_force to the smtp transport.
533 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
535 PH/49 Installed latest patch from the Sieve maintainer.
537 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
538 was also set, the appendfile transport was unnecessarily scanning a
539 directory of message files (e.g. for maildir delivery) to find the count
540 of files (along with the size), even though it did not need this
541 information. It now does the scan only if it needs to find either the
542 size of the count of files.
544 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
546 PH/52 Two bugs concerned with error handling when the smtp transport is
549 (i) Exim was not creating retry information for temporary errors given
550 for individual recipients after the DATA command when the smtp transport
551 was used in LMTP mode. This meant that they could be retried too
552 frequently, and not timed out correctly.
554 (ii) Exim was setting the flag that allows error details to be returned
555 for LMTP errors on RCPT commands, but not for LMTP errors for individual
556 recipients that were returned after the DATA command.
558 PH/53 This is related to PH/52, but is more general: for any failing address,
559 when detailed error information was permitted to be returned to the
560 sender, but the error was temporary, then after the final timeout, only
561 "retry timeout exceeded" was returned. Now it returns the full error as
562 well as "retry timeout exceeded".
564 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
565 do this, and (what is worse) MTAs that accept it.
567 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
568 will now be deprecated.
570 PH/56 New os.c-cygwin from the Cygwin maintainer.
572 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
573 formats (previously only available in exim formats via -bpr, -bpru,
574 and -bpra. Now also available in native and exiqgrep formats)
576 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
577 with very large, slow to parse queues
579 JJ/08 exipick: added ! as generic prefix to negate any criteria format
581 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
583 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
584 responses to authentication challenges, though it was showing the
585 challenges; (ii) I've removed the CR characters from the debug output for
588 PH/58 Allow for the insertion of a newline as well as a space when a string
589 is turned into more than one encoded-word during RFC 2047 encoding. The
590 Sieve code now uses this.
592 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
593 data_4xx, lost_connection, tls_required.
595 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
596 message was being sent as an SMTP response.
598 PH/61 Add -l and -k options to exicyclog.
600 PH/62 When verifying, if an address was redirected to one new address, so that
601 verification continued, and the new address failed or deferred after
602 having set something in $address_data, the value of $address_data was not
603 passed back to the ACL. This was different to the case when no
604 redirection occurred. The value is now passed back in both cases.
606 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
607 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
608 use login_cap.h, so on its own it isn't the distinguishing feature. The
609 new name refers directly to the setclassresources() function.
611 PH/65 Added configuration files for NetBSD3.
613 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
615 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
616 is preferred over IPv4.
618 PH/68 The bounce_return_message and bounce_return_body options were not being
619 honoured for bounces generated during the reception of non-SMTP messages.
620 In particular, this applied to messages rejected by the ACL. This bug has
621 been fixed. However, if bounce_return_message is true and bounce_return_
622 body is false, the headers that are returned for a non-SMTP message
623 include only those that have been read before the error was detected.
624 (In the case of an ACL rejection, they have all been read.)
626 PH/69 The HTML version of the specification is now built in a directory called
627 spec_html instead of spec.html, because the latter looks like a path with
628 a MIME-type, and this confuses some software.
630 PH/70 Catch two compiler warnings in sieve.c.
632 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
633 function verify_get_ident() calls ip_connect() to connect a socket, but
634 if the "connect()" function timed out, ip_connect() used to close the
635 socket. However, verify_get_ident() also closes the socket later, and in
636 between Exim writes to the log, which may get opened at this point. When
637 the socket was closed in ip_connect(), the log could get the same file
638 descriptor number as the socket. This naturally causes chaos. The fix is
639 not to close the socket in ip_connect(); the socket should be closed by
640 the function that creates it. There was only one place in the code where
641 this was missing, in the iplookup router, which I don't think anybody now
642 uses, but I've fixed it anyway.
644 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
645 well as to direct DNS lookups. Otherwise the handling of names in host
646 lists is inconsistent and therefore confusing.
652 PH/01 Two changes to the default runtime configuration:
654 (1) Move the checks for relay_from_hosts and authenticated clients from
655 after to before the (commented out) DNS black list checks.
657 (2) Add control=submission to the relay_from_hosts and authenticated
658 clients checks, on the grounds that messages accepted by these
659 statements are most likely to be submissions.
661 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
663 (1) Generate an error if the third argument for the ${prvs expansion is
666 (2) Treat a missing third argument of ${prvscheck as if it were an empty
669 (3) Reset the variables that are obtained from the first argument of
670 ${prvscheck and used in the second argument before leaving the code,
671 because their memory is reclaimed, so using them afterwards may do
674 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
675 one (it's much easier than Tom thought :-).
677 (5) Because of (4), we can now allow for the use of $prvscheck_result
678 inside the third argument.
680 PH/03 For some reason, the default setting of PATH when running a command from
681 a pipe transport was just "/usr/bin". I have changed it to
684 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
685 anything to be listed in the output from -bV.
687 PH/05 When a filter generated an autoreply, the entire To: header line was
688 quoted in the delivery log line, like this:
690 => >A.N.Other <ano@some.domain> <original@ddress> ...
692 This has been changed so that it extracts the operative address. There
693 may be more than one such address. If so, they are comma-separated, like
696 => >ano@some.domain,ona@other.domain <original@ddress> ...
698 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
699 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
700 not being looked up in the rDNS to get a host name, Exim was showing the
701 IP address twice in Received: lines, even though the IP addresses were
702 identical. For example:
704 Received: from [1.2.3.4] (helo=[1.2.3.4])
706 However, if the real host name was known, it was omitting the HELO data
707 if it matched the actual IP address. This has been tidied up so that it
708 doesn't show the same IP address twice.
710 PH/07 When both +timestamp and +memory debugging was on, the value given by
711 $tod_xxx expansions could be wrong, because the tod_stamp() function was
712 called by the debug printing, thereby overwriting the timestamp buffer.
713 Debugging no longer uses the tod_stamp() function when +timestamp is set.
715 PH/08 When the original message was included in an autoreply transport, it
716 always said "this is a copy of the message, including all the headers",
717 even if body_only or headers_only was set. It now gives an appropriate
720 PH/09 Applied a patch from the Sieve maintainer which:
722 o fixes some comments
723 o adds the (disabled) notify extension core
724 o adds some debug output for the result of if/elsif tests
725 o points to the current vacation draft in the documentation
726 and documents the missing references header update
730 o fixes a bug in processing the envelope test (when testing
731 multiple envelope elements, the last element determinted the
734 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
735 Electronic Mail") by including:
737 Auto-submitted: auto-generated
739 in the messages that it generates (bounce messages and others, such as
740 warnings). In the case of bounce messages for non-SMTP mesages, there was
741 also a typo: it was using "Auto_submitted" (underscore instead of
742 hyphen). Since every message generated by Exim is necessarily in response
743 to another message, thes have all been changed to:
745 Auto-Submitted: auto-replied
747 in accordance with these statements in the RFC:
749 The auto-replied keyword:
751 - SHOULD be used on messages sent in direct response to another
752 message by an automatic process,
754 - MUST NOT be used on manually-generated messages,
756 - MAY be used on Delivery Status Notifications (DSNs) and Message
757 Disposition Notifications (MDNs),
759 - MUST NOT be used on messages generated by automatic or periodic
760 processes, except for messages which are automatic responses to
763 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
764 to the default Received: header definition.
766 PH/12 Added log selector acl_warn_skipped (default on).
768 PH/13 After a successful wildlsearch lookup, discard the values of numeric
769 variables because (a) they are in the wrong storage pool and (b) even if
770 they were copied, it wouldn't work properly because of the caching.
772 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
773 checking when decoding. Apparently there are clients that generate
774 overlong encoded strings. Why am I not surprised?
776 PH/15 If the first argument of "${if match_address" was not empty, but did not
777 contain an "@" character, Exim crashed. Now it writes a panic log message
778 and treats the condition as false.
780 PH/16 In autoreply, treat an empty string for "once" the same as unset.
782 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
783 extension "envelope-auth". The code is finished and in agreement with
784 other implementations, but there is no documentation so far and in fact,
785 nobody wrote the draft yet. This extension is currently #undef'ed, thus
786 not changing the active code.
788 Print executed "if" and "elsif" statements when debugging is used. This
789 helps a great deal to understand what a filter does.
791 Document more things not specified clearly in RFC3028. I had all this
792 sorted out, when out of a sudden new issues came to my mind. Oops."
794 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
797 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
798 canonical form (as documented). However, after a host name lookup from
799 the IP address, check_host() was doing a simple string comparison with
800 addresses acquired from the DNS when checking that the found name did
801 have the original IP as one of its addresses. Since any found IPv6
802 addresses are likely to be in abbreviated form, the comparison could
803 fail. Luckily, there already exists a function for doing the comparison
804 by converting both addresses to binary, so now that is used instead of
807 PH/20 There was another similar case to PH/19, when a complete host name was
808 given in a host list; looking up its IP address could give an abbreviated
809 form, whereas the current host's name might or might not be abbreviated.
810 The same fix has been applied.
816 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
817 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
820 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
821 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
823 PH/03 Typo: missing ".o" in src/pcre/Makefile.
825 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
826 header line, restrict the check to what is listed in RFCs 2369 and 2929.
827 Also, for "Auto-Submitted", treat anything other than "no" as
828 non-personal, in accordance with RFC 3834. (Previously it treated
829 anything starting "auto-" as non-personal.)
831 TF/01 The control=submission/name=... option had a problem with syntax
832 errors if the name included a slash character. The /name= option
833 now slurps the rest of the string, so it can include any characters
834 but it must come last in the list of options (after /sender_retain
837 PH/05 Some modifications to the interface to the fake nameserver for the new
845 TK/01 Added the "success_on_redirect" address verification option. See
846 NewStuff for rationale and an example.
848 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
850 PH/02 Patch to exigrep to allow it to work on syslog lines.
852 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
853 fread() to skip over the body file's header line, because in Cygwin the
854 header line is locked and is inaccessible.
856 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
857 co-exist for some time) to make it clear that it is the Exim ID that is
858 referenced, not the Message-ID: header line.
860 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
861 string_format() function, because snprintf() does not exist on all
864 PH/06 The use of forbid_filter_existstest now also locks out the use of the
865 ${stat: expansion item.
867 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
868 protocol synchronization error", to keep the pedants happy.
870 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
871 well as for IRIX systems, when gcc is being used. See the host.c source
874 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
876 PH/10 Named domain lists were not working if used in a queue_smtp_domains
879 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
880 transport and to the smtp transport in LMTP mode.
882 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
884 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
885 run a filter in a subprocess. This could lead to confusion in subsequent
886 lookups in the parent process. There should also be a search_tidyup() at
887 the end of the subprocess.
889 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
890 only if the host matched helo_try_verify_hosts, which caused the
891 verification to occur when the EHLO/HELO command was issued. The ACL just
892 tested the remembered result. Now, if a previous verification attempt has
893 not happened, "verify = helo" does it there and then.
895 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
897 TK/03 Fix log output including CR from clamd.
899 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
900 exist provoked a memory error which could cause a segfault.
902 PH/15 Installed PCRE 6.2
904 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
906 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
907 of the problem. Specifically, suggested +O2 rather than +O1 for the
910 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
912 PH/20 If a delivery was routed to a non-standard port by means of an SRV
913 record, the port was not correctly logged when the outgoing_port log
914 selector was set (it logged the transort's default port).
916 PH/21 Added support for host-specific ports to manualroute, queryprogram,
917 fallback_hosts, and "hosts" in the smtp transport.
919 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
920 host errors such as "Connection refused".
922 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
923 authentication with radiusclient 0.4.9:
925 - Error returned from rc_read_config was caught wrongly
926 - Username/password not passed on to radius server due to wrong length.
928 The presumption is that some radiusclient API changes for 4.51/PH/17
929 were not taken care of correctly. The code is still untested by me (my
930 Linux distribution still has 0.3.2 of radiusclient), but it was
931 contributed by a Radius user.
933 PH/24 When doing a callout, the value of $domain wasn't set correctly when
934 expanding the "port" option of the smtp transport.
936 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
937 while reading a MIME header. Thanks to Tom Hughes for a patch.
939 PH/24 Include config.h inside local_scan.h so that configuration settings are
942 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
943 that in an ACL for RCPT (for example), you can examine exactly what was
946 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
947 commands, but it was not correctly comparing the address with the actual
948 client host address. Thus, it would show the EHLO address in Received:
949 header lines when this was not necessary.
951 PH/27 Added the % operator to ${eval:}.
953 PH/28 Exim tries to create and chdir to its spool directory when it starts;
954 it should be ignoring failures (because with -C, for example, it has lost
955 privilege). It wasn't ignoring creation failures other than "already
958 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
961 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
962 because an input line was too long, either on its own, or by virtue of
963 too many continuations, the temporary file was not being removed, and the
964 return code was incorrect.
966 PH/31 Missing "BOOL" in function definition in filtertest.c.
968 PH/32 Applied Sieve patches from the maintainer.
970 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
972 PH/33 Added "verify = not_blind".
974 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
975 Local/Makefile (with some defaults set). These are used in built scripts
976 such as exicyclog, but they have never been used in the exim_install
977 script (though there are many overriding facilities there). I have
978 arranged that the exim_install script now takes note of these two
981 PH/35 Installed configuration files for Dragonfly.
983 PH/36 When a locally submitted message by a trusted user did not contain a
984 From: header, and the sender address was obtained from -f or from an SMTP
985 MAIL command, and the trusted user did not use -F to supply a sender
986 name, $originator_name was incorrectly used when constructing a From:
987 header. Furthermore, $originator_name was used for submission mode
988 messages from external hosts without From: headers in a similar way,
989 which is clearly wrong.
991 PH/37 Added control=suppress_local_fixups.
993 PH/38 When log_selector = +received_sender was set, and the addition of the
994 sender made the log line's construction buffer exactly full, or one byte
995 less than full, an overflow happened when the terminating "\n" was
998 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
999 when the result of a list match is failure because a DNS lookup failed.
1001 PH/40 RM_COMMAND is now used in the building process.
1003 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1004 the "build-* directories that it finds.
1006 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1007 address in a domain literal was a prefix of an interface address.
1009 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1010 when verifying a sender address, unless rewrite_headers is false.
1012 PH/44 Wrote a long comment about why errors_to addresses are verified as
1013 recipients, not senders.
1015 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1016 the ratelimit ACL was added.
1018 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1020 PH/46 Added extra information about PostgreSQL errors to the error string.
1022 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1023 avoiding the need to install special zones in a real server. This is
1024 backwards compatible; if it can't find the fake resolver, it drops back.
1025 Thus, both old and new test suites can be run.
1027 TF/02 Added util/ratelimit.pl
1029 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1030 clock is set back in time.
1032 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1033 Candler <B.Candler@pobox.com>.
1035 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1036 OK for addresses that are the result of redirections.
1038 PH/48 A number of further additions for the benefit of the new test suite,
1039 including a fake gethostbyname() that interfaces to the fake DNS resolver
1042 TF/06 The fix for widen_domains has also been applied to qualify_single and
1043 search_parents which are the other dnslookup options that can cause
1046 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1049 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1051 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1052 service" if it encountered an MX record with an empty target hostname.
1053 The message is now "an MX or SRV record indicated no SMTP service".
1055 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1056 if the DNS of the sending site is misconfigured. This is quite a
1057 common situation. This change restores the behaviour of treating a
1058 helo verification defer as a failure.
1060 PH/51 If self=fail was set on a router, the bounce message did not include the
1061 actual error message.
1067 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1069 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1070 command itself ended in error, the underlying message about the transport
1071 filter timeout was being overwritten with the pipe command error. Now the
1072 underlying error message should be appended to the second error message.
1074 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1076 PH/02 Reduce the amount of output that "make" produces by default. Full output
1077 can still be requested.
1079 PH/03 The warning log line about a condition test deferring for a "warn" verb
1080 was being output only once per connection, rather than after each
1081 occurrence (because it was using the same function as for successful
1082 "warn" verbs). This seems wrong, so I have changed it.
1084 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1085 it should not have, which might have caused a crash in the right
1086 circumstances, but probably never did.
1088 PH/04 Installed a modified version of Tony Finch's patch to make submission
1089 mode fix the return path as well as the Sender: header line, and to
1090 add a /name= option so that you can make the user's friendly name appear
1093 TF/03 Added the control = fakedefer ACL modifier.
1095 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1096 Mark Lowes for thorough testing.
1098 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1100 TK/03 Merged latest SRS patch from Miles Wilton.
1102 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1103 with the definition in sysexits.h (which is #included earlier).
1104 Fortunately, Exim does not actually use EX_OK. The code used to try to
1105 preserve the sysexits.h value, by assumimg that macro definitions were
1106 scanned for macro replacements. I have been disabused of this notion,
1107 so now the code just undefines EX_OK before #including unistd.h.
1109 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1110 in the smtp transport. When a block could not be written in a single
1111 write() function, the timeout was being re-applied to each part-write.
1112 This seems wrong - if the receiver was accepting one byte at a time it
1113 would take for ever. The timeout is now adjusted when this happens. It
1114 doesn't have to be particularly precise.
1116 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1117 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1119 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1120 without a domain if the check to <postmaster@domain> fails.
1122 SC/01 Eximstats: added -xls and the ability to specify output files
1123 (patch written by Frank Heydlauf).
1125 SC/02 Eximstats: use FileHandles for outputing results.
1127 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1129 SC/04 Eximstats: fixed display of large numbers with -nvr option
1131 SC/05 Eximstats: fixed merging of reports with empty tables.
1133 SC/06 Eximstats: added the -include_original_destination flag
1135 SC/07 Eximstats: removed tabs and trailing whitespace.
1137 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1139 TK/06 MBOX spool code: Add real "From " MBOX separator line
1140 so the .eml file is really in mbox format (even though
1141 most programs do not really care). Patch from Alex Miller.
1143 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1144 The latter is generated from $received_to and is only set if the
1145 message has one envelope recipient. SA can use these headers,
1146 obviously out-of-the-box. Patch from Alex Miller.
1148 PH/08 The ${def test on a variable was returning false if the variable's
1149 value was "0", contrary to what the specification has always said!
1150 The result should be true unless the variable is empty.
1152 PH/09 The syntax error of a character other than { following "${if
1153 def:variable_name" (after optional whitespace) was not being diagnosed.
1154 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1155 accidental colon was present, for example, could give incorrect results.
1157 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1158 result is used (not including appendfile, where other changes are about
1161 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1162 This involved changing a lot of size variables from int to off_t. It
1163 should work with maildirs and everything.
1165 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1166 spamd dying while we are connected to it.
1168 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1171 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1172 function for BDB 4.3.
1174 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1176 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1177 the sources for PCRE have been re-arranged and more files are now
1180 PH/15 The code I had for printing potentially long long variables in PH/11
1181 above was not the best (it lost precision). The length of off_t variables
1182 is now inspected at build time, and an appropriate printing format (%ld
1183 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1184 to be "long long int" or "long int". This is needed for the internal
1185 formatting function string_vformat().
1187 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1188 the configuration file to be ":syslog", then the script "guesses" where
1189 the logs files are, rather than using the compiled in default. In our
1190 case the guess is not the same as the compiled default, so the script
1191 suddenly stopped working when I started to use syslog. The patch checks
1192 to see if log_file_path is "". If so, it attempts to read it from exim
1193 with no configuration file to get the compiled in version, before it
1194 falls back to the previous guessing code."
1196 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1197 implementing BATV in an Exim configuration. See NewStuff for the gory
1200 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1201 Makefile that are specific to HP-UX.
1203 PH/18 If the "use_postmaster" option was set for a recipient callout together
1204 with the "random" option, the postmaster address was used as the MAIL
1205 FROM address for the random test, but not for the subsequent recipient
1206 test. It is now used for both.
1208 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1209 patch removes a few documentation additions to RFC 3028, because the
1210 latest draft now contains them. It adds the new en;ascii-case comparator
1211 and a new error check for 8bit text in MIME parts. Comparator and
1212 require names are now matched exactly. I enabled the subaddress
1213 extension, but it is not well tested yet (read: it works for me)."
1215 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1216 rework some of the code of TK/09 above to avoid the hardwired use of
1217 "%lld" and "long long". Replaced the call to snprintf() with a call to
1220 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1221 records point to non-existent hosts", "retry timeout exceeded", and
1222 "retry time not reached for any host after a long failure period".
1224 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1225 experimental DomainKeys support:
1227 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1228 (2) On an error such as an illegally used "control", the wrong name for
1229 the control was given.
1231 These problems did NOT occur unless DomainKeys support was compiled.
1233 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1235 PH/24 Added ${if match_ip condition.
1237 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1238 databases so that it will be absolutely obvious if a crash occurs in the
1239 DB library. This is a regular occurrence (often caused by mis-matched
1242 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1243 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1244 picked up on a user's system that detects such things. There doesn't seem
1245 to be a gcc warning option for this - only an attribute that has to be
1246 put on the function's prototype. It seems that in Fedora Core 4 they have
1247 set this on a number of new functions. No doubt there will be more in due
1250 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1251 specify a transport. However, if an address that was verified by such a
1252 router was the subject of a callout, Exim crashed because it tried to
1253 read the rcpt_include_affixes from the non-existent transport. Now it
1254 just assumes that the setting of that option is false. This bug was
1255 introduced by 4.51/PH/31.
1257 PH/28 Changed -d+all to exclude +memory, because that information is very
1258 rarely of interest, but it makes the output a lot bigger. People tend to
1259 do -d+all out of habit.
1261 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1262 code in os-type was giving problems when libc.so lives in lib64, like on
1265 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1266 aren't the modern standard, and it seems that some systems' include files
1267 don't always have them. Exim was already checking for some of the newer
1268 ones like T_AAAA, and defining it itself. I've added checks for all the
1269 record types that Exim uses.
1271 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1272 not automatically generating a new one, as it is supposed to. This
1273 prevented TLS from working. If the file did exist, but contained invalid
1274 data, a new version was generated, as expected. It was only the case of a
1275 non-existent file that was broken.
1277 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1278 with a change in libdomainkeys > 0.64.
1280 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1281 from DNS. If the selector record carries the flag, it now has
1282 precedence over the domain-wide flag.
1284 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1286 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1287 the use of a port name that isn't defined in /etc/services) occurred, the
1288 message was deferred as in a normal delivery, and thus remained on the
1289 spool, instead of being failed because of the mua_wrapper setting. This
1290 is now fixed, and I tidied up some of the mua_wrapper messages at the
1293 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1294 the messages in a hash of arrays rather than using individual hashes.
1295 This is a bit cleaner and results in dramatic memory savings, albeit
1296 at a slight CPU cost.
1298 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1299 as requested by Marc Sherman.
1301 SC/10 Eximstats: added histograms for user specified patterns as requested
1304 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1306 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1307 fopen() in the content-scanning modules that did not already have it.
1313 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1314 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1316 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1318 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1320 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1321 to test Sieve filters that use "vacation".
1323 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1324 that changes the way the GnuTLS parameters are stored in the cache file.
1325 The new format can be generated externally. For backward compatibility,
1326 if the data in the cache doesn't make sense, Exim assumes it has read an
1327 old-format file, and it generates new data and writes a new file. This
1328 means that you can't go back to an older release without removing the
1331 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1332 work if there are any delivery delays because "one_time" forces the
1333 parent to be marked "delivered", so its unseen clone is never tried
1334 again. For this reason, Exim now forbids the simultaneous setting of
1337 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1338 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1339 entry if you want to know the details. The fix, however, neglected to
1340 consider the case when local delivery batching is involved. The test for
1341 "previously delivered" was not happening when checking to see if an
1342 address could be batched with a previous (undelivered) one; under
1343 certain circumstances this could lead to multiple deliveries to the same
1346 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1347 in its include files, and this causes problems building Exim.
1349 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1350 header_syntax) but Exim was just ignoring anything given after a slash.
1351 In particular, this caused confusion with an attempt to use "verify =
1352 reverse_host_lookup/defer_ok". An error is now given when options are
1353 supplied for verify items that do not have them. (Maybe reverse_host_
1354 lookup should have a defer_ok option, but that's a different point.)
1356 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1357 defined by RFC 821) to 2048, because there were problems with some AUTH
1358 commands, and RFC 1869 says the size should be increased for extended
1359 SMTP commands that take arguments.
1361 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1364 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1365 "unknown" error; now it says that the functionality isn't in the binary.
1367 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1368 an address' error message when a string expansion fails (syntax or
1369 whatever). Otherwise the password may appear in the log. Following change
1370 PH/42 below, there is no longer a chance of it appearing in a bounce
1373 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1375 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1376 of its addresses was ever tried. (Bugzilla bug #2.)
1378 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1379 the result incorrectly in the debug output. (It correctly added a newline
1380 to what was transported.)
1382 TF/01 Added $received_time.
1384 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1385 commented out examples of how to interface to a virus scanner and to
1386 SpamAssassin. Also added commented examples of av_scanner and
1387 spamd_address settings.
1389 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1390 and controls are allowed in which ACLs. There were a couple of minor
1391 errors. Some of the entries in the conditions table (which is a table of
1392 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1393 negation of where the condition IS allowed.
1395 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1397 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1398 header file does not have a version number, so I've had to invent a new
1399 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1400 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1401 radiusclient), but it was contributed by a Radius user.
1403 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1404 files or directories, for OpenSSL.
1406 PH/19 When an Exim process that is running as root has to create an Exim log
1407 file, it does so in a subprocess that runs as exim:exim so as to get the
1408 ownership right at creation (otherwise, other Exim processes might see
1409 the file with the wrong ownership). There was no test for failure of this
1410 fork() call, which would lead to the process getting stuck as it waited
1411 for a non-existent subprocess. Forks do occasionally fail when resources
1412 run out. I reviewed all the other calls to fork(); they all seem to check
1415 PH/20 When checking for unexpected SMTP input at connect time (before writing
1416 the banner), Exim was not dealing correctly with a non-positive return
1417 from the read() function. If the client had disconnected by this time,
1418 the result was a log entry for a synchronization error with an empty
1419 string after "input=" when read() returned zero. If read() returned -1
1420 (an event I could not check), uninitialized data bytes were printed.
1421 There were reports of junk text (parts of files, etc) appearing after
1424 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1426 PH/22 Added support for macro redefinition, and (re)definition in between
1427 driver and ACL definitions.
1429 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1430 forgetting to use the resulting value; it was using the unexpanded value.
1432 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1433 hadn't been configured. The fix is from Juergen Kreileder, who
1434 understands it better than I do:
1436 "Here's what I see happening with three configured cyrus_sasl
1437 authenticators configured (plain, login, cram-md5):
1439 On startup auth_cyrus_sasl_init() gets called for each of these.
1440 This means three calls to sasl_listmech() without a specified mech_list.
1441 => SASL tests which mechs of all available mechs actually work
1442 => three warnings about OTP not working
1443 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1445 With the patch, sasl_listmech() also gets called three times. But now
1446 SASL's mech_list option is set to the server_mech specified in the the
1447 authenticator. Or in other words, the answer from sasl_listmech()
1448 gets limited to just the mech you're testing for (which is different
1450 => the return list contains just 'plain' or 'login', 'cram-md5' or
1451 nothing depending on the value of ob->server_mech.
1453 I've just tested the patch: Authentication still works fine,
1454 unavailable mechs specified in the exim configuration are still
1455 caught, and the auth.log warnings about OTP are gone."
1457 PH/25 When debugging is enabled, the contents of the command line are added
1458 to the debugging output, even when log_selector=+arguments is not
1461 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1462 answer is "GNU", and only if the return is "GNU/something" is the answer
1465 PH/27 $acl_verify_message is now set immediately after the failure of a
1466 verification in an ACL, and so is available in subsequent modifiers. In
1467 particular, the message can be preserved by coding like this:
1469 warn !verify = sender
1470 set acl_m0 = $acl_verify_message
1472 Previously, $acl_verify_message was set only while expanding "message"
1473 and "log_message" when a very denied access.
1475 PH/28 Modified OS/os.c-Linux with
1477 -#ifndef OS_LOAD_AVERAGE
1478 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1480 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1481 nomenclature these days.)
1483 PH/29 Installed patch from the Sieve maintainer that adds the options
1484 sieve_useraddress and sieve_subaddress to the redirect router.
1486 PH/30 In these circumstances:
1487 . Two addresses routed to the same list of hosts;
1488 . First host does not offer TLS;
1489 . First host accepts first address;
1490 . First host gives temporary error to second address;
1491 . Second host offers TLS and a TLS session is established;
1492 . Second host accepts second address.
1493 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1494 and peerdn, if requested) that were in fact used only for the second
1497 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1498 attention to any local part prefix or suffix that was matched by the
1499 router that accepted the address. It now behaves in the same way as it
1500 does for delivery: the affixes are removed from the local part unless
1501 rcpt_include_affixes is set on the transport.
1503 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1504 timeout during the DATA phase of an incoming message.
1506 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1507 applied a patch sanctioned by the Sieve maintainer.
1509 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1510 the uid or gid is negative. A case of a negative gid caused this to be
1511 noticed. The fix allows for either to be negative.
1513 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1514 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1517 PH/36 The change PH/12 above was broken. Fixed it.
1519 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1520 the grounds that routing the same address twice would always produce the
1521 same answer. This might have been true once, but it is certainly no
1522 longer true now. Routing a child address may depend on the previous
1523 routing that produced that child. Some complicated redirection strategies
1524 went wrong when messages had multiple recipients, and made Exim's
1525 behaviour dependent on the order in which the addresses were given.
1527 I have moved the duplicate checking until after the routing is complete.
1528 Exim scans the addresses that are assigned to local and remote
1529 transports, and removes any duplicates. This means that more work will be
1530 done, as duplicates will always all be routed, but duplicates are
1531 presumably rare, so I don't expect this is of any significance.
1533 For deliveries to pipes, files, and autoreplies, the duplicate checking
1534 still happens during the routing process, since they are not going to be
1537 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1538 It corrects a timeout issue with spamd. This is Ian's comment: "The
1539 background is that sometimes spamd either never reads data from a
1540 connection it has accepted, or it never writes response data. The exiscan
1541 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1542 blindly assumes that writes won't block so it may never time out."
1544 PH/39 Allow G after quota size as well as K and M.
1546 PH/40 The value set for $authenticated_id in an authenticator may not contain
1547 binary zeroes or newlines because the value is written to log lines and
1548 to spool files. There was no check on this. Now the value is run through
1549 the string_printing() function so that such characters are converted to
1550 printable escape sequences.
1552 PH/41 $message_linecount is a new variable that contains the total number of
1553 lines in the message. Compare $body_linecount, which is the count for the
1556 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1557 bounce and delay warning messages, except in certain special cases, which
1560 (a) An SMTP error message from a remote host;
1561 (b) A message specified in a :fail: redirection;
1562 (c) A message specified in a "fail" command in a system filter;
1563 (d) A message specified in a FAIL return from the queryprogram router;
1564 (e) A message specified by the cannot_route_message router option.
1566 In these cases only, Exim does include the error details in bounce and
1567 warning messages. There are also a few cases where bland messages such
1568 as "unrouteable address" or "local delivery error" are given.
1570 PH/43 $value is now also set for the "else" part of a ${run expansion.
1572 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1573 being worked on, but at least Exim now implements the latest version to
1576 PH/45 In a pipe transport, although a timeout while waiting for the pipe
1577 process to complete was treated as a delivery failure, a timeout while
1578 writing the message to the pipe was logged, but erroneously treated as a
1579 successful delivery. Such timeouts include transport filter timeouts. For
1580 consistency with the overall process timeout, these timeouts are now
1581 treated as errors, giving rise to delivery failures by default. However,
1582 there is now a new Boolean option for the pipe transport called
1583 timeout_defer, which, if set TRUE, converts the failures into defers for
1584 both kinds of timeout. A transport filter timeout is now identified in
1587 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1588 systems where "make" and "gmake" are different, calling "gmake" at top
1589 level broke things. I've arranged for the value of $(MAKE) to be passed
1590 from the Makefile to this script so that it can call the same version of
1594 A note about Exim versions 4.44 and 4.50
1595 ----------------------------------------
1597 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1598 changes of various kinds. As a consequence, a big documentation update was
1599 needed. This delayed the release for rather longer than seemed good, especially
1600 in the light of a couple of (minor) security issues. Therefore, the changes
1601 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1602 release. So 4.44 and 4.50 are in effect two different branches that both start
1605 I have left the 4.50 change log unchanged; it contains all the changes since
1606 4.43. The change log for 4.44 is below; many of its items are identical to
1607 those for 4.50. This seems to be the most sensible way to preserve the
1608 historical information.
1614 1. Minor wording change to the doc/README.SIEVE file.
1616 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1617 computation of the current number of files was incorrect.
1619 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1620 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1623 4. Give more explanation in the error message when the command for a transport
1624 filter fails to execute.
1626 5. There are several places where Exim runs a non-Exim command in a
1627 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1628 was being done only for the command run by the queryprogram router. It is
1629 now done for all such subprocesses. The other cases are: ${run, transport
1630 filters, and the commands run by the lmtp and pipe transports.
1632 6. Added CONFIGURE_GROUP build-time option.
1634 7. Some older OS have a limit of 256 on the maximum number of file
1635 descriptors. Exim was using setrlimit() to set 1000 as a large value
1636 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1637 systems. I've change it so that if it can't get 1000, it tries for 256.
1639 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1640 was an oversight, and furthermore, ever since the addition of extra
1641 controls (e.g. 4.43/32), the checks on when to allow different forms of
1642 "control" were broken. There should now be diagnostics for all cases when a
1643 control that does not make sense is encountered.
1645 9. Added the /retain_sender option to "control=submission".
1647 10. $recipients is now available in the predata ACL (oversight).
1649 11. Tidy the search cache before the fork to do a delivery from a message
1650 received from the command line. Otherwise the child will trigger a lookup
1651 failure and thereby defer the delivery if it tries to use (for example) a
1652 cached ldap connection that the parent has called unbind on.
1654 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1655 of $address_data from the recipient verification was clobbered by the
1656 sender verification.
1658 13. The value of address_data from a sender verification is now available in
1659 $sender_address_data in subsequent conditions in the ACL statement.
1661 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1663 15. Added a new option "connect=<time>" to callout options, to set a different
1666 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1667 was its contents. (It was OK if the option was not defined at all.)
1669 17. A "Completed" log line is now written for messages that are removed from
1670 the spool by the -Mrm option.
1672 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1673 information about exactly what failed.
1675 19. Added -dd to debug only the daemon process.
1677 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1678 handles timeouts, both on the server side and network timeouts. Renamed the
1679 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1681 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1682 It is now set to "smtps".
1684 22. $host_address is now set to the target address during the checking of
1685 ignore_target_hosts.
1687 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1688 being passed; this would have caused $sender_host_name to have been used if
1689 matching the list had actually called for a host name (not very likely,
1690 since this list is usually IP addresses). A host name is now passed as
1693 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1694 code that specifies a non-restarting handler (typically sigaction() in
1695 modern systems) in an attempt to fix a rare and obscure crash bug.
1697 25. Narrowed the window for a race in the daemon that could cause it to ignore
1698 SIGCHLD signals. This is not a major problem, because they are used only to
1699 wake it up if nothing else does.
1701 26. A malformed maildirsize file could cause Exim to calculate negative values
1702 for the mailbox size or file count. Odd effects could occur as a result.
1703 The maildirsize information is now recalculated if the size or filecount
1706 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1707 support for a long time. Removed HAVE_SYS_VFS_H.
1709 28. Installed the latest version of exipick from John Jetmore.
1711 29. In an address list, if the pattern was not a regular expression, an empty
1712 subject address (from a bounce message) matched only if the pattern was an
1713 empty string. Non-empty patterns were not even tested. This was the wrong
1714 because it is perfectly reasonable to use an empty address as part of a
1715 database query. An empty address is now tested by patterns that are
1716 lookups. However, all the other forms of pattern expect the subject to
1717 contain a local part and a domain, and therefore, for them, an empty
1718 address still always fails if the pattern is not itself empty.
1720 30. Exim went into a mad DNS loop when attempting to do a callout where the
1721 host was specified on an smtp transport, and looking it up yielded more
1722 than one IP address.
1724 31. Re-factored the code for checking spool and log partition space into a
1725 function that finds that data and another that does the check. The former
1726 is then used to implement four new variables: $spool_space, $log_space,
1727 $spool_inodes, and $log_inodes.
1729 32. The RFC2047 encoding function was originally intended for short strings
1730 such as real names; it was not keeping to the 75-character limit for
1731 encoded words that the RFC imposes. It now respects the limit, and
1732 generates multiple encoded words if necessary. To be on the safe side, I
1733 have increased the buffer size for the ${rfc2047: expansion operator from
1736 33. It is now permitted to omit both strings after an "if" condition; if the
1737 condition is true, the result is "true". As before, when the second string
1738 is omitted, a false condition yields an empty string. This makes it less
1739 cumbersome to write custom ACL and router conditions.
1741 34. Failure to deliver a bounce message always caused it to be frozen, even if
1742 there was an errors_to setting on the router. The errors_to setting is now
1745 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1746 canonical form (fully expanded) before being placed in
1747 $sender_host_address.
1749 36. The table in the code that translates DNS record types into text (T_A to
1750 "A" for instance) was missing entries for NS and CNAME. It is just possible
1751 that this could have caused confusion if both these types were looked up
1752 for the same domain, because the text type is used as part of Exim's
1753 per-process caching. But the chance of anyone hitting this buglet seems
1756 37. The dnsdb lookup has been extended in a number of ways.
1758 (1) There is a new type, "zns", which walks up the domain tree until it
1759 finds some nameserver records. It should be used with care.
1761 (2) There is a new type, "mxh", which is like "mx" except that it returns
1762 just the host names, not the priorities.
1764 (3) It is now possible to give a list of domains (or IP addresses) to be
1765 looked up. The behaviour when one of the lookups defers can be
1766 controlled by a keyword.
1768 (4) It is now possible to specify the separator character for use when
1769 multiple records are returned.
1771 38. The dnslists ACL condition has been extended: it is now possible to supply
1772 a list of IP addresses and/or domains to be looked up in a particular DNS
1775 39. Added log_selector=+queue_time_overall.
1777 40. When running the queue in the test harness, wait just a tad after forking a
1778 delivery process, to get repeatability of debugging output.
1780 41. Include certificate and key file names in error message when GnuTLS fails
1781 to set them up, because the GnuTLS error message doesn't include the name
1782 of the failing file when there is a problem reading it.
1784 42. Allow both -bf and -bF in the same test run.
1786 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1788 44. The "Exiscan patch" is now merged into the mainline Exim source.
1790 45. Sometimes the final signoff response after QUIT could fail to get
1791 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1792 < 0 before doing a fflush(). This bug looks as though it goes back to the
1793 introduction of TLS in release 3.20, but "sometimes" must have been rare
1794 because the tests only now provoked it.
1796 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1797 (this can affect the format of dates).
1799 47. exim_tidydb, when checking for the continued existence of a message for
1800 which it has found a message-specific retry record, was not finding
1801 messages that were in split spool directories. Consequently, it was
1802 deleting retry records that should have stayed in existence.
1804 48. Steve fixed some bugs in eximstats.
1806 49. The SPA authentication driver was not abandoning authentication and moving
1807 on to the next authenticator when an expansion was forced to fail,
1808 contradicting the general specification for all authenticators. Instead it
1809 was generating a temporary error. It now behaves as specified.
1811 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1812 (the order specifies the preference for clients). The order is now AES256,
1813 AES128, 3DES, ARCFOUR128.
1815 51. Small patch to Sieve code - explicitly set From: when generating an
1818 52. Exim crashed if a remote delivery caused a very long error message to be
1819 recorded - for instance if somebody sent an entire SpamAssassin report back
1820 as a large number of 550 error lines. This bug was coincidentally fixed by
1821 increasing the size of one of Exim's internal buffers (big_buffer) that
1822 happened as part of the Exiscan merge. However, to be on the safe side, I
1823 have made the code more robust (and fixed the comments that describe what
1826 53. Now that there can be additional text after "Completed" in log lines (if
1827 the queue_time_overall log selector is set), a one-byte patch to exigrep
1828 was needed to allow it to recognize "Completed" as not the last thing in
1831 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1832 patch that reportedly fixes this has been added. I am not expert enough to
1833 create a test for it. This is what the patch creator wrote:
1835 "I found a little strange behaviour of ldap code when working with
1836 Windows 2003 AD Domain, where users was placed in more than one
1837 Organization Units. When I tried to give exim partial DN, the exit code
1838 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1839 But simultaneously result of request was absolutely normal ldap result,
1840 so I produce this patch..."
1842 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1843 REFERENCE, so I have modified the code to exclude the patch when that macro
1846 55. Some experimental protocols are using DNS PTR records for new purposes. The
1847 keys for these records are domain names, not reversed IP addresses. The
1848 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1849 leaves it alone. Component reversal etc. now happens only for IP addresses.
1852 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1854 57. Double the size of the debug message buffer (to 2048) so that more of very
1855 long debug lines gets shown.
1857 58. The exicyclog utility now does better if the number of log files to keep
1858 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1860 59. Two changes related to the smtp_active_hostname option:
1862 (1) $smtp_active_hostname is now available as a variable.
1863 (2) The default for smtp_banner uses $smtp_active_hostname instead
1864 of $primary_hostname.
1866 60. The host_aton() function is supposed to be passed a string that is known
1867 to be a valid IP address. However, in the case of IPv6 addresses, it was
1868 not checking this. This is a hostage to fortune. Exim now panics and dies
1869 if the condition is not met. A case was found where this could be provoked
1870 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1871 components; fortuitously, this particular loophole had already been fixed
1872 by change 4.50/55 above.
1874 If there are any other similar loopholes, the new check in host_aton()
1875 itself should stop them being exploited. The report I received stated that
1876 data on the command line could provoke the exploit when Exim was running as
1877 exim, but did not say which command line option was involved. All I could
1878 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1879 running as the user.
1882 61. There was a buffer overflow vulnerability in the SPA authentication code
1883 (which came originally from the Samba project). I have added a test to the
1884 spa_base64_to_bits() function which I hope fixes it.
1887 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1888 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1890 63. The daemon start-up calls getloadavg() while still root for those OS that
1891 need the first call to be done as root, but it missed one case: when
1892 deliver_queue_load_max is set with deliver_drop_privilege. This is
1893 necessary for the benefit of the queue runner, because there is no re-exec
1894 when deliver_drop_privilege is set.
1896 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1897 This has been fixed.
1899 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1900 was in use, was not putting the data itself into the right store pool;
1901 consequently, it could be overwritten for a subsequent message in the same
1902 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1905 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1907 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1908 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1909 the function were treating the return as a boolean value, which happened to
1910 work because 0=false and not-0=true, but is not correct code.
1912 68. The host_aton() function was not handling scoped IPv6 addresses (those
1913 with, for example, "%eth0" on the end) correctly.
1915 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1916 negated items (that is, ~something) in unsigned ints. Some compilers
1917 apparently mutter when there is no cast.
1919 70. If an address verification called from an ACL failed, and did not produce a
1920 user-specific message (i.e. there was only a "system" message), nothing was
1921 put in $acl_verify_message. In this situation, it now puts the system
1924 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1925 session; change 4.31/43 added the unwanted input to the log line - except
1926 that it did not do this in the start of session case. It now does.
1928 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1929 This is wrong and can cause the other end to generate a synchronization
1930 error if it is another Exim or anything else that does the synchronization
1931 check. A QUIT command is no longer sent after a timeout.
1933 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1934 during host lookups.
1936 74. The defer_ok option of callout verification was not working if it was used
1937 when verifying addresses in header lines, that is, for this case:
1939 verify = header_sender/callout=defer_ok
1941 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1942 those file descriptors could be used for SMTP connections. If anything
1943 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1944 could be sent to the SMTP client, causing chaos. The daemon now opens
1945 stdin, stdout, and stderr to /dev/null when it puts itself into the
1948 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1949 log by default. The user can override this with suitable Perl magic.
1951 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1952 the log message when discard triggers, was not working for the DATA ACL or
1953 for the non-SMTP ACL.
1955 78. Error message wording change in sieve.c.
1957 79. If smtp_accept_max_per_host was set, the number of connections could be
1958 restricted to fewer than expected, because the daemon was trying to set up
1959 a new connection before checking whether the processes handling previous
1960 connections had finished. The check for completed processes is now done
1961 earlier. On busy systems, this bug wouldn't be noticed because something
1962 else would have woken the daemon, and it would have reaped the completed
1965 80. If a message was submitted locally by a user whose login name contained one
1966 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1967 It caused a spool format error. I have fixed the spool reading code. A
1968 related problem was that the "from" clause in the Received: line became
1969 illegal because of the space(s). It is now covered by ${quote_local_part.
1971 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
1974 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
1976 83. Patch from Sieve maintainer to fix unterminated string problem in
1977 "vacation" handling.
1979 84. Some minor changes to the Linux configuration files to help with other
1980 OS variants using glibc.
1982 85. One more patch for Sieve to update vacation handling to latest spec.
1985 ----------------------------------------------------
1986 See the note above about the 4.44 and 4.50 releases.
1987 ----------------------------------------------------
1993 1. Change 4.43/35 introduced a bug that caused file counts to be
1994 incorrectly computed when quota_filecount was set in an appendfile
1997 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1998 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2001 3. Give more explanation in the error message when the command for a transport
2002 filter fails to execute.
2004 4. There are several places where Exim runs a non-Exim command in a
2005 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2006 was being done only for the command run by the queryprogram router. It is
2007 now done for all such subprocesses. The other cases are: ${run, transport
2008 filters, and the commands run by the lmtp and pipe transports.
2010 5. Some older OS have a limit of 256 on the maximum number of file
2011 descriptors. Exim was using setrlimit() to set 1000 as a large value
2012 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2013 systems. I've change it so that if it can't get 1000, it tries for 256.
2015 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2016 was an oversight, and furthermore, ever since the addition of extra
2017 controls (e.g. 4.43/32), the checks on when to allow different forms of
2018 "control" were broken. There should now be diagnostics for all cases when a
2019 control that does not make sense is encountered.
2021 7. $recipients is now available in the predata ACL (oversight).
2023 8. Tidy the search cache before the fork to do a delivery from a message
2024 received from the command line. Otherwise the child will trigger a lookup
2025 failure and thereby defer the delivery if it tries to use (for example) a
2026 cached ldap connection that the parent has called unbind on.
2028 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2029 of $address_data from the recipient verification was clobbered by the
2030 sender verification.
2032 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2033 was its contents. (It was OK if the option was not defined at all.)
2035 11. A "Completed" log line is now written for messages that are removed from
2036 the spool by the -Mrm option.
2038 12. $host_address is now set to the target address during the checking of
2039 ignore_target_hosts.
2041 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2042 being passed; this would have caused $sender_host_name to have been used if
2043 matching the list had actually called for a host name (not very likely,
2044 since this list is usually IP addresses). A host name is now passed as
2047 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2048 code that specifies a non-restarting handler (typically sigaction() in
2049 modern systems) in an attempt to fix a rare and obscure crash bug.
2051 15. Narrowed the window for a race in the daemon that could cause it to ignore
2052 SIGCHLD signals. This is not a major problem, because they are used only to
2053 wake it up if nothing else does.
2055 16. A malformed maildirsize file could cause Exim to calculate negative values
2056 for the mailbox size or file count. Odd effects could occur as a result.
2057 The maildirsize information is now recalculated if the size or filecount
2060 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2061 support for a long time. Removed HAVE_SYS_VFS_H.
2063 18. Updated exipick to current release from John Jetmore.
2065 19. Allow an empty sender to be matched against a lookup in an address list.
2066 Previously the only cases considered were a regular expression, or an
2069 20. Exim went into a mad DNS lookup loop when doing a callout where the
2070 host was specified on the transport, if the DNS lookup yielded more than
2073 21. The RFC2047 encoding function was originally intended for short strings
2074 such as real names; it was not keeping to the 75-character limit for
2075 encoded words that the RFC imposes. It now respects the limit, and
2076 generates multiple encoded words if necessary. To be on the safe side, I
2077 have increased the buffer size for the ${rfc2047: expansion operator from
2080 22. Failure to deliver a bounce message always caused it to be frozen, even if
2081 there was an errors_to setting on the router. The errors_to setting is now
2084 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2085 canonical form (fully expanded) before being placed in
2086 $sender_host_address.
2088 24. Updated eximstats to version 1.33
2090 25. Include certificate and key file names in error message when GnuTLS fails
2091 to set them up, because the GnuTLS error message doesn't include the name
2092 of the failing file when there is a problem reading it.
2094 26. Expand error message when OpenSSL has problems setting up cert/key files.
2097 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2098 (this can affect the format of dates).
2100 28. exim_tidydb, when checking for the continued existence of a message for
2101 which it has found a message-specific retry record, was not finding
2102 messages that were in split spool directories. Consequently, it was
2103 deleting retry records that should have stayed in existence.
2105 29. eximstats updated to version 1.35
2106 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2107 1.35 - bugfix such that pie charts by volume are generated correctly
2109 30. The SPA authentication driver was not abandoning authentication and moving
2110 on to the next authenticator when an expansion was forced to fail,
2111 contradicting the general specification for all authenticators. Instead it
2112 was generating a temporary error. It now behaves as specified.
2114 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2115 (the order specifies the preference for clients). The order is now AES256,
2116 AES128, 3DES, ARCFOUR128.
2118 31. Small patch to Sieve code - explicitly set From: when generating an
2121 32. Exim crashed if a remote delivery caused a very long error message to be
2122 recorded - for instance if somebody sent an entire SpamAssassin report back
2123 as a large number of 550 error lines. This bug was coincidentally fixed by
2124 increasing the size of one of Exim's internal buffers (big_buffer) that
2125 happened as part of the Exiscan merge. However, to be on the safe side, I
2126 have made the code more robust (and fixed the comments that describe what
2129 33. Some experimental protocols are using DNS PTR records for new purposes. The
2130 keys for these records are domain names, not reversed IP addresses. The
2131 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2132 leaves it alone. Component reversal etc. now happens only for IP addresses.
2135 34. The host_aton() function is supposed to be passed a string that is known
2136 to be a valid IP address. However, in the case of IPv6 addresses, it was
2137 not checking this. This is a hostage to fortune. Exim now panics and dies
2138 if the condition is not met. A case was found where this could be provoked
2139 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2140 components; fortuitously, this particular loophole had already been fixed
2141 by change 4.50/55 or 4.44/33 above.
2143 If there are any other similar loopholes, the new check in host_aton()
2144 itself should stop them being exploited. The report I received stated that
2145 data on the command line could provoke the exploit when Exim was running as
2146 exim, but did not say which command line option was involved. All I could
2147 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2148 running as the user.
2151 35. There was a buffer overflow vulnerability in the SPA authentication code
2152 (which came originally from the Samba project). I have added a test to the
2153 spa_base64_to_bits() function which I hope fixes it.
2156 36. The daemon start-up calls getloadavg() while still root for those OS that
2157 need the first call to be done as root, but it missed one case: when
2158 deliver_queue_load_max is set with deliver_drop_privilege. This is
2159 necessary for the benefit of the queue runner, because there is no re-exec
2160 when deliver_drop_privilege is set.
2162 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2163 was in use, was not putting the data itself into the right store pool;
2164 consequently, it could be overwritten for a subsequent message in the same
2165 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2168 38. Sometimes the final signoff response after QUIT could fail to get
2169 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2170 < 0 before doing a fflush(). This bug looks as though it goes back to the
2171 introduction of TLS in release 3.20, but "sometimes" must have been rare
2172 because the tests only now provoked it.
2178 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2179 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2180 Now it returns an integer. A number of calls were still expecting a T/F
2181 return. Fortuitously, in all cases, the tests worked in OK situations,
2182 which is the norm. However, things would have gone wrong on any write
2183 failures on the smtp file descriptor. This function is used when sending
2184 messages over SMTP and also when doing verify callouts.
2186 2. When Exim is called to do synchronous delivery of a locally submitted
2187 message (the -odf or -odi options), it no longer closes stderr before doing
2190 3. Implemented the mua_wrapper option.
2192 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2194 5. Implemented the functions header_remove(), header_testname(),
2195 header_add_at_position(), and receive_remove_recipient(), and exported them
2198 6. If an ACL "warn" statement specified the addition of headers, Exim already
2199 inserted X-ACL-Warn: at the start if there was no header name. However, it
2200 was not making this test for the second and subsequent header lines if
2201 there were newlines in the string. This meant that an invalid header could
2202 be inserted if Exim was badly configured.
2204 7. Allow an ACL "warn" statement to add header lines at the start or after all
2205 the Received: headers, as well as at the end.
2207 8. Added the rcpt_4xx retry error code.
2209 9. Added postmaster_mailfrom=xxx to callout verification option.
2211 10. Added mailfrom=xxxx to the callout verification option, for verify=
2214 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2215 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2217 12. Inserted some casts to stop certain compilers warning when using pointer
2218 differences as field lengths or precisions in printf-type calls (mostly
2219 affecting debugging statements).
2221 13. Added optional readline() support for -be (dynamically loaded).
2223 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2224 same clock tick as a message's arrival, so that its received time was the
2225 same as the "first fail" time on the retry record, and that message
2226 remained on the queue past the ultimate address timeout, every queue runner
2227 would try a delivery (because it was past the ultimate address timeout) but
2228 after another failure, the ultimate address timeout, which should have then
2229 bounced the address, did not kick in. This was a "< instead of <=" error;
2230 in most cases the first failure would have been in the next clock tick
2231 after the received time, and all would be well.
2233 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2234 being recognized when the domain list was tested by the match_domain
2235 condition in an expansion string.
2237 16. Added the ${str2b64: operator.
2239 17. Exim was always calling setrlimit() to set a large limit for the number of
2240 processes, without checking whether the existing limit was already
2241 adequate. (It did check for the limit on file descriptors.) Furthermore,
2242 errors from getrlimit() and setrlimit() were being ignored. Now they are
2243 logged to the main and panic logs, but Exim does carry on, to try to do its
2244 job under whatever limits there are.
2246 18. Imported PCRE 5.0.
2248 19. Trivial typo in log message " temporarily refused connection" (the leading
2251 20. If the log selector return_path_on_delivery was set and an address was
2252 redirected to /dev/null, the delivery process crashed because it assumed
2253 that a return path would always be set for a "successful" delivery. In this
2254 case, the whole delivery is bypassed as an optimization, and therefore no
2257 21. Internal re-arrangement: the function for sending a challenge and reading
2258 a response while authentication was assuming a zero-terminated challenge
2259 string. It's now changed to take a pointer and a length, to allow for
2260 binary data in such strings.
2262 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2264 23. Exim was not respecting finduser_retries when seeking the login of the
2265 uid under which it was called; it was always trying 10 times. (The default
2266 setting of finduser_retries is zero.) Also, it was sleeping after the final
2267 failure, which is pointless.
2269 24. Implemented tls_on_connect_ports.
2271 25. Implemented acl_smtp_predata.
2273 26. If the domain in control=submission is set empty, Exim assumes that the
2274 authenticated id is a complete email address when it generates From: or
2275 Sender: header lines.
2277 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2278 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2279 chown and chgrp in /bin and hostname in /usr/bin.
2281 28. Exim was keeping the "process log" file open after each use, just as it
2282 does for the main log. This opens the possibility of it remaining open for
2283 long periods when the USR1 signal hits a daemon. Occasional processlog
2284 errors were reported, that could have been caused by this. Anyway, it seems
2285 much more sensible not to leave this file open at all, so that is what now
2288 29. The long-running daemon process does not normally write to the log once it
2289 has entered its main loop, and it closes the log before doing so. This is
2290 so that log files can straightforwardly be renamed and moved. However,
2291 there are a couple of unusual error situations where the daemon does write
2292 log entries, and I had neglected to close the log afterwards.
2294 30. The text of an SMTP error response that was received during a remote
2295 delivery was being truncated at 512 bytes. This is too short for some of
2296 the long messages that one sometimes sees. I've increased the limit to
2299 31. It is now possible to make retry rules that apply only when a message has a
2300 specific sender, in particular, an empty sender.
2302 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2303 it possible to be selective about when SMTP synchronization is enforced.
2305 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2307 32. Implemented hosts_connection_nolog.
2309 33. Added an ACL for QUIT.
2311 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2314 35. Added mailbox_size and mailbox_filecount to appendfile.
2316 36. Added control = no_multiline_responses to ACLs.
2318 37. There was a bug in the logic of the code that waits for the clock to tick
2319 in the case where the clock went backwards by a substantial amount such
2320 that the microsecond fraction of "now" was more than the microsecond
2321 fraction of "then" (but the whole seconds number was less).
2323 38. Added support for the libradius Radius client library this is found on
2324 FreeBSD (previously only the radiusclient library was supported).
2330 1. When certain lookups returned multiple values in the form name=value, the
2331 quoting of the values was not always being done properly. Specifically:
2332 (a) If the value started with a double quote, but contained no whitespace,
2334 (b) If the value contained whitespace other than a space character (i.e.
2335 tabs or newlines or carriage returns) it was not quoted.
2336 This fix has been applied to the mysql and pgsql lookups by writing a
2337 separate quoting function and calling it from the lookup code. The fix
2338 should probably also be applied to nisplus, ibase and oracle lookups, but
2339 since I cannot test any of those, I have not disturbed their existing code.
2341 2. A hit in the callout cache for a specific address caused a log line with no
2342 reason for rejecting RCPT. Now it says "Previous (cached) callout
2343 verification failure".
2345 3. There was an off-by-one bug in the queryprogram router. An over-long
2346 return line was truncated at 256 instead of 255 characters, thereby
2347 overflowing its buffer with the terminating zero. As well as fixing this, I
2348 have increased the buffer size to 1024 (and made a note to document this).
2350 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2351 when Exim is waiting for an SMTP response from a remote server, Exim
2352 restarts its select() call on the socket, thereby resetting its timeout.
2353 This is not a problem when such interrupts are rare. Somebody set up a cron
2354 job to run exiwhat every 2 minutes, which is less than the normal select()
2355 timeout (5 or 10 minutes). This meant that the select() timeout never
2356 kicked in because it was always reset. I have fixed this by comparing the
2357 time when an interrupt arrives with the time at the start of the first call
2358 to select(). If more time than the timeout has elapsed, the interrupt is
2359 treated as a timeout.
2361 5. Some internal re-factoring in preparation for the addition of Sieve
2362 extensions (by MH). In particular, the "personal" test is moved to a
2363 separate function, and given an option for scanning Cc: and Bcc: (which is
2364 not set for Exim filters).
2366 6. When Exim created an email address using the login of the caller as the
2367 local part (e.g. when creating a From: or Sender: header line), it was not
2368 quoting the local part when it contained special characters such as @.
2370 7. Installed new OpenBSD configuration files.
2372 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2373 try to make them clearer.
2375 9. Callout options, other than the timeout value, were being ignored when
2376 verifying sender addresses in header lines. For example, when using
2378 verify = header_sender/callout=no_cache
2380 the cache was (incorrectly) being used.
2382 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2383 only the Cygwin environment.
2385 11. When return_path_on_delivery was set as a log selector, if different remote
2386 addresses in the same message used different return paths and parallel
2387 remote delivery occurred, the wrong values would sometimes be logged.
2388 (Whenever a remote delivery process finished, the return path value from
2389 the most recently started remote delivery process was logged.)
2391 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2392 lines when AUTH and/or TLS are in use. This is the "received protocol"
2393 field. Exim used to use "asmtp" for authenticated SMTP, without any
2394 indication (in the protocol name) for TLS use. Now it follows the RFC and
2395 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2396 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2397 names appear in log lines as well as in Received: header lines.
2399 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2400 extensions, and comparison tests, and to fix some bugs.
2402 14. Changes to the "personal" filter test:
2404 (1) The test was buggy in that it was just doing the equivalent of
2405 "contains" tests on header lines. For example, if a user's address was
2406 anne@some.where, the "personal" test would incorrectly be true for
2408 To: susanne@some.where
2410 This test is now done by extracting each address from the header in turn,
2411 and checking the entire address. Other tests that are part of "personal"
2412 are now done using regular expressions (for example, to check local parts
2413 of addresses in From: header lines).
2415 (2) The list of non-personal local parts in From: addresses has been
2416 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2417 taken from the Sieve specification recommendations.
2419 (3) If the message contains any header line starting with "List-" it is
2420 treated as non-personal.
2422 (4) The test for "circular" in the Subject: header line has been removed
2423 because it now seems ill-conceived.
2425 15. Minor typos in src/EDITME comments corrected.
2427 16. Installed latest exipick from John Jetmore.
2429 17. If headers_add on a router specified a text string that was too long for
2430 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2431 of string_sprintf() is now avoided.
2433 18. $message_body_size was not set (it was always zero) when running the DATA
2434 ACL and the local_scan() function.
2436 19. For the "mail" command in an Exim filter, no default was being set for
2437 the once_repeat time, causing a random time value to be used if "once" was
2438 specified. (If the value happened to be <= 0, no repeat happened.) The
2439 default is now 0s, meaning "never repeat". The "vacation" command was OK
2440 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2441 (I found it when inspecting the code).
2443 20. There is now an overall timeout for performing a callout verification. It
2444 defaults to 4 times the callout timeout, which applies to individual SMTP
2445 commands during the callout. The overall timeout applies when there is more
2446 than one host that can be tried. The timeout is checked before trying the
2447 next host. This prevents very long delays if there are a large number of
2448 hosts and all are timing out (e.g. when the network connections are timing
2449 out). The value of the overall timeout can be changed by specifying an
2450 additional sub-option for "callout", called "maxwait". For example:
2452 verify = sender/callout=5s,maxwait=20s
2454 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2455 to the end before writing, but this should make it even safer).
2457 22. Exim was forgetting that it had advertised PIPELINING for the second and
2458 subsequent messages on an SMTP connection. It was also not resetting its
2459 memory on STARTTLS and an internal HELO.
2461 23. When Exim logs an SMTP synchronization error within a session, it now
2462 records whether PIPELINING has been advertised or not.
2464 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2465 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2466 rather than long int.
2468 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2470 26. Added the never_mail option to autoreply.
2476 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2477 crash if the getsockname() call failed; this can happen if a connection is
2478 closed very soon after it is established. The problem was simply in the
2479 order in which certain operations were done, causing Exim to try to write
2480 to the SMTP stream before it had set up the file descriptor. The bug has
2481 been fixed by making things happen in the correct order.
2487 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2488 before the connection was closed, thus losing the rejection response.
2490 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2491 some early Solaris releases, but causes trouble in current releases where
2492 socklen_t is defined.
2494 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2497 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2500 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2501 that releasing the top of it at the end releases what was used for sub-
2502 expansions (unless the block got too big). However, discard this block if
2503 the first thing is a variable or header, so that we can use its block when
2504 it is dynamic (useful for very large $message_headers, for example).
2506 6. Lookups now cache *every* query, not just the most recent. A new, separate
2507 store pool is used for this. It can be recovered when all lookup caches are
2508 flushed. Lookups now release memory at the end of their result strings.
2509 This has involved some general refactoring of the lookup sources.
2511 7. Some code has been added to the store_xxx() functions to reduce the amount
2512 of flapping under certain conditions.
2514 8. log_incoming_interface used to affect only the <= reception log lines. Now
2515 it causes the local interface and port to be added to several more SMTP log
2516 lines, for example "SMTP connection from", and rejection lines.
2518 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2520 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2522 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2523 could be overwritten at the end of the current message (or the start of a
2524 new message if it was set in a HELO ACL). The value is now preserved for
2525 the duration of the SMTP connection.
2527 12. If a transport had a headers_rewrite setting, and a matching header line
2528 contained an unqualified address, that address was qualified, even if it
2529 did not match any rewriting rules. The underlying bug was that the values
2530 of the flags that permit the existence of unqualified sender and recipient
2531 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2532 non-local messages, and by -bnq for local messages) were not being
2533 preserved with the message after it was received.
2535 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2536 "next input=" as part of the text comprising the host identity instead of
2537 the correct text. The code was using the same buffer for two different
2538 strings. However, depending on which order the printing function evaluated
2539 its arguments, the bug did not always show up. Under Linux, for example, my
2540 test suite worked just fine.
2542 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2543 doesn't work with some older versions of Perl. It has been changed to "my",
2544 which in any case is probably the better facility to use.
2546 15. A really picky compiler found some instances of statements for creating
2547 error messages that either had too many or two few arguments for the format
2550 16. The size of the buffer for calls to the DNS resolver has been increased
2551 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2552 for addresses that have a lot of PTR records. This alleviates a problem; it
2553 does not fully solve it.
2555 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2556 buffer now truncates the list and logs the incident, which is the same
2557 action as happens when Exim is looking up a host name and its aliases.
2558 Previously in this situation something unpredictable would happen;
2559 sometimes it was "internal error: store_reset failed".
2561 18. If a server dropped the connection unexpectedly when an Exim client was
2562 using GnuTLS and trying to read a response, the client delivery process
2563 crashed while trying to generate an error log message.
2565 19. If a "warn" verb in an ACL added multiple headers to a message in a single
2566 string, for example:
2568 warn message = H1: something\nH2: something
2570 the text was added as a single header line from Exim's point of view
2571 though it ended up OK in the delivered message. However, searching for the
2572 second and subsequent header lines using $h_h2: did not work. This has been
2573 fixed. Similarly, if a system filter added multiple headers in this way,
2574 the routers could not see them.
2576 20. Expanded the error message when iplsearch is called with an invalid key to
2577 suggest using net-iplsearch in a host list.
2579 21. When running tests using -bh, any delays imposed by "delay" modifiers in
2580 ACLs are no longer actually imposed (and a message to that effect is
2583 22. If a "gecos" field in a passwd entry contained escaped characters, in
2584 particular, if it contained a \" sequence, Exim got it wrong when building
2585 a From: or a Sender: header from that name. A second bug also caused
2586 incorrect handling when an unquoted " was present following a character
2587 that needed quoting.
2589 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2590 was not being matched caselessly.
2592 24. Arranged for all hyphens in the exim.8 source to be escaped with
2595 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2596 properly. Recipient callout cache records were still being keyed to include
2597 the sender, even when use_sender was set false. This led to far more
2598 callouts that were necessary. The sender is no longer included in the key
2599 when use_sender is false.
2601 26. Added "control = submission" modifier to ACLs.
2603 27. Added the ${base62d: operator to decode base 62 numbers.
2605 28. dnsdb lookups can now access SRV records.
2607 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2608 the configuration file.
2610 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2611 (-v) mode. This makes the output for a verbose queue run more intelligible.
2613 31. Added a use_postmaster feature to recipient callouts.
2615 32. Added the $body_zerocount variable, containing the number of binary zero
2616 bytes in the message body.
2618 33. The time of last modification of the "new" subdirectory is now used as the
2619 "mailbox time last read" when there is a quota error for a maildir
2622 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2624 35. Added +ignore_unknown as a special item in host lists.
2626 36. Code for decoding IPv6 addresses in host lists is now included, even if
2627 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2628 address was recognized as an IP address, but was then not correctly decoded
2629 into binary, causing unexpected and incorrect effects when compared with
2636 1. Very minor rewording of debugging text in manualroute to say "list of
2637 hosts" instead of "hostlist".
2639 2. If verify=header_syntax was set, and a header line with an unqualified
2640 address (no domain) and a large number of spaces between the end of the
2641 name and the colon was received, the reception process suffered a buffer
2642 overflow, and (when I tested it) crashed. This was caused by some obsolete
2643 code that should have been removed. The fix is to remove it!
2645 3. When running in the test harness, delay a bit after writing a bounce
2646 message to get a bit more predictability in the log output.
2648 4. Added a call to search_tidyup() just before forking a reception process. In
2649 theory, someone could use a lookup in the expansion of smtp_accept_max_
2650 per_host which, without the tidyup, could leave open a database connection.
2652 5. Added the variables $recipient_data and $sender_data which get set from a
2653 lookup success in an ACL "recipients" or "senders" condition, or a router
2654 "senders" option, similar to $domain_data and $local_part_data.
2656 6. Moved the writing of debug_print from before to after the "senders" test
2659 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2660 problems for message scanning, either using a data ACL, or using
2661 local_scan() because the Received: header was not generated till after they
2662 were called (in order to set the time as the time of reception completion).
2663 I have revised the way this works. The header is now generated after the
2664 body is received, but before the ACL or local_scan() are called. After they
2665 are run, the timestamp in the header is updated.
2671 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2672 before starting a queue runner without re-exec. This happened only when
2673 deliver_drop_privilege was set or when the Exim user was set to root. The
2674 effect of the bug was that timeouts during subsequent deliveries caused
2675 crashes instead of being properly handled. The handler is now left at its
2676 default (and expected) setting.
2678 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2679 message, again when deliver_drop_privilege is set or Exim is run as root.
2680 The bug described in (1) was not present in this case, but the tidying up
2681 of the other signals was missing. I have made the two cases consistent.
2683 3. The ignore_target_hosts setting on a manualroute router was being ignored
2684 for hosts that were looked up using the /MX notation.
2686 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2689 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2690 operated on the sender address. After changing the $sender_address to <>
2691 for the sender address verify, Exim was re-instated it as the original
2692 (before rewriting) address, but remembering that it had rewritten it, so it
2693 wasn't rewriting it again. This bug also had the effect of breaking the
2694 sender address verification caching when the sender address was rewritten.
2696 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2697 This has been changed so that if the ip literal address matches
2698 ignore_target_hosts, the router declines.
2700 7. Added expansion conditions match_domain, match_address, and match_local_
2701 part (NOT match_host).
2703 8. The placeholder for the Received: header didn't have a length field set.
2705 9. Added code to Exim itself and to exim_lock to test for a specific race
2706 condition that could lead to file corruption when using MBX delivery. The
2707 issue is with the lockfile that is created in /tmp. If this file is removed
2708 after a process has opened it but before that process has acquired a lock,
2709 there is the potential for a second process to recreate the file and also
2710 acquire a lock. This could lead to two Exim processes writing to the file
2711 at the same time. The added code performs the same test as UW imapd; it
2712 checks after acquiring the lock that its file descriptor still refers to
2713 the same named file.
2715 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2716 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2719 11. Added the smtp_active_hostname option. If used, this will typically be made
2720 to depend on the incoming interface address. Because $interface_address is
2721 not set up until the daemon has forked a reception process, error responses
2722 that can happen earlier (such as "too many connections") no longer contain
2725 12. If an expansion in a condition on a "warn" statement fails because a lookup
2726 defers, the "warn" statement is abandoned, and the next ACL statement is
2727 processed. Previously this caused the whole ACL to be aborted.
2729 13. Added the iplsearch lookup type.
2731 14. Added ident_timeout as a log selector.
2733 15. Added tls_certificate_verified as a log selector.
2735 16. Added a global option tls_require_ciphers (compare the smtp transport
2736 option of the same name). This controls incoming TLS connections.
2738 17. I finally figured out how to make tls_require_ciphers do a similar thing
2739 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2740 before starting the TLS session.
2742 18. Tabs are now shown as \t in -bP output.
2744 19. If the log selector return_path_on_delivery was set, Exim crashed when
2745 bouncing a message because it had too many Received: header lines.
2747 20. If two routers both had headers_remove settings, and the first one included
2748 a superfluous trailing colon, the final name in the first list and the
2749 first name in the second list were incorrectly joined into one item (with a
2750 colon in the middle).
2756 1. Added -C and -D options to the exinext utility, mainly to make it easier
2757 to include in the automated testing, but these could be helpful when
2758 multiple configurations are in use.
2760 2. The exinext utility was not formatting the output nicely when there was
2761 an alternate port involved in the retry record key, nor when there was a
2762 message id as well (for retries that were specific to a specific message
2763 and a specific host). It was also confused by IPv6 addresses, because of
2764 the additional colons they contain. I have fixed the IPv4 problem, and
2765 patched it up to do a reasonable job for IPv6.
2767 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2768 delivery, the log line now contains "pipelined" if PIPELINING was used.
2770 4. An SMTP transport process used to panic and die if the bind() call to set
2771 an explicit outgoing interface failed. This has been changed; it is now
2772 treated in the same way as a connect() failure.
2774 5. A reference to $sender_host_name in the part of a conditional expansion
2775 that was being skipped was still causing a DNS lookup. This no longer
2778 6. The def: expansion condition was not recognizing references to header lines
2779 that used bh_ and bheader_.
2781 7. Added the _cache feature to named lists.
2783 8. The code for checking quota_filecount in the appendfile transport was
2784 allowing one more file than it should have been.
2786 9. For compatibility with Sendmail, the command line option
2794 and sets the incoming protocol and host name (for trusted callers). The
2795 host name and its colon can be omitted when only the protocol is to be set.
2796 Note the Exim already has two private options, -pd and -ps, that refer to
2797 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2798 "s", but I don't think that's a major issue.
2800 10. A number of refactoring changes to the code, none of which should affect
2803 (a) The number of logging options was getting close to filling up the
2804 32-bit word that was used as a bit map. I have split them into two classes:
2805 those that are passed in the argument to log_write(), and those that are
2806 only ever tested independently outside of that function. These are now in
2807 separate 32-bit words, so there is plenty of room for expansion again.
2808 There is no change in the user interface or the logging behaviour.
2810 (b) When building, for example, log lines, the code previously used a
2811 macro that called string_cat() twice, in order to add two strings. This is
2812 not really sufficiently general. Furthermore, there was one instance where
2813 it was actually wrong because one of the argument was used twice, and in
2814 one call a function was used. (As it happened, calling the function twice
2815 did not affect the overall behaviour.) The macro has been replaced by a
2816 function that can join an arbitrary number of extra strings onto a growing
2819 (c) The code for expansion conditions now uses a table and a binary chop
2820 instead of a serial search (which was left over from when there were very
2821 few conditions). Also, it now recognizes conditions like "pam" even when
2822 the relevant support is not compiled in: a suitably worded error message is
2823 given if an attempt is made to use such a condition.
2825 11. Added ${time_interval:xxxxx}.
2827 12. A bug was causing one of the ddress fields not to be passed back correctly
2828 from remote delivery subprocesses. The field in question was not being
2829 subsequently used, so this caused to problems in practice.
2831 13. Added new log selectors queue_time and deliver_time.
2833 14. Might have fixed a bug in maildirsizefile handling that threw up
2834 "unexpected character" debug warnings, and recalculated the data
2835 unnecessarily. In any case, I expanded the warning message to give more
2838 15. Added the message "Restricted characters in address" to the statements in
2839 the default ACL that block characters like @ and % in local parts.
2841 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2842 Three changes have been made:
2844 (a) There was a serious bug; a negative response to MAIL caused the whole
2845 recipient domain to be cached as invalid, thereby blocking all messages
2846 to all local parts at the same domain, from all senders. This bug has
2847 been fixed. The domain is no longer cached after a negative response to
2848 MAIL if the sender used is not empty.
2850 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2853 (c) A new callout option, "use_sender" has been added for people who want
2854 the modified behaviour.
2860 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2863 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2864 indeed breaks things for older releases.
2866 3. Added additional logging to the case where there is a problem reading data
2867 from a filter that is running in a subprocess using a pipe, in order to
2868 try to track down a specific problem.
2870 4. Testing facility fudge: when running in the test harness and attempting
2871 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2872 getting "No route to host". Convert this to a timeout.
2874 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2877 6. Some OS don't have socklen_t but use size_t instead. This affects the
2878 fifth argument of getsockopt() amongst other things. This is now
2879 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2880 can be set for individual OS. I have set it for SunOS5, OSF1, and
2881 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2882 some earlier ones do not.
2884 7. Change 4.30/15 was not doing the test caselessly.
2886 8. The standard form for an IPv6 address literal was being rejected by address
2887 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2888 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2889 this, as well as the form without the "IPv6" on the front (but only when
2890 address literals are enabled, of course).
2892 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2894 10. Exim crashed if a message with an empty sender address specified by -f
2895 encountered a router with an errors_to setting. This could be provoked only
2896 by a command such as
2900 where an empty string was supplied; "<>" did not hit this bug.
2902 11. Installed PCRE release 4.5.
2904 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2905 remained set. It is now erased.
2907 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2908 times from message ids (which are base 36 rather than the normal 62).
2910 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2911 were being counted as actual protocol errors, and logged if the log
2912 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2913 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2914 and DATA following a set of rejected RCPTs do not count as protocol errors.
2915 In other words, Exim assumes they were pipelined, though this may not
2916 actually be the case. Of course, in all cases the client gets an
2917 appropriate error code.
2919 15. If a lookup fails in an ACL condition, a message about the failure may
2920 be available; it is used if testing the ACL cannot continue, because most
2921 such messages specify what the cause of the deferral is. However, some
2922 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2923 that caused an old message to be retained and used if a later statement
2924 caused a defer, replacing the real cause of the deferral.
2926 16. If an IP address had so many PTR records that the DNS lookup buffer
2927 was not large enough to hold them, Exim could crash while trying to process
2928 the truncated data. It now detects and logs this case.
2930 17. Further to 4.21/58, another change has been made: if (and only if) the
2931 first line of a message (the first header line) ends with CRLF, a bare LF
2932 in a subsequent header line has a space inserted after it, so as not to
2933 terminate the header.
2935 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2936 unnecessarily, used atoi() instead of strtol(), and didn't check the
2937 termination when getting file sizes from file names by regex.
2939 19. Completely re-implemented the support for maildirsize files, in the light
2940 of a number of problems with the previous contributed implementation
2941 (4.30/29). In particular:
2943 . If the quota is zero, the maildirsize file is maintained, but no quota is
2946 . If the maildir directory does not exist, it is created before any attempt
2947 to write a maildirsize file.
2949 . The quota value in the file is just a cache; if the quota is changed in
2950 the transport, the new value overrides.
2952 . A regular expression is available for excluding directories from the
2955 20. The autoreply transport checks the characters in options that define the
2956 message's headers; it allows continued headers, but it was checking with
2957 isspace() after an embedded newline instead of explicitly looking for a
2960 21. If all the "regular" hosts to which an address was routed had passed their
2961 expiry times, and had not reached their retry times, the address was
2962 bounced, even if fallback hosts were defined. Now Exim should go on to try
2965 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2966 equivalent code in the SMTP transport. Some hosts send humungous responses
2967 to HELO/EHLO, more than 1024 it seems.
2969 23. Refactoring: code in filter.c used (void *) for "any old type" but this
2970 gives compiler warnings in some environments. I've now done it "properly",
2973 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2974 (because of problems with the built-in one) was declared to return uschar *
2975 instead of char *, causing compiler failure.
2977 25. Fixed a file descriptor leak when processing alias/forward files.
2979 26. Fixed a minor format string issue in dbfn.c.
2981 27. Typo in exim.c: ("dmbnz" for "dbmnz").
2983 28. If a filter file refered to $h_xxx or $message_headers, and the headers
2984 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2987 29. When a sender address is verified, it is cached, to save repeating the test
2988 when there is more than one recipient in a message. However, when the
2989 verification involves a callout, it is possible for different callout
2990 options to be set for different recipients. It is too complicated to keep
2991 track of this in the cache, so now Exim always runs a verification when a
2992 callout is required, relying on the callout cache for the optimization.
2993 The overhead is duplication of the address routing, but this should not be
2996 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
2997 to be verified with callout=postmaster, and the main callout worked but the
2998 postmaster check failed, the verification correctly failed. However, if a
2999 subsequent RCPT command asked for sender verification *without* the
3000 postmaster check, incorrect caching caused this verification also to fail,
3003 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3004 it was not caching the DNS options (qualify_single, search_parents) that
3005 were used when the lookup failed. A subsequent lookup with different
3006 options therefore always gave the same answer, though there were cases
3007 where it should not have. (Example: a "domains = !$mx_any" option on a
3008 dnslookup router: the "domains" option is always processed without any
3009 widening, but the router might have qualify_single set.) Now Exim uses the
3010 cached value only when the same options are set.
3012 32. Added John Jetmore's "exipick" utility to the distribution.
3014 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3015 than a timeout (e.g. a certificate is required, and is not provided), an
3016 Exim server now closes the connection immediately. Previously it waited for
3017 the client to close - but if the client is SSL, it seems that they each
3018 wait for each other, leading to a delay before one of them times out.
3020 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3021 maintained 0.8.x compatibility because I don't think many are using it, and
3022 it is clearly obsolete.
3024 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3027 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3028 client certificate was expired. A simple patch fixes this, though I don't
3029 understand the full logic of why the verify callback is called multiple
3032 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3033 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3034 which causes problems with some clients (such as the Certicom SSL Plus
3035 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3036 disables the coutermeasure allowing Eudora to connect."
3038 38. Exim was not checking that a write() to a log file succeeded. This could
3039 lead to Bad Things if a log got too big, in particular if it hit a file
3040 size limit. Exim now panics and dies if it cannot write to a log file, just
3041 as it does if it cannot open a log file.
3043 39. Modified OS/Makefile-Linux so that it now contains
3045 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3047 The two -D definitions ensure that Exim is compiled with large file
3048 support, which makes it possible to handle log files that are bigger than
3051 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3052 instance) a domain was checked against a named list that involved a lookup,
3053 causing $domain_data to be set, then another domain was checked against the
3054 same list, then the first domain was re-checked, the value of $domain_data
3055 after the final check could be wrong. In particular, if the second check
3056 failed, it could be set empty. This bug probably also applied to
3059 41. The strip_trailing_dot option was not being applied to the address given
3060 with the -f command-line option.
3062 42. The code for reading a message's header from the spool was incrementing
3063 $received_count, but never initializing it. This meant that the value was
3064 incorrect (doubled) while delivering a message in the same process in which
3065 it was received. In the most common configuration of Exim, this never
3066 happens - a fresh exec is done - but it can happen when
3067 deliver_drop_privilege is set.
3069 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3070 it now includes up to 150 characters of the unexpected data in the log
3073 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3074 and building data strings. The size of both of these buffers was 10 000
3075 bytes - far larger than anybody would *ever* want, thought I. Needless to
3076 say, somebody hit the limit. I have increased the maximum line length to
3077 20 000 and the maximum data length of concatenated lines to 100 000. I have
3078 also fixed two bugs, because there was no checking on these buffers. Tsk,
3079 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3080 buffer is too small.
3082 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3083 lsearch lookups. Now it does.
3085 46. When parsing a route_list item in a manualroute router, a fixed-length
3086 buffer was used for the list of hosts. I made this 1024 bytes long,
3087 thinking that nobody would ever have a list of hosts that long. Wrong.
3088 Somebody had a whole pile of complicated expansion conditions, and the
3089 string was silently truncated, leading to an expansion error. It turns out
3090 that it is easier to change to an unlimited length (owing to other changes
3091 that have happened since this code was originally written) than to build
3092 structure for giving a limitation error. The length of the item that
3093 expands into the list of hosts is now unlimited.
3095 47. The lsearch lookup could not handle data where the length of text line was
3096 more than 4095 characters. Such lines were truncated, leading to shortened
3097 data being returned. It should now handle lines of any length.
3099 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3100 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3103 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3104 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3105 no longer happens in comment lines. A list of replacements is now placed
3106 at the head of all of the source files, except those whose only change is
3107 to replace PERL_COMMAND in the very first #! line.
3109 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3110 messages on the queue, with a bottom-up merge sort, using code contributed
3111 by Michael Haardt. This should make operations like -bp somewhat faster on
3112 large queues. It won't affect queue runners, except when queue_run_in_order
3115 51. Installed eximstats 1.31 in the distribution.
3117 52. Added support for SRV lookups to the dnslookup router.
3119 53. If an ACL referred to $message_body or $message_body_end, the value was not
3120 reset for any messages that followed in the same SMTP session.
3122 54. The store-handling optimization for building very long strings was not
3123 differentiating between the different store pools. I don't think this
3124 actually made any difference in practice, but I've tidied it.
3126 55. While running the routers to verify a sender address, $sender_address
3127 was still set to the sender address. This is wrong, because when routing to
3128 send a bounce to the sender, it would be empty. Therefore, I have changed
3129 it so that, while verifying a sender address, $sender_address is set to <>.
3130 (There is no change to what happens when verifying a recipient address.)
3132 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3133 target A or AAAA records (if not already returned) without resetting the
3134 qualify_single or search_parents options of the DNS resolver. These are
3135 inappropriate in this case because the targets of MX and SRV records must
3136 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3137 target that, when qualified, matched something in the local domain. These
3138 two options are now turned off when doing these lookups.
3140 57. It seems that at least some releases of Reiserfs (which does not have the
3141 concept of a fixed number of inodes) returns zero and not -1 for the
3142 number of available inodes. This interacted badly with check_spool_inodes,
3143 which assumed that -1 was the "no such thing" setting. What I have done is
3144 to check that the total number of inodes is greater than zero before doing
3145 the test of how many are available.
3147 58. When a "warn" ACL statement has a log_message modifier, the message is
3148 remembered, and not repeated. This is to avoid a lot of repetition when a
3149 message has many recipients that cause the same warning to be written.
3150 Howewer, Exim was preserving the list of already written lines for an
3151 entire SMTP session, which doesn't seem right. The memory is now reset if a
3152 new message is started.
3154 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3155 debugging output unless log_rewrite was also set.
3157 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3158 of (int)(handle) when we know that handle contains (void *)(-1).
3160 61. The Exim daemon panic-logs an error return when it closes the incoming
3161 connection. However "connection reset by peer" seems to be common, and
3162 isn't really an error worthy of noting specially, so that particular error
3165 62. When Exim is trying to find all the local interfaces, it used to panic and
3166 die if the ioctl to get the interface flags failed. However, it seems that
3167 on at least one OS (Solaris 9) it is possible to have an interface that is
3168 included in the list of interfaces, but for which you get a failure error
3169 for this call. This happens when the interface is not "plumbed" into a
3170 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3171 failure of the "get flags" call assumes that the interface is down.
3173 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3174 makes life easier for people who are doing arithmetic on fields extracted
3175 from dates, where you often get leading zeros that should not be
3176 interpreted as octal.
3178 64. Added qualify_domain to the redirect router, to override the global
3181 65. If a pathologically long header line contained very many addresses (the
3182 report of this problem mentioned 10 000) and each of them was rewritten,
3183 Exim could use up a very large amount of memory. (It kept on making new
3184 copies of the header line as it rewrote, and never released the old ones.)
3185 At the expense of a bit more processing, the header rewriting function has
3186 been changed so that it no longer eats memory in this way.
3188 66. The generation of the Received: header has been moved from the time that a
3189 message starts to be received, to the time that it finishes. The timestamp
3190 in the Received: header should now be very close to that of the <= log
3191 line. There are two side-effects of this change:
3193 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3194 logged header lines no longer include the local Received: line, because
3195 it has not yet been created. The same applies to a copy of the message
3196 that is returned to a non-SMTP sender when a message is rejected.
3198 (b) When a filter file is tested using -bf, no additional Received: header
3199 is added to the test message. After some thought, I decided that this
3202 This change does not affect the value of $received_for. It is still set
3203 after address rewriting, but before local_scan() is called.
3205 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3207 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3208 gave an unhelpful panic error message, and a defer error. I have managed to
3209 change this behaviour so that it now rejects any supplied certificate,
3210 which seems right, as the list of acceptable certificates is empty.
3212 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3213 gave an unhelpful defer error. I have not managed to make this reject any
3214 supplied certificates, but the error message it gives is "no certificate
3215 supplied", which is not helpful.
3217 70. exigrep's output now also includes lines that are not associated with any
3218 message, but which match the given pattern. Implemented by a patch from
3219 Martin Sluka, which also tidied up the Perl a bit.
3221 71. Recipient callout verification, like sender verification, was using <> in
3222 the MAIL FROM command. This isn't really the right thing, since the actual
3223 sender may affect whether the remote host accepts the recipient or not. I
3224 have changed it to use the actual sender in the callout; this means that
3225 the cache record is now keyed on a recipient/sender pair, not just the
3226 recipient address. There doesn't seem to be a real danger of callout loops,
3227 since a callout by the remote host to check the sender would use <>.
3228 [SEE ABOVE: changed after hitting problems.]
3230 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3231 temporary errors. However, in the case of such a code being given after
3232 the end of a data transmission (i.e. after ".") Exim was failing to write
3233 a retry record for the message. (Yes, there was some broken host that was
3234 actually sending 8xx at this point.)
3236 73. An unknown lookup type in a host list could cause Exim to panic-die when
3237 the list was checked. (An example that provoked this was putting <; in the
3238 middle of a list instead of at the start.) If this happened during a DATA
3239 ACL check, a -D file could be left lying around. This kind of configuration
3240 error no longer causes Exim to die; instead it causes a defer errror. The
3241 incident is still logged to the main and panic logs.
3243 74. Buglet left over from Exim 3 conversion. The message "too many messages
3244 in one connection" was written to the rejectlog but not the mainlog, except
3245 when address rewriting (yes!) was being logged.
3247 75. Added write_rejectlog option.
3249 76. When a system filter was run not as root (that is, when system_filter_user
3250 was set), the values of the $n variables were not being returned to the
3251 main process; thus, they were not subsequently available in the $sn
3254 77. Added +return_path_on_delivery log selector.
3256 78. A connection timeout was being treated differently from recipients deferred
3257 when testing hosts_max_try with a message that was older than the host's
3258 retry timeout. (The host should not be counted, thus allowing all hosts to
3259 be tried at least once before bouncing.) This may have been the cause of an
3260 occasionally reported bug whereby a message would remain on the queue
3261 longer than the retry timeout, but would be bounced if a delivery was
3262 forced. I say "may" because I never totally pinned down the problem;
3263 setting up timeout/retry tests is difficult. See also the next item.
3265 79. The ultimate address timeout was not being applied to errors that involved
3266 a combination of host plus message (for example, a timeout on a MAIL
3267 command). When an address resolved to a number of possible hosts, and they
3268 were not all tried for each delivery (e.g. because of hosts_max_try), a
3269 message could remain on the queue longer than the retry timeout.
3271 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3274 81. Fixed an obscure SMTP outgoing bug which required at least the following
3275 conditions: (a) there was another message waiting for the same server;
3276 (b) the server returned 5xx to all RCPT commands in the first message so
3277 that the message was not completed; (c) the server dropped the connection
3278 or gave a negative response to the RSET that Exim sends to abort the
3279 transaction. The observed case was a dropped connection after DATA that had
3280 been sent in pipelining mode. That is, the server had advertised PIPELINING
3281 but was not implementing it correctly. The effect of the bug was incorrect
3282 behaviour, such as trying another host, and this could lead to a crash.
3288 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3289 and daemon.c were passed as pointers to ints; they should have been
3290 pointers to socklen_t variables (which are typically unsigned ints).
3292 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3295 3. Fixed a really odd bug that affected only the testing scheme; patching a
3296 certain fixed string in the binary changed the value of another string that
3297 happened to be identical to the end of the original first string.
3299 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3300 name", it returns that address as the IP address. On some operating
3301 systems (e.g. Solaris), it also passes back the IP address string as the
3302 "host name". However, on others (e.g. Linux), it passes back an empty
3303 string. Exim wasn't checking for this, and was changing the host name to an
3304 empty string, assuming it had been canonicized.
3306 5. Although rare, it is permitted to have more than one PTR record for a given
3307 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3308 all the names associated with an address, because they do in Solaris.
3309 However, it seems that they do not in Linux for data that comes from the
3310 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3311 I found this out when I moved to a new Linux workstation and tried to run
3312 the Exim test suite.
3314 To get round this problem I have changed the code so that it now does its
3315 own call to the DNS to look up PTR records when searching for a host name.
3316 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3317 addresses that are only in /etc/hosts are still found.
3319 This behaviour is, however, controlled by an option called host_lookup_
3320 order, which defaults to "bydns:byaddr". If people want to use the other
3321 order, or indeed, just use one or the other means of lookup, they can
3322 specify it in this variable.
3324 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3325 some operating systems, this comes back from gethostbyaddr() as an empty
3326 string, and this is what Exim used to test for. However, it seems that in
3327 other systems, "." is yielded. Exim now tests for this case too.
3329 7. The values of check_spool_space and check_log_space are now held internally
3330 as a number of kilobytes instead of an absolute number of bytes. If a
3331 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3332 kilobyte. This means that much larger values can be stored.
3334 8. Exim monitor: an attempt to get the action menu when not actually pointing
3335 at a message produces an empty menu entitled "No message selected". This
3336 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3337 no entries in it ("Shell widget menu has zero width and/or height"). So I
3338 have added a single, blank menu entry in this case.
3340 9. Added ${quote_local_part.
3342 10. MIME decoding is now applied to the contents of Subject: header lines when
3345 11. Now that a reference to $sender_host_address automatically causes a reverse
3346 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3347 host lookup before query-style lookups in lists that might use this
3348 variable. This has therefore been abolished, and the "net-" prefix is no
3349 longer necessary for query-style lookups.
3351 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3352 have been a typo for LFLAGS, so it has been changed.
3354 13. The install script calls Exim with "-C /dev/null" in order to find the
3355 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3356 to be output. Howeve, since Exim outputs its version number before the
3357 error, it didn't break the script. It just looked ugly. I fixed this by
3358 always allowing "-C /dev/null" if the caller is root.
3360 14. Ignore overlarge ACL variable number when reading spool file - insurance
3361 against a later release with more variables having written the file.
3363 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3364 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3365 this, as well as the form without the "IPv6" on the front.
3367 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3368 OS/Makefile-Darwin file.
3370 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3371 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3373 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3374 variable. However, some OS use ints for inodes. I've added cast to long int
3375 to get rid of the compiler warning.
3377 19. I had forgotten to lock out "/../" in configuration file names when
3378 ALT_CONFIG_PREFIX was set.
3380 20. Routers used for verification do not need to specify transports. However,
3381 if such a router generated a host list, and callout was configured, Exim
3382 crashed, because it could not find a port number from the (non-existent)
3383 transport. It now assumes port 25 in this circumstance.
3385 21. Added the -t option to exigrep.
3387 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3388 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3389 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3390 them is set, LOOKUP_LSEARCH is forced.
3392 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3394 24. Added sender and host information to the "rejected by local_scan()" log
3395 line; previously there was no indication of these.
3397 25. Added .include_if_exists.
3399 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3400 Linux. It turns out that not all file systems support this. Apparently some
3401 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3402 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3403 not supported on the file descriptor, is now ignored when Exim is trying to
3404 sync a directory. This applies only to Linux.
3406 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3408 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3409 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3412 29. Added support for maildirsize files from supplied patch (modified a bit).
3414 30. The use of :fail: followed by an empty string could lead Exim to respond to
3415 sender verification failures with (e.g.):
3417 550 Verification failed for <xxx>
3418 550 Sender verify failed
3420 where the first response line was missing the '-' that indicates it is not
3421 the final line of the response.
3423 31. The loop for finding the name of the user that called Exim had a hardwired
3424 limit of 10; it now uses the value of finduser_retries, which is used for
3425 all other user lookups.
3427 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3430 33. Exim was neglecting to zero errno before one call of strtol() when
3431 expanding a string and expecting an integer value. On some systems this
3432 resulted in spurious "integer overflow" errors. Also, it was casting the
3433 result into an int without checking.
3435 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3436 did not work. The code looks as if it has *never* worked, though it appears
3437 to have been documented since at least releast 1.62. I have made it work.
3439 35. The "timeout_DNS" error in retry rules, also documented since at least
3440 1.62, also never worked. As it isn't clear exactly what this means, and
3441 clearly it isn't a major issue, I have abolished the feature by treating it
3442 as "timeout", and writing a warning to the main and panic logs.
3444 36. The display of retry rules for -brt wasn't always showing the error code
3447 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3448 timeout_connect_A, timeout_connect_MX.
3450 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3451 to the empty sender.
3453 39. The daemon was not analysing the content of -oX till after it had closed
3454 stderr and disconnected from the controlling terminal. This meant that any
3455 syntax errors were only noted on the panic log, and the return code from
3456 the command was 0. By re-arranging the code a little, I've made the
3457 decoding happen first, so such errors now appear on stderr, and the return
3458 code is 1. However, the actual setting up of the sockets still happens in
3459 the disconnected process, so errors there are still only recorded on the
3462 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3463 connections (as happens on some IP stacks) was logged at start up time as
3464 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3465 it from "IPv6 and IPv4", which means that two separate sockets are being
3468 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3469 says whether AF_INET or AF_INET6 was passed as an argument.
3471 42. Exiwhat output was messed up when time zones were included in log
3474 43. Exiwhat now gives more information about the daemon's listening ports,
3475 and whether -tls-on-connect was used.
3477 44. The "port" option of the smtp transport is now expanded.
3479 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3480 silently ignored. Now an error message is written to the main and panic
3483 46. There's a new ACL modifier called "logwrite" which writes to a log file
3484 as soon as it is encountered.
3486 47. Added $local_user_uid and $local_user_gid at routing time.
3488 48. Exim crashed when trying to verify a sender address that was being
3491 49. Exim was recognizing only a space character after ".include". It now also
3492 recognizes a tab character.
3494 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3495 extracting the relevant information from the specification. The man page no
3496 longer contains scrambled data for the -d option, and I've added a section
3497 at the front about calling Exim under different names.
3499 51. Added "extra_headers" argument to the "mail" command in filter files.
3501 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3504 53. Installed eximstats 1.29.
3506 54. Added transport_filter_timeout as a generic transport option.
3508 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3509 Cc: header lines. This was required by RFC 822, but it not required by RFC
3512 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3513 incoming messages that did not have them. Now it does so only if the
3514 message originates locally, that is, if there is no associated remote host
3515 address. When Resent- header lines are present, this applies to the Resent-
3516 lines rather than the non-Resent- lines.
3518 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3519 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3521 58. Messages for configuration errors now include the name of the main
3522 configuration file - useful now that there may be more than one file in a
3523 list (.included file names were always shown).
3525 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3526 for those rare installations that do not start the daemon as root or run it
3527 setuid root. I've cut out the call to initgroups() if the daemon is not
3530 60. The Exim user and group can now be bound into the binary as text strings
3531 that are looked up at the start of Exim's processing.
3533 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3535 62. Added $mailstore_basename variable.
3537 63. Installed patch to sieve.c from Michael Haardt.
3539 64. When Exim failed to open the panic log after failing to open the main log,
3540 the original message it was trying to log was written to stderr and debug
3541 output, but if they were not available (the usual case in production), it
3542 was lost. Now it is written to syslog before the two lines that record the
3543 failures to open the logs.
3545 65. Users' Exim filters run in subprocesses under the user's uid. It is
3546 possible for a "deliver" command or an alias in a "personal" command to
3547 provoke an address rewrite. If logging of address rewriting is configured,
3548 this fails because the process is not running as root or exim. There may be
3549 a better way of dealing with this, but for the moment (because 4.30 needs
3550 to be released), I have disabled address rewrite logging when running a
3551 filter in a non-root, non-exim process.
3557 1. The buildconfig auxiliary program wasn't quoting the value set for
3558 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3559 not defined. This bug was masked in 4.22 by the effect that was fixed in
3562 2. Some messages that were rejected after a message id was allocated were
3563 shown as "incomplete" by exigrep. It no longer does this for messages that
3564 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3566 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3567 have allow_domain_literals set, the ID did not get logged in the <= line.
3568 Domain literals are now always recognized in Message-ID: header lines.
3570 4. The first argument for a ${extract expansion item is the key name or field
3571 number. Leading and trailing spaces in this item were not being ignored,
3572 causing some misleading effects.
3574 5. When deliver_drop_privilege was set, single queue runner processes started
3575 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3576 same command in the process it spins off) were not dropping privilege.
3578 6. When the daemon running as "exim" started a queue runner, it always
3579 re-executed Exim in the spun-off process. This is a waste of effort when
3580 deliver_drop_privilege is set. The new process now just calls the
3581 queue-runner function directly.
3587 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3590 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3591 ignored. Though the use of -oP was forcing the writing of a pid file, it
3592 was always written to the default place.
3594 3. If the message "no IP address found for host xxxx" is generated during
3595 incoming verification, it is now followed by identification of the incoming
3596 connection (so you can more easily find what provoked it).
3598 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3600 5. Added some features to "harden" Exim a bit more against certain attacks:
3602 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3603 be put in Local/Makefile. This is like the never_users runtime option,
3604 but it cannot be overridden. The default setting is "root".
3606 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3607 prefix string with which any file named in a -C command line option
3610 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3611 is retained for -C and -D only if the caller of Exim is root. Without
3612 it, the exim user may also use -C and -D and retain privilege.
3614 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3615 command line option is disabled.
3617 6. Macro names set by the -D option must start with an upper case letter, just
3618 like macro names defined in the configuration file.
3620 7. Added "dereference=" facility to LDAP.
3622 8. Two instances of the typo "uknown" in the source files are fixed.
3624 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3625 the Configure-Makefile script screwed up while processing it.
3627 10. Incorporated PCRE 4.4.
3629 11. The SMTP synchronization check was not operating right at the start of an
3630 SMTP session. For example, it could not catch a HELO sent before the client
3631 waited for the greeting. There is now a check for outstanding input at the
3632 point when the greeting is written. Because of the duplex, asynchronous
3633 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3634 way, but not yet received, when the check is performed.
3636 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3637 on TCP/IP sockets, because this apparently causes some broken clients to
3640 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3641 unchanged) from the Cygwin maintainer.
3643 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3644 was supported instead of testing for mbx. Effectively a typo.
3646 15. The spa authenticator server code was not checking that the input it
3647 received was valid base64.
3649 16. The debug output line for the "set" modifier in ACLs was not showing the
3650 name of the variable that was being set.
3652 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3654 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3655 lookup on its own. Something else was needed to trigger the lookup. For
3656 example, a match in host_lookup or the need for a host name in a host list.
3657 Now, if $sender_host_name is referenced and the host name has not yet been
3658 looked up, a lookup is performed. If the lookup fails, the variable remains
3659 empty, and $host_lookup_failed is set to "1".
3661 19. Added "eqi" as a case-independent comparison operator.
3663 20. The saslauthd authentication condition could segfault if neither service
3664 nor realm was specified.
3666 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3667 error message that was logged was misleading, and incoming SMTP
3668 connections were dropped. The message is now more accurate, and temporary
3669 errors are given to SMTP connections.
3671 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3672 (except RSET) with a 421 error, until QUIT is received. However, it was
3673 failing to send a response to QUIT.
3675 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3676 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3677 ACL is now run after the helo_try_verify_hosts code.
3679 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3680 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3681 case-independent, so other case variants are also recognized). Apparently
3682 some systems use these upper case variants.
3684 25. If more than two messages were waiting for the same host, and a transport
3685 filter was specified for the transport, Exim sent two messages over the
3686 same TCP/IP connection, and then failed with "socket operation on non-
3687 socket" when it tried to send the third.
3689 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3691 27. The extern definition of crypt16() in expand.c was not being excluded when
3692 the OS had its own crypt16() function.
3694 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3695 as a preferred synonym for return_size_limit, both as an option and as an
3698 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3700 30. Changed the default configuration ACL to relax the local part checking rule
3701 for addresses that are not in any local domains. For these addresses,
3702 slashes and pipe symbols are allowed within local parts, but the sequence
3703 /../ is explicitly forbidden.
3705 31. SPA server authentication was not clearing the challenge buffer before
3708 32. log_message in a "warn" ACL statement was writing to the reject log as
3709 well as to the main log, which contradicts the documentation and doesn't
3710 seem right (because no rejection is happening). So I have stopped it.
3712 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3713 However, I am unable to do any testing of this.
3715 34. Fixed an infelicity in the appendfile transport. When checking directories
3716 for a mailbox, to see if any needed to be created, it was accidentally
3717 using path names with one or more superfluous leading slashes; tracing
3718 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3720 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3721 contents are added to the log line that is written for every discarded
3722 recipient. (Previously a log_message setting was ignored.)
3724 36. The ${quote: operator now quotes the string if it is empty.
3726 37. The install script runs exim in order to find its version number. If for
3727 some reason other than non-existence or emptiness, which it checks, it
3728 could not run './exim', it was installing it with an empty version number,
3729 i.e. as "exim-". This error state is now caught, and the installation is
3732 38. An argument was missing from the function that creates an error message
3733 when Exim fails to connect to the socket for saslauthd authentication.
3734 This could cause Exim to crash, or give a corrupted message.
3736 39. Added isip, isip4, and isip6 to ${if conditions.
3738 40. The ACL variables $acl_xx are now saved with the message, and can be
3739 accessed later in routers, transports, and filters.
3741 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3742 strings in the file are not string-expanded.
3744 42. If a MAIL command specified a SIZE value that was too large to fit into an
3745 int variable, the check against message_size_limit failed. Such values are
3746 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3747 day this will have to be increased, but I don't think I want to be around
3748 when emails are that large.
3755 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3756 iconv() is not standard in FreeBSD.
3758 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3759 IPv6 enabled. The observed symptom was a segmentation fault on return from
3760 the function os_common_find_running_interfaces() in src/os.c.
3762 3. In the check_special_case() function in daemon.c I had used "errno" as an
3763 argument name, which causes warnings on some systems. This was basically a
3764 typo, since it was named "eno" in the comments!
3766 4. The code that waits for the clock to tick (at a resolution of some fraction
3767 of a second) so as to ensure message-id uniqueness was always waiting for
3768 at least one whole tick, when it could have waited for less. [This is
3769 almost certainly not relevant at current processor speeds, where it is
3770 unlikely to ever wait at all. But we try to future-proof.]
3772 5. The function that sleeps for a time interval that includes fractions of a
3773 second contained a race. It did not block SIGALRM between setting the
3774 timer, and suspending (a couple of lines later). If the interval was short
3775 and the sigsuspend() was delayed until after it had expired, the suspension
3776 never ended. On busy systems this could lead to processes getting stuck for
3779 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3780 process, before it forks any delivery processes. The open lookup caching
3781 mechanism meant that the open file or database connection was passed into
3782 the delivery process. The problem was that delivery processes always tidy
3783 up cached lookup data. This could cause a problem for the next delivery
3784 process started by the queue runner, because the external queue runner
3785 process does not know about the closure. So the next delivery process
3786 still has data in the lookup cache. In the case of a file lookup, there was
3787 no problem because closing a file descriptor in a subprocess doesn't affect
3788 the parent. However, if the lookup was caching a connection to a database,
3789 the connection was closed, and the second delivery process was likely to
3790 see errors such as "PGSQL: query failed: server closed the connection
3791 unexpectedly". The problem has been fixed by closing all cached lookups
3792 in a queue runner before running a delivery process.
3794 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3795 seem to have the "const" qualifier which it has on other OS. I've
3798 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3799 *of the same type* (client or server) with the same public name that an
3800 error should be diagnosed.
3802 9. When Exim looked up a host name for an IP address, but failed to find the
3803 original IP address when looking up the host name (a safety check), it
3804 output the message "<ip address> does not match any IP for NULL", which was
3805 confusing, to say the least. The bug was that the host name should have
3806 appeared instead of "NULL".
3808 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3809 user that is built into the binary, and the -C or -D options is used, root
3810 privilege is dropped before the configuration file is read. In addition,
3811 logging is switched to stderr instead of the normal log files. If the
3812 configuration then re-defines the Exim user, the unprivileged environment
3813 is probably not what is expected, so Exim logs a panic warning message (but
3816 However, if deliver_drop_privilege is set, the unprivileged state may well
3817 be exactly what is intended, so the warning has been cut out in that case,
3818 and Exim is allowed to try to write to its normal log files.
3824 1. smtp_return_error_details was not giving details for temporary sender
3825 or receiver verification errors.
3827 2. Diagnose a configuration error if two authenticators have the same public
3830 3. Exim used not to create the message log file for a message until the first
3831 delivery attempt. This could be confusing when incoming messages were held
3832 for policy or load reasons. The message log file is now created at the time
3833 the message is received, and an initial "Received" line is written to it.
3835 4. The automatically generated man page for command line options had a minor
3836 bug that caused no ill effects; however, a more serious problem was that
3837 the procedure for building the man page automatically didn't always
3838 operate. Consequently, release 4.20 contains an out-of-date version. This
3839 shouldn't happen again.
3841 5. When building Exim with embedded Perl support, the script that builds the
3842 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3843 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3845 6. The freeze_tell option was not being used for messages that were frozen on
3846 arrival, either by an ACL or by local_scan().
3848 7. Added the smtp_incomplete_transaction log selector.
3850 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3851 was accepting AUTH without a new EHLO.
3853 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3854 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3857 10. Logging of TCP/IP connections (when configured) now happens in the main
3858 daemon process instead of the child process, so that the TCP/IP connection
3859 count is more accurate (but it can never be perfect).
3861 11. The use of "drop" in a nested ACL was not being handled correctly in the
3862 outer ACL. Now, if condition failure induced by the nested "drop" causes
3863 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3864 or "require"), the connection is dropped.
3866 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3867 that yield "discard" can now be used with an "accept" or a "discard" verb,
3868 but an error is generated for any others (because I can't see a useful way
3869 to define what should happen).
3871 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3872 are now processed in the same way as lines in the Exim configuration file.
3873 In particular, continuation lines are supported.
3875 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3877 15. Added -ti meaning -t -i.
3879 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3880 domains, and warn by logging if others are found.
3882 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3883 returned from GIFCONF when Exim is trying to find the list of interfaces on
3884 a host. The code in os.c has been modified to copy each ifreq to an aligned
3885 structure in all cases.
3887 Also, in some cases, the returned ifreq's were being copied to a 'struct
3888 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3889 means the last couple of bytes of an IPv6 address could be chopped if the
3890 ifreq contained only a normal sockaddr (14 bytes storage).
3892 18. Named domain lists were not supported in the hosts_treat_as_local option.
3893 An entry such as +xxxx was not recognized, and was treated as a literal
3896 19. Ensure that header lines added by a DATA ACL are included in the reject log
3897 if the ACL subsequently rejects the message.
3899 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3900 MD5 (which is deprecated).
3902 21. When testing a filter file using -bf, Exim was writing a message when it
3903 took the sender from a "From " line in the message, but it was not doing so
3904 when it took $return_path from a Return-Path: header line. It now does.
3906 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3907 with a valid header line field name (a series of printing characters
3908 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3910 23. Changed "disc" in the source to "disk" to conform to the documentation and
3911 the book and for uniformity.
3913 24. Ignore Sendmail's -Ooption=value command line item.
3915 25. When execve() failed while trying to run a command in a pipe transport,
3916 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3917 could be confused with a return value of 69 from the command itself. This
3918 has been changed to 127, the value the shell returns if it is asked to run
3919 a non-existent command. The wording for the related log line suggests a
3920 non-existent command as the problem.
3922 26. If received_header_text expands to an empty string, do not add a Received:
3923 header line to the message. (Well, it adds a token one on the spool, but
3924 marks it "old" so that it doesn't get used or transmitted.)
3926 27. Installed eximstats 1.28 (addition of -nt option).
3928 28. There was no check for failure on the call to getsockname() in the daemon
3929 code. This can fail if there is a shortage of resources on the system, with
3930 ENOMEM, for example. A temporary error is now given on failure.
3932 29. Contrary to the C standard, it seems that in some environments, the
3933 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3934 program. Exim now does this explicitly; it affects the formatting of
3935 timestamps using strftime().
3937 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3938 complaints. I've now initialized all the variables, to avoid this.
3940 32. Header lines added by a system filter were not being "seen" during
3941 transport-time rewrites.
3943 33. The info_callback() function passed to OpenSSL is set up with type void
3944 (*)(SSL *, int, int), as described somewhere. However, when calling the
3945 function (actually a macro) that sets it up, the type void(*)() is
3946 expected. I've put in a cast to prevent warnings from picky compilers.
3948 34. If a DNS black list lookup found a CNAME record, but there were no A
3949 records associated with the domain it pointed at, Exim crashed.
3951 35. If a DNS black list lookup returned more than one A record, Exim ignored
3952 all but the first. It now scans all returned addresses if a particular IP
3953 value is being sought. In this situation, the contents of the
3954 $dnslist_value variable are a list of all the addresses, separated by a
3957 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3958 to accept a host name and all its aliases if the forward lookup for any of
3959 them yielded the IP address of the incoming connection. Now it accepts only
3960 those names whose forward lookup yields the correct IP address. Any other
3961 names are discarded. This closes a loophole whereby a rogue DNS
3962 administrator could create reverse DNS records to break through a
3963 wildcarded host restriction in an ACL.
3965 37. If a user filter or a system filter that ran in a subprocess used any of
3966 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3967 the wrong values were passed to the pipe command ($thisaddress had the
3968 value of $0, $0 had the value of $1, etc). This bug was introduced by
3969 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3971 38. Improved the line breaking for long SMTP error messages from ACLs.
3972 Previously, if there was no break point between 40 and 75 characters, Exim
3973 left the rest of the message alone. Two changes have been made: (a) I've
3974 reduced the minimum length to 35 characters; (b) if it can't find a break
3975 point between 35 and 75 characters, it looks ahead and uses the first one
3976 that it finds. This may give the occasional overlong line, but at least the
3977 remaining text gets split now.
3979 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3980 file descriptors might be low, and that setting 1000 would always raise it.
3981 It turns out that in some environments, the limit is already over 1000 and
3982 that lowering it causes trouble. So now Exim takes care not to decrease it.
3984 40. When delivering a message, the value of $return_path is set to $sender_
3985 address at the start of routing (routers may change the value). By an
3986 oversight, this default was not being set up when an address was tested by
3987 -bt or -bv, which affected the outcome if any router or filter referred to
3990 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
3991 log only when "message" or "log_message" are set. However, if one of the
3992 conditions was an address verification, or a call to a nested ACL, the
3993 messages generated by the underlying test were being passed through. This
3994 no longer happens. The underlying message is available in $acl_verify_
3995 message for both "message" and "log_message" expansions, so it can be
3996 passed through if needed.
3998 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3999 new expansion $bh_ to give the encoded byte string without charset
4000 translation. Translation happens only if iconv() is available; HAVE_ICONV
4001 indicates this at build time. HEADERS_CHARSET gives the charset to
4002 translate to; headers_charset can change it in the configuration, and
4003 "headers charset" can change it in an individual filter file.
4005 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4006 that creates RFC 2047 encoded "words" labels them as that charset instead
4007 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4008 expansion operator; (ii) when Exim creates a From: line for a local
4009 message; (iii) when a header line is rewritten to include a "phrase" part.
4011 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4012 buggy, causing it to skip the first lines of messages whose message ID
4013 ended in 'D'. This would not have bitten before Exim release 4.14, because
4014 message IDs were unlikely to end in 'D' before then. The effect was to have
4015 incorrect size information for certain domains.
4017 45. #include "config.h" was missing at the start of the crypt16.c module. This
4018 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4021 46. If there was a timeout during a "random" callout check, Exim treated it as
4022 a failure of the random address, and carried on sending RSET and the real
4023 address. If the delay was just some slowness somewhere, the response to the
4024 original RCPT would be taken as a response to RSET and so on, causing
4025 mayhem of various kinds.
4027 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4028 when I implemented it. It didn't allow for the fact that some option values
4029 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4030 the right test for positive values.
4032 48. Domain names in DNS records are case-independent. Exim always looks them up
4033 in lower case. Some resolvers return domain names in exactly the case they
4034 appear in the zone file, that is, they may contain uppercase letters. Not
4035 all resolvers do this - some return always lower case. Exim was treating a
4036 change of case by a resolver as a change of domain, similar to a widening
4037 of a domain abbreviation. This triggered its re-routing code and so it was
4038 trying to route what was effectively the same domain again. This normally
4039 caused routing to fail (because the router wouldn't handle the domain
4040 twice). Now Exim checks for this case specially, and just changes the
4041 casing of the domain that it ultimately uses when it transmits the message
4044 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4047 50. If a filter generated a file delivery with a non-absolute name (possible if
4048 no home directory exists for the router), the forbid_file option was not
4051 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4052 the existing equality matching.
4054 52. Exim was using ints instead of ino_t variables in some places where it was
4055 dealing with inode numbers.
4057 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4058 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4059 called TMPDIR, and if it finds it is different, it changes its value.
4061 54. The smtp_printf() function is now made available to local_scan() so
4062 additional output lines can be written before returning. There is also an
4063 smtp_fflush() function to enable the detection of a dropped connection.
4064 The variables smtp_input and smtp_batched_input are exported to
4067 55. Changed the default runtime configuration: the message "Unknown user"
4068 has been removed from the ACL, and instead placed on the localuser router,
4069 using the cannot_route_message feature. This means that any verification
4070 failures that generate their own messages won't get overridden. Similarly,
4071 the "Unrouteable address" message that was in the ACL for unverifiable
4072 relay addresses has also been removed.
4074 56. Added hosts_avoid_esmtp to the smtp transport.
4076 57. The exicyclog script was not checking for the esoteric option
4077 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4078 will work only if exicyclog is run under the appropriate euid.
4080 58. Following a discussion on the list, the rules by which Exim recognises line
4081 endings on incoming messages have been changed. The -dropcr and drop_cr
4082 options are now no-ops, retained only for backwards compatibility. The
4083 following line terminators are recognized: LF CRLF CR. However, special
4084 processing applies to CR:
4086 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4087 nor a local message in the state where . is a terminator.
4089 (ii) If a bare CR is encountered in a header line, an extra space is added
4090 after the line terminator so as not to end the header. The reasoning
4091 behind this is that bare CRs in header lines are most likely either
4092 to be mistakes, or people trying to play silly games.
4094 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4095 was being incorrectly given as 18 bytes larger than it should have been.
4096 This is a VOB (very old bug).
4098 60. This may never have affected anything current, but just in case it has:
4099 When the local host is found other than at the start of a list of hosts,
4100 the local host, those with the same MX, and any that follow, are discarded.
4101 When the list in question was part of a longer list of hosts, the following
4102 hosts (not currently being processed) were also being discarded. This no
4103 longer happens. I'm not sure if this situation could ever has previously
4106 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4109 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4110 header. This is something that is recommended in a new Internet Draft, and
4111 is something that is documented as being done by Sendmail. There are two
4112 possible values. For messages generated by the autoreply transport, Exim
4115 Auto-Submitted: auto-replied
4117 whereas for all other generated messages (e.g. bounces) it adds
4119 Auto-Submitted: auto-generated
4121 63. The "personal" condition in filters now includes a test for the
4122 Auto-Submitted: header. If it contains the string "auto-" the message it
4123 not considered personal.
4125 64. Added rcpt_include_affixes as a generic transport option.
4127 65. Added queue_only_override (default true).
4129 66. Added the syslog_duplication option.
4131 67. If what should have been the first header line of a message consisted of
4132 a space followed by a colon, Exim was mis-interpreting it as a header line.
4133 It isn't of course - it is syntactically invalid and should therefore be
4134 treated as the start of the message body. The misbehaviour could have
4135 caused a number of strange effects, including loss of data in subsequent
4136 header lines, and spool format errors.
4138 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4139 client host had authenticated. This control can now be exercised by an ACL
4140 for more flexibility.
4142 69. By default, callouts do not happen when testing with -bh. There is now a
4143 variant, -bhc, which does actually run the callout code, including
4144 consulting and updating the callout cache.
4146 70. Added support for saslauthd authentication, courtesy of Alexander
4149 71. If statvfs() failed on the spool or log directories while checking their
4150 size for availability, Exim confusingly gave the error "space shortage".
4151 Furthermore, in debugging mode it crashed with a floating point exception.
4152 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4153 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4154 serious problem, Exim now writes to the main and panic logs when this
4155 happens, with details of the failure. It then refuses to accept the
4156 incoming message, giving the message "spool directory problem" or "log
4157 directory problem" with a 421 code for SMTP messages.
4159 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4160 0, 1, and 2 exist, because some OS complain for execs without them (see
4161 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4162 descriptors. However, the code omitted to check that the open succeeded,
4163 causing mysterious errors if for some reason the permissions on /dev/null
4164 got screwed. Now Exim writes a message to the main and panic logs, and
4165 bombs out if it can't open /dev/null.
4167 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4168 interact so that it is all more flexible. It is supposed to remain
4169 backwards compatible. Also added extra_local_interfaces.
4171 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4172 to bomb out with an assertion failure - to the client this appears as a
4173 connection drop. This problem occurs in the part of the code that was taken
4174 from the Samba project. Fortunately, the assertion is in a very simple
4175 function, so I have fixed this by reproducing the function inline in the
4176 one place where it is called, and arranging for authentication to fail
4177 instead of killing the process with assert().
4179 75. The SPA client code was not working when the server requested OEM rather
4180 than Unicode encoding.
4182 76. Added code to make require_files with a specific uid setting more usable in
4183 the case where statting the file as root fails - usually a non-root-mounted
4184 NFS file system. When this happens and the failure is EACCES, Exim now
4185 forks a subprocess and does the per-uid checking as the relevant uid.
4187 77. Added process_log_path.
4189 78. If log_file_path was not explicitly set, a setting of check_log_space or
4190 check_log_inodes was ignored.
4192 79. If a space check for the spool or log partitions fails, the incident is now
4193 logged. Of course, in the latter case the data may get lost...
4195 80. Added the %p formatting code to string_format() so that it can be used to
4196 print addresses in debug_print(). Adjusted all the address printing in the
4197 debugging in store.c to use %p rather than %d.
4199 81. There was a concern that a line of code in smtp_in.c could overflow a
4200 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4201 initially expressed, the concern was not well-founded, because trailing
4202 spaces are removed early. However, if the trailing spaces were followed by
4203 a NULL, they did not get removed, so the overflow was possible. Two fixes
4206 (a) I re-wrote the offending code in a cleaner fashion.
4207 (b) If an incoming SMTP command contains a NULL character, it is rejected
4210 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4211 runs initgroups(), so that if the Exim user is in any additional groups,
4212 they will be used during message reception.
4218 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob