1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.402 2006/10/03 10:25:55 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.
98 SC/01 Use a glob alias rather than an array ref in eximstats generated
99 parser. This improves both readability and performance.
101 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
102 Don't display local sender or destination tables in eximstats unless
103 there is data to show.
104 Added average volumes into the eximstats top table text output.
106 SC/03 Collect data on the number of addresses (recipients) as well
107 as the number of messages in eximstats.
109 TF/01 Correct an error in the documentation for the redirect router. Exim
110 does (usually) call initgroups() when daemonizing.
112 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
113 with consistent privilege compared to when running as a daemon.
115 TF/03 Note in the spec that $authenticated_id is not set for local
116 submissions from trusted users.
118 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
119 Thanks to Dean Brooks <dean@iglou.com> for the patch.
121 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
122 by adding some example configuration directives to the default
123 configuration file. A little bit of work is required to uncomment the
124 directives and define how usernames and passwords are checked, but
125 there is now a framework to start from.
127 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
128 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
129 without this. I don't know how relevant this is to other LDAP libraries.
131 PH/02 Add the verb name to the "unknown ACL verb" error.
133 PH/03 Magnus Holmgren's patch for filter_prepend_home.
135 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
137 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
138 directory not expanded when it should be if an expanded home directory
139 was set for the address (which is overridden by the transport).
141 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
144 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
145 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
146 because it is too late at that time, and has no effect.
148 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
149 security issue with \' (bugzilla #107). I could not use the
150 PQescapeStringConn() function, because it needs a PGconn value as one of
153 PH/08 When testing addresses using -bt, indicate those final addresses that
154 are duplicates that would not cause an additional delivery. At least one
155 person was confused, thinking that -bt output corresponded to deliveries.
156 (Suppressing duplicates isn't a good idea as you lose the information
157 about possibly different redirections that led to the duplicates.)
159 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
160 systems where poll() doesn't work, in particular OS X.
162 PH/10 Added more information to debugging output for retry time not reached.
164 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
165 operations in malware.c.
167 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
170 PH/13 If write_rejectlog was set false when logging was sent to syslog with
171 syslog_duplication set false, log lines that would normally be written
172 both the the main log and to the reject log were not written to syslog at
175 PH/14 In the default configuration, change the use of "message" in ACL warn
176 statements to "add_header".
178 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
179 not followed by a command (e.g. "seen endif").
181 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
182 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
185 PH/17 Added extra conditions to the default value of delay_warning_condition
189 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
190 { match{$h_precedence:}{(?i)bulk|list|junk} } \
191 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
194 The Auto-Submitted: and various List- headers are standardised, whereas I
195 don't think Precedence: ever was.
197 PH/18 Refactored debugging code in route_finduser() to show more information,
198 in particular, the error code if getpwnam() issues one.
200 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
201 This is apparently needed in addition to the PH/07 change above to avoid
202 any possible encoding problems.
204 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
205 but not after initializing Perl.
207 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
208 output them only if debugging. By default they are written stderr,
209 apparently, which is not desirable.
211 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
214 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
217 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
219 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
220 authenticated or an ident call has been made. Suppress the default
221 values for $authenticated_id and $authenticated_sender (but permit -oMai
222 and -oMas) when testing with -bh.
224 PH/24 Re-jigged the order of the tests in the default configuration so that the
225 tests for valid domains and recipients precede the DNS black list and CSA
226 tests, on the grounds that those ones are more expensive.
228 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
229 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
230 This bug exists in every version of Exim that I still have, right back to
233 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
234 However, an attempt to turn on case-sensitivity in a regex key by
235 including (?-i) didn't work because the subject string was already
236 lowercased, and the effects were non-intuitive. It turns out that a
237 one-line patch can be used to allow (?-i) to work as expected.
243 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
244 other effects) broke the use of negated acl sub-conditions.
246 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
249 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
250 "Deny" causes Exim to reject the incoming connection with a 554 error.
251 Unfortunately, if there is a major crisis, such as a disk failure,
252 tcp-wrappers gives "deny", whereas what one would like would be some
253 kind of temporary error. A kludge has been added to help with this.
254 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
255 554 error is used if errno is still zero or contains ENOENT (which occurs
256 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
259 PH/03 Add -lutil to the default FreeBSD LIBS setting.
261 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
262 errors. Otherwise a message that provokes a temporary error (when other
263 messages do not) can cause a whole host to time out.
265 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
266 addresses were routed directly to files or pipes from a redirect router.
267 File deliveries just didn't batch; pipe deliveries might have suffered
270 PH/06 A failure to get a lock for a hints database would erroneously always say
271 "Failed to get write lock", even when it was really a read lock.
273 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
274 of 0600. This has been changed to use the value of the lockfile_mode
275 option (which defaults to 0600).
277 PH/08 Applied small patch from the Sieve maintainer.
279 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
280 folder from quota calculations, a direct delivery into this folder messed
281 up the contents of the maildirsize file. This was because the regex was
282 used only to exclude .Trash (or whatever) when the size of the mailbox
283 was calculated. There was no check that a delivery was happening into an
284 excluded directory. This bug has been fixed by ignoring all quota
285 processing for deliveries into excluded directories.
287 PH/10 Added the maildirfolder_create_regex option to appendfile.
293 PH/01 The code for finding all the local interface addresses on a FreeBSD
294 system running IPv6 was broken. This may well have applied to all BSD
295 systems, as well as to others that have similar system calls. The broken
296 code found IPv4 interfaces correctly, but gave incorrect values for the
297 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
298 that it would not match correctly against @[] and not recognize the IPv6
301 PH/02 The ipliteral router was not recognizing addresses of the form user@
302 [ipv6:....] because it didn't know about the "ipv6:" prefix.
304 PH/03 Added disable_ipv6.
306 PH/04 Changed $reply_address to use the raw form of the headers instead of the
307 decoded form, because it is most often used to construct To: headers
308 lines in autoreplies, and the decoded form may well be syntactically
309 invalid. However, $reply_address has leading white space removed, and all
310 newlines turned into spaces so that the autoreply transport does not
313 PH/05 If group was specified without a user on a router, and no group or user
314 was specified on a transport, the group from the router was ignored.
316 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
317 for visible compile-time settings that can be used to change these
318 numbers, for those that want even more. Backwards compatibility with old
319 spool files has been maintained. However, going back to a previous Exim
320 release will lost any variables that are in spool files.
322 PH/07 Two small changes when running in the test harness: increase delay when
323 passing a TCP/IP connection to a new process, in case the original
324 process has to generate a bounce, and remove special handling of
325 127.0.0.2 (sic), which is no longer necessary.
327 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
328 be the same on different OS.
330 PH/09 Moved a debug statement in filter processing to avoid a race problem when
333 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
334 whether --show-vars was specified or not
336 JJ/02 exipick: Added support for new ACL variable spool format introduced
339 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
340 syntactically invalid From: or Reply-to: line, and a filter used this to
341 generate an autoreply, and therefore failed to obtain an address for the
342 autoreply, Exim could try to deliver to a non-existent relative file
343 name, causing unrelated and misleading errors. What now happens is that
344 it logs this as a hard delivery error, but does not attempt to create a
347 PH/11 The exinext utility has a -C option for testing purposes, but although
348 the given file was scanned by exinext itself; it wasn't being passed on
349 when Exim was called.
351 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
352 an end-of-file indication when reading a command response.
354 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
355 compiled. In many other places in Exim, IPv6 addresses are always
356 recognized, so I have changed this. It also means that IPv4 domain
357 literals of the form [IPV4:n.n.n.n] are now always recognized.
359 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
360 used if the router is not running as root, for example, when verifying at
361 ACL time, or when using -bh. The debugging output from this situation was
362 non-existent - all you got was a failure to exec. I have made two
365 (a) Failures to set uid/gid, the current directory, or a process leader
366 in a subprocess such as that created by queryprogram now generate
367 suitable debugging ouput when -d is set.
369 (b) The queryprogram router detects when it is not running as root,
370 outputs suitable debugging information if -d is set, and then runs
371 the subprocess without attempting to change uid/gid.
373 PH/15 Minor change to Makefile for building test_host (undocumented testing
376 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
377 additional section of a DNS packet that returns MX or SRV records.
378 Instead, it always explicitly searches for A/AAAA records. This avoids
379 major problems that occur when a DNS server includes only records of one
380 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
381 fixed another bug: if SRV records were looked up and the corresponding
382 address records were *not* found in the additional section, the port
383 values from the SRV records were lost.
385 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
386 using the correct key (the original address) when searching the retry
387 rules in order to find which one to use for generating the retry hint.
389 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
390 adding the default one. Similarly, if it contains a Reply-To: header, the
391 errors_reply_to option, if set, is not used.
393 PH/19 When calculating a retry time, Exim used to measure the "time since
394 failure" by looking at the "first failed" field in the retry record. Now
395 it does not use this if it is later than than the arrival time of the
396 message. Instead it uses the arrival time. This makes for better
397 behaviour in cases where some deliveries succeed, thus re-setting the
398 "first failed" field. An example is a quota failure for a huge message
399 when small messages continue to be delivered. Without this change, the
400 "time since failure" will always be short, possible causing more frequent
401 delivery attempts for the huge message than are intended.
402 [Note: This change was subsequently modified - see PH/04 for 4.62.]
404 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
405 $1, $2, $3) because the numerical variables can be reset during some
406 expansion items (e.g. "match"), thereby losing the authentication data.
408 PH/21 Make -bV show the size of off_t variables so that the test suite can
409 decide whether to run tests for quotas > 2G.
411 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
412 mailbox_size, and mailbox_filecount in the appendfile transport. If a
413 filecount value is greater than 2G or if a quota value is greater than 2G
414 on a system where the size of off_t is not greater than 4, a panic error
417 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
418 never match. The debug and -bh output now contains an explicit error
419 message indicating a malformed IPv4 address or mask.
421 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
422 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
425 PH/25 Do not write to syslog when running in the test harness. The only
426 occasion when this arises is a failure to open the main or panic logs
427 (for which there is an explicit test).
429 PH/26 Added the /no_tell option to "control=freeze".
431 PH/27 If a host name lookup failed very early in a connection, for example, if
432 the IP address matched host_lookup and the reverse lookup yielded a name
433 that did not have a forward lookup, an error message of the form "no IP
434 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
435 could be logged. Now it outputs the IP address instead of "NULL".
437 PH/28 An enabling patch from MH: add new function child_open_exim2() which
438 allows the sender and the authenticated sender to be set when
439 submitting a message from within Exim. Since child_open_exim() is
440 documented for local_scan(), the new function should be too.
442 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
443 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
444 results in an empty string is now treated as unset.
446 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
448 PH/31 Added sender_verify_fail logging option.
450 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
451 needed by RFC 822 but not by RFC 2822 was commented out. I have now
452 tidied the source and removed it altogether.
454 PH/33 When a queue run was abandoned because the load average was too high, a
455 log line was always written; now it is written only if the queue_run log
456 selector is set. In addition, the log line for abandonment now contains
457 information about the queue run such as the pid. This is always present
458 in "start" and "stop" lines but was omitted from the "abandon" line.
460 PH/34 Omit spaces between a header name and the colon in the error message that
461 is given when verify = headers_syntax fails (if there are lots of them,
462 the message gets confusing).
464 PH/35 Change the default for dns_check_names_pattern to allow slashes within
465 names, as there are now some PTR records that contain slashes. This check
466 is only to protect against broken name servers that fall over on strange
467 characters, so the fact that it applies to all lookups doesn't matter.
469 PH/36 Now that the new test suite is complete, we can remove some of the
470 special code in Exim that was needed for the old test suite. For example,
471 sorting DNS records because real resolvers return them in an arbitrary
472 order. The new test suite's fake resolver always returns records in the
475 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
476 bounces) except when queue_only is set, to avoid logging races between
477 the different processes.
479 PH/38 Panic-die if .include specifies a non-absolute path.
481 PH/39 A tweak to the "H" retry rule from its user.
483 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
484 a label. They prevented compilation on older perls.
486 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
487 a warning to be raised on newish perls.
489 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
490 on queue. Changes to match documented behaviour of showing count of
491 messages matching specified criteria.
493 PH/40 Changed the default ident timeout from 30s to 5s.
495 PH/41 Added support for the use of login_cap features, on those BSD systems
496 that have them, for controlling the resources used by pipe deliveries.
498 PH/42 The content-scanning code uses fopen() to create files in which to put
499 message data. Previously it was not paying any attention to the mode of
500 the files. Exim runs with umask(0) because the rest of the code creates
501 files with open(), and sets the required mode explicitly. Thus, these
502 files were ending up world-writeable. This was not a big issue, because,
503 being within the spool directory, they were not world-accessible. I have
504 created a function called modefopen, which takes an additional mode
505 argument. It sets umask(777), creates the file, chmods it to the required
506 mode, then resets the umask. All the relevant calls to fopen() in the
507 content scanning code have been changed to use this function.
509 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
510 to 24 hours. This avoids potential overflow problems when processing G
511 and H retry rules. I suspect nobody ever tinkers with this value.
513 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
515 PH/45 When the plaintext authenticator is running as a client, the server's
516 challenges are checked to ensure they are valid base64 strings. By
517 default, the authentication attempt is cancelled if an invalid string is
518 received. Setting client_ignore_invalid_base64 true ignores these errors.
519 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
520 they are received. Thus, the responses can be made to depend on the
521 challenges. If an invalid string is ignored, an empty string is placed in
524 PH/46 Messages that are created by the autoreply transport now contains a
525 References: header, in accordance with RFCs 2822 and 3834.
527 PH/47 Added authenticated_sender_force to the smtp transport.
529 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
531 PH/49 Installed latest patch from the Sieve maintainer.
533 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
534 was also set, the appendfile transport was unnecessarily scanning a
535 directory of message files (e.g. for maildir delivery) to find the count
536 of files (along with the size), even though it did not need this
537 information. It now does the scan only if it needs to find either the
538 size of the count of files.
540 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
542 PH/52 Two bugs concerned with error handling when the smtp transport is
545 (i) Exim was not creating retry information for temporary errors given
546 for individual recipients after the DATA command when the smtp transport
547 was used in LMTP mode. This meant that they could be retried too
548 frequently, and not timed out correctly.
550 (ii) Exim was setting the flag that allows error details to be returned
551 for LMTP errors on RCPT commands, but not for LMTP errors for individual
552 recipients that were returned after the DATA command.
554 PH/53 This is related to PH/52, but is more general: for any failing address,
555 when detailed error information was permitted to be returned to the
556 sender, but the error was temporary, then after the final timeout, only
557 "retry timeout exceeded" was returned. Now it returns the full error as
558 well as "retry timeout exceeded".
560 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
561 do this, and (what is worse) MTAs that accept it.
563 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
564 will now be deprecated.
566 PH/56 New os.c-cygwin from the Cygwin maintainer.
568 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
569 formats (previously only available in exim formats via -bpr, -bpru,
570 and -bpra. Now also available in native and exiqgrep formats)
572 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
573 with very large, slow to parse queues
575 JJ/08 exipick: added ! as generic prefix to negate any criteria format
577 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
579 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
580 responses to authentication challenges, though it was showing the
581 challenges; (ii) I've removed the CR characters from the debug output for
584 PH/58 Allow for the insertion of a newline as well as a space when a string
585 is turned into more than one encoded-word during RFC 2047 encoding. The
586 Sieve code now uses this.
588 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
589 data_4xx, lost_connection, tls_required.
591 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
592 message was being sent as an SMTP response.
594 PH/61 Add -l and -k options to exicyclog.
596 PH/62 When verifying, if an address was redirected to one new address, so that
597 verification continued, and the new address failed or deferred after
598 having set something in $address_data, the value of $address_data was not
599 passed back to the ACL. This was different to the case when no
600 redirection occurred. The value is now passed back in both cases.
602 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
603 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
604 use login_cap.h, so on its own it isn't the distinguishing feature. The
605 new name refers directly to the setclassresources() function.
607 PH/65 Added configuration files for NetBSD3.
609 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
611 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
612 is preferred over IPv4.
614 PH/68 The bounce_return_message and bounce_return_body options were not being
615 honoured for bounces generated during the reception of non-SMTP messages.
616 In particular, this applied to messages rejected by the ACL. This bug has
617 been fixed. However, if bounce_return_message is true and bounce_return_
618 body is false, the headers that are returned for a non-SMTP message
619 include only those that have been read before the error was detected.
620 (In the case of an ACL rejection, they have all been read.)
622 PH/69 The HTML version of the specification is now built in a directory called
623 spec_html instead of spec.html, because the latter looks like a path with
624 a MIME-type, and this confuses some software.
626 PH/70 Catch two compiler warnings in sieve.c.
628 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
629 function verify_get_ident() calls ip_connect() to connect a socket, but
630 if the "connect()" function timed out, ip_connect() used to close the
631 socket. However, verify_get_ident() also closes the socket later, and in
632 between Exim writes to the log, which may get opened at this point. When
633 the socket was closed in ip_connect(), the log could get the same file
634 descriptor number as the socket. This naturally causes chaos. The fix is
635 not to close the socket in ip_connect(); the socket should be closed by
636 the function that creates it. There was only one place in the code where
637 this was missing, in the iplookup router, which I don't think anybody now
638 uses, but I've fixed it anyway.
640 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
641 well as to direct DNS lookups. Otherwise the handling of names in host
642 lists is inconsistent and therefore confusing.
648 PH/01 Two changes to the default runtime configuration:
650 (1) Move the checks for relay_from_hosts and authenticated clients from
651 after to before the (commented out) DNS black list checks.
653 (2) Add control=submission to the relay_from_hosts and authenticated
654 clients checks, on the grounds that messages accepted by these
655 statements are most likely to be submissions.
657 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
659 (1) Generate an error if the third argument for the ${prvs expansion is
662 (2) Treat a missing third argument of ${prvscheck as if it were an empty
665 (3) Reset the variables that are obtained from the first argument of
666 ${prvscheck and used in the second argument before leaving the code,
667 because their memory is reclaimed, so using them afterwards may do
670 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
671 one (it's much easier than Tom thought :-).
673 (5) Because of (4), we can now allow for the use of $prvscheck_result
674 inside the third argument.
676 PH/03 For some reason, the default setting of PATH when running a command from
677 a pipe transport was just "/usr/bin". I have changed it to
680 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
681 anything to be listed in the output from -bV.
683 PH/05 When a filter generated an autoreply, the entire To: header line was
684 quoted in the delivery log line, like this:
686 => >A.N.Other <ano@some.domain> <original@ddress> ...
688 This has been changed so that it extracts the operative address. There
689 may be more than one such address. If so, they are comma-separated, like
692 => >ano@some.domain,ona@other.domain <original@ddress> ...
694 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
695 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
696 not being looked up in the rDNS to get a host name, Exim was showing the
697 IP address twice in Received: lines, even though the IP addresses were
698 identical. For example:
700 Received: from [1.2.3.4] (helo=[1.2.3.4])
702 However, if the real host name was known, it was omitting the HELO data
703 if it matched the actual IP address. This has been tidied up so that it
704 doesn't show the same IP address twice.
706 PH/07 When both +timestamp and +memory debugging was on, the value given by
707 $tod_xxx expansions could be wrong, because the tod_stamp() function was
708 called by the debug printing, thereby overwriting the timestamp buffer.
709 Debugging no longer uses the tod_stamp() function when +timestamp is set.
711 PH/08 When the original message was included in an autoreply transport, it
712 always said "this is a copy of the message, including all the headers",
713 even if body_only or headers_only was set. It now gives an appropriate
716 PH/09 Applied a patch from the Sieve maintainer which:
718 o fixes some comments
719 o adds the (disabled) notify extension core
720 o adds some debug output for the result of if/elsif tests
721 o points to the current vacation draft in the documentation
722 and documents the missing references header update
726 o fixes a bug in processing the envelope test (when testing
727 multiple envelope elements, the last element determinted the
730 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
731 Electronic Mail") by including:
733 Auto-submitted: auto-generated
735 in the messages that it generates (bounce messages and others, such as
736 warnings). In the case of bounce messages for non-SMTP mesages, there was
737 also a typo: it was using "Auto_submitted" (underscore instead of
738 hyphen). Since every message generated by Exim is necessarily in response
739 to another message, thes have all been changed to:
741 Auto-Submitted: auto-replied
743 in accordance with these statements in the RFC:
745 The auto-replied keyword:
747 - SHOULD be used on messages sent in direct response to another
748 message by an automatic process,
750 - MUST NOT be used on manually-generated messages,
752 - MAY be used on Delivery Status Notifications (DSNs) and Message
753 Disposition Notifications (MDNs),
755 - MUST NOT be used on messages generated by automatic or periodic
756 processes, except for messages which are automatic responses to
759 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
760 to the default Received: header definition.
762 PH/12 Added log selector acl_warn_skipped (default on).
764 PH/13 After a successful wildlsearch lookup, discard the values of numeric
765 variables because (a) they are in the wrong storage pool and (b) even if
766 they were copied, it wouldn't work properly because of the caching.
768 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
769 checking when decoding. Apparently there are clients that generate
770 overlong encoded strings. Why am I not surprised?
772 PH/15 If the first argument of "${if match_address" was not empty, but did not
773 contain an "@" character, Exim crashed. Now it writes a panic log message
774 and treats the condition as false.
776 PH/16 In autoreply, treat an empty string for "once" the same as unset.
778 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
779 extension "envelope-auth". The code is finished and in agreement with
780 other implementations, but there is no documentation so far and in fact,
781 nobody wrote the draft yet. This extension is currently #undef'ed, thus
782 not changing the active code.
784 Print executed "if" and "elsif" statements when debugging is used. This
785 helps a great deal to understand what a filter does.
787 Document more things not specified clearly in RFC3028. I had all this
788 sorted out, when out of a sudden new issues came to my mind. Oops."
790 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
793 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
794 canonical form (as documented). However, after a host name lookup from
795 the IP address, check_host() was doing a simple string comparison with
796 addresses acquired from the DNS when checking that the found name did
797 have the original IP as one of its addresses. Since any found IPv6
798 addresses are likely to be in abbreviated form, the comparison could
799 fail. Luckily, there already exists a function for doing the comparison
800 by converting both addresses to binary, so now that is used instead of
803 PH/20 There was another similar case to PH/19, when a complete host name was
804 given in a host list; looking up its IP address could give an abbreviated
805 form, whereas the current host's name might or might not be abbreviated.
806 The same fix has been applied.
812 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
813 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
816 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
817 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
819 PH/03 Typo: missing ".o" in src/pcre/Makefile.
821 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
822 header line, restrict the check to what is listed in RFCs 2369 and 2929.
823 Also, for "Auto-Submitted", treat anything other than "no" as
824 non-personal, in accordance with RFC 3834. (Previously it treated
825 anything starting "auto-" as non-personal.)
827 TF/01 The control=submission/name=... option had a problem with syntax
828 errors if the name included a slash character. The /name= option
829 now slurps the rest of the string, so it can include any characters
830 but it must come last in the list of options (after /sender_retain
833 PH/05 Some modifications to the interface to the fake nameserver for the new
841 TK/01 Added the "success_on_redirect" address verification option. See
842 NewStuff for rationale and an example.
844 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
846 PH/02 Patch to exigrep to allow it to work on syslog lines.
848 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
849 fread() to skip over the body file's header line, because in Cygwin the
850 header line is locked and is inaccessible.
852 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
853 co-exist for some time) to make it clear that it is the Exim ID that is
854 referenced, not the Message-ID: header line.
856 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
857 string_format() function, because snprintf() does not exist on all
860 PH/06 The use of forbid_filter_existstest now also locks out the use of the
861 ${stat: expansion item.
863 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
864 protocol synchronization error", to keep the pedants happy.
866 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
867 well as for IRIX systems, when gcc is being used. See the host.c source
870 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
872 PH/10 Named domain lists were not working if used in a queue_smtp_domains
875 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
876 transport and to the smtp transport in LMTP mode.
878 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
880 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
881 run a filter in a subprocess. This could lead to confusion in subsequent
882 lookups in the parent process. There should also be a search_tidyup() at
883 the end of the subprocess.
885 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
886 only if the host matched helo_try_verify_hosts, which caused the
887 verification to occur when the EHLO/HELO command was issued. The ACL just
888 tested the remembered result. Now, if a previous verification attempt has
889 not happened, "verify = helo" does it there and then.
891 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
893 TK/03 Fix log output including CR from clamd.
895 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
896 exist provoked a memory error which could cause a segfault.
898 PH/15 Installed PCRE 6.2
900 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
902 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
903 of the problem. Specifically, suggested +O2 rather than +O1 for the
906 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
908 PH/20 If a delivery was routed to a non-standard port by means of an SRV
909 record, the port was not correctly logged when the outgoing_port log
910 selector was set (it logged the transort's default port).
912 PH/21 Added support for host-specific ports to manualroute, queryprogram,
913 fallback_hosts, and "hosts" in the smtp transport.
915 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
916 host errors such as "Connection refused".
918 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
919 authentication with radiusclient 0.4.9:
921 - Error returned from rc_read_config was caught wrongly
922 - Username/password not passed on to radius server due to wrong length.
924 The presumption is that some radiusclient API changes for 4.51/PH/17
925 were not taken care of correctly. The code is still untested by me (my
926 Linux distribution still has 0.3.2 of radiusclient), but it was
927 contributed by a Radius user.
929 PH/24 When doing a callout, the value of $domain wasn't set correctly when
930 expanding the "port" option of the smtp transport.
932 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
933 while reading a MIME header. Thanks to Tom Hughes for a patch.
935 PH/24 Include config.h inside local_scan.h so that configuration settings are
938 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
939 that in an ACL for RCPT (for example), you can examine exactly what was
942 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
943 commands, but it was not correctly comparing the address with the actual
944 client host address. Thus, it would show the EHLO address in Received:
945 header lines when this was not necessary.
947 PH/27 Added the % operator to ${eval:}.
949 PH/28 Exim tries to create and chdir to its spool directory when it starts;
950 it should be ignoring failures (because with -C, for example, it has lost
951 privilege). It wasn't ignoring creation failures other than "already
954 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
957 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
958 because an input line was too long, either on its own, or by virtue of
959 too many continuations, the temporary file was not being removed, and the
960 return code was incorrect.
962 PH/31 Missing "BOOL" in function definition in filtertest.c.
964 PH/32 Applied Sieve patches from the maintainer.
966 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
968 PH/33 Added "verify = not_blind".
970 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
971 Local/Makefile (with some defaults set). These are used in built scripts
972 such as exicyclog, but they have never been used in the exim_install
973 script (though there are many overriding facilities there). I have
974 arranged that the exim_install script now takes note of these two
977 PH/35 Installed configuration files for Dragonfly.
979 PH/36 When a locally submitted message by a trusted user did not contain a
980 From: header, and the sender address was obtained from -f or from an SMTP
981 MAIL command, and the trusted user did not use -F to supply a sender
982 name, $originator_name was incorrectly used when constructing a From:
983 header. Furthermore, $originator_name was used for submission mode
984 messages from external hosts without From: headers in a similar way,
985 which is clearly wrong.
987 PH/37 Added control=suppress_local_fixups.
989 PH/38 When log_selector = +received_sender was set, and the addition of the
990 sender made the log line's construction buffer exactly full, or one byte
991 less than full, an overflow happened when the terminating "\n" was
994 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
995 when the result of a list match is failure because a DNS lookup failed.
997 PH/40 RM_COMMAND is now used in the building process.
999 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1000 the "build-* directories that it finds.
1002 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1003 address in a domain literal was a prefix of an interface address.
1005 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1006 when verifying a sender address, unless rewrite_headers is false.
1008 PH/44 Wrote a long comment about why errors_to addresses are verified as
1009 recipients, not senders.
1011 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1012 the ratelimit ACL was added.
1014 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1016 PH/46 Added extra information about PostgreSQL errors to the error string.
1018 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1019 avoiding the need to install special zones in a real server. This is
1020 backwards compatible; if it can't find the fake resolver, it drops back.
1021 Thus, both old and new test suites can be run.
1023 TF/02 Added util/ratelimit.pl
1025 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1026 clock is set back in time.
1028 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1029 Candler <B.Candler@pobox.com>.
1031 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1032 OK for addresses that are the result of redirections.
1034 PH/48 A number of further additions for the benefit of the new test suite,
1035 including a fake gethostbyname() that interfaces to the fake DNS resolver
1038 TF/06 The fix for widen_domains has also been applied to qualify_single and
1039 search_parents which are the other dnslookup options that can cause
1042 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1045 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1047 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1048 service" if it encountered an MX record with an empty target hostname.
1049 The message is now "an MX or SRV record indicated no SMTP service".
1051 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1052 if the DNS of the sending site is misconfigured. This is quite a
1053 common situation. This change restores the behaviour of treating a
1054 helo verification defer as a failure.
1056 PH/51 If self=fail was set on a router, the bounce message did not include the
1057 actual error message.
1063 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1065 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1066 command itself ended in error, the underlying message about the transport
1067 filter timeout was being overwritten with the pipe command error. Now the
1068 underlying error message should be appended to the second error message.
1070 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1072 PH/02 Reduce the amount of output that "make" produces by default. Full output
1073 can still be requested.
1075 PH/03 The warning log line about a condition test deferring for a "warn" verb
1076 was being output only once per connection, rather than after each
1077 occurrence (because it was using the same function as for successful
1078 "warn" verbs). This seems wrong, so I have changed it.
1080 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1081 it should not have, which might have caused a crash in the right
1082 circumstances, but probably never did.
1084 PH/04 Installed a modified version of Tony Finch's patch to make submission
1085 mode fix the return path as well as the Sender: header line, and to
1086 add a /name= option so that you can make the user's friendly name appear
1089 TF/03 Added the control = fakedefer ACL modifier.
1091 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1092 Mark Lowes for thorough testing.
1094 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1096 TK/03 Merged latest SRS patch from Miles Wilton.
1098 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1099 with the definition in sysexits.h (which is #included earlier).
1100 Fortunately, Exim does not actually use EX_OK. The code used to try to
1101 preserve the sysexits.h value, by assumimg that macro definitions were
1102 scanned for macro replacements. I have been disabused of this notion,
1103 so now the code just undefines EX_OK before #including unistd.h.
1105 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1106 in the smtp transport. When a block could not be written in a single
1107 write() function, the timeout was being re-applied to each part-write.
1108 This seems wrong - if the receiver was accepting one byte at a time it
1109 would take for ever. The timeout is now adjusted when this happens. It
1110 doesn't have to be particularly precise.
1112 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1113 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1115 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1116 without a domain if the check to <postmaster@domain> fails.
1118 SC/01 Eximstats: added -xls and the ability to specify output files
1119 (patch written by Frank Heydlauf).
1121 SC/02 Eximstats: use FileHandles for outputing results.
1123 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1125 SC/04 Eximstats: fixed display of large numbers with -nvr option
1127 SC/05 Eximstats: fixed merging of reports with empty tables.
1129 SC/06 Eximstats: added the -include_original_destination flag
1131 SC/07 Eximstats: removed tabs and trailing whitespace.
1133 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1135 TK/06 MBOX spool code: Add real "From " MBOX separator line
1136 so the .eml file is really in mbox format (even though
1137 most programs do not really care). Patch from Alex Miller.
1139 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1140 The latter is generated from $received_to and is only set if the
1141 message has one envelope recipient. SA can use these headers,
1142 obviously out-of-the-box. Patch from Alex Miller.
1144 PH/08 The ${def test on a variable was returning false if the variable's
1145 value was "0", contrary to what the specification has always said!
1146 The result should be true unless the variable is empty.
1148 PH/09 The syntax error of a character other than { following "${if
1149 def:variable_name" (after optional whitespace) was not being diagnosed.
1150 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1151 accidental colon was present, for example, could give incorrect results.
1153 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1154 result is used (not including appendfile, where other changes are about
1157 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1158 This involved changing a lot of size variables from int to off_t. It
1159 should work with maildirs and everything.
1161 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1162 spamd dying while we are connected to it.
1164 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1167 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1168 function for BDB 4.3.
1170 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1172 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1173 the sources for PCRE have been re-arranged and more files are now
1176 PH/15 The code I had for printing potentially long long variables in PH/11
1177 above was not the best (it lost precision). The length of off_t variables
1178 is now inspected at build time, and an appropriate printing format (%ld
1179 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1180 to be "long long int" or "long int". This is needed for the internal
1181 formatting function string_vformat().
1183 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1184 the configuration file to be ":syslog", then the script "guesses" where
1185 the logs files are, rather than using the compiled in default. In our
1186 case the guess is not the same as the compiled default, so the script
1187 suddenly stopped working when I started to use syslog. The patch checks
1188 to see if log_file_path is "". If so, it attempts to read it from exim
1189 with no configuration file to get the compiled in version, before it
1190 falls back to the previous guessing code."
1192 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1193 implementing BATV in an Exim configuration. See NewStuff for the gory
1196 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1197 Makefile that are specific to HP-UX.
1199 PH/18 If the "use_postmaster" option was set for a recipient callout together
1200 with the "random" option, the postmaster address was used as the MAIL
1201 FROM address for the random test, but not for the subsequent recipient
1202 test. It is now used for both.
1204 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1205 patch removes a few documentation additions to RFC 3028, because the
1206 latest draft now contains them. It adds the new en;ascii-case comparator
1207 and a new error check for 8bit text in MIME parts. Comparator and
1208 require names are now matched exactly. I enabled the subaddress
1209 extension, but it is not well tested yet (read: it works for me)."
1211 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1212 rework some of the code of TK/09 above to avoid the hardwired use of
1213 "%lld" and "long long". Replaced the call to snprintf() with a call to
1216 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1217 records point to non-existent hosts", "retry timeout exceeded", and
1218 "retry time not reached for any host after a long failure period".
1220 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1221 experimental DomainKeys support:
1223 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1224 (2) On an error such as an illegally used "control", the wrong name for
1225 the control was given.
1227 These problems did NOT occur unless DomainKeys support was compiled.
1229 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1231 PH/24 Added ${if match_ip condition.
1233 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1234 databases so that it will be absolutely obvious if a crash occurs in the
1235 DB library. This is a regular occurrence (often caused by mis-matched
1238 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1239 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1240 picked up on a user's system that detects such things. There doesn't seem
1241 to be a gcc warning option for this - only an attribute that has to be
1242 put on the function's prototype. It seems that in Fedora Core 4 they have
1243 set this on a number of new functions. No doubt there will be more in due
1246 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1247 specify a transport. However, if an address that was verified by such a
1248 router was the subject of a callout, Exim crashed because it tried to
1249 read the rcpt_include_affixes from the non-existent transport. Now it
1250 just assumes that the setting of that option is false. This bug was
1251 introduced by 4.51/PH/31.
1253 PH/28 Changed -d+all to exclude +memory, because that information is very
1254 rarely of interest, but it makes the output a lot bigger. People tend to
1255 do -d+all out of habit.
1257 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1258 code in os-type was giving problems when libc.so lives in lib64, like on
1261 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1262 aren't the modern standard, and it seems that some systems' include files
1263 don't always have them. Exim was already checking for some of the newer
1264 ones like T_AAAA, and defining it itself. I've added checks for all the
1265 record types that Exim uses.
1267 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1268 not automatically generating a new one, as it is supposed to. This
1269 prevented TLS from working. If the file did exist, but contained invalid
1270 data, a new version was generated, as expected. It was only the case of a
1271 non-existent file that was broken.
1273 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1274 with a change in libdomainkeys > 0.64.
1276 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1277 from DNS. If the selector record carries the flag, it now has
1278 precedence over the domain-wide flag.
1280 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1282 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1283 the use of a port name that isn't defined in /etc/services) occurred, the
1284 message was deferred as in a normal delivery, and thus remained on the
1285 spool, instead of being failed because of the mua_wrapper setting. This
1286 is now fixed, and I tidied up some of the mua_wrapper messages at the
1289 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1290 the messages in a hash of arrays rather than using individual hashes.
1291 This is a bit cleaner and results in dramatic memory savings, albeit
1292 at a slight CPU cost.
1294 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1295 as requested by Marc Sherman.
1297 SC/10 Eximstats: added histograms for user specified patterns as requested
1300 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1302 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1303 fopen() in the content-scanning modules that did not already have it.
1309 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1310 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1312 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1314 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1316 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1317 to test Sieve filters that use "vacation".
1319 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1320 that changes the way the GnuTLS parameters are stored in the cache file.
1321 The new format can be generated externally. For backward compatibility,
1322 if the data in the cache doesn't make sense, Exim assumes it has read an
1323 old-format file, and it generates new data and writes a new file. This
1324 means that you can't go back to an older release without removing the
1327 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1328 work if there are any delivery delays because "one_time" forces the
1329 parent to be marked "delivered", so its unseen clone is never tried
1330 again. For this reason, Exim now forbids the simultaneous setting of
1333 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1334 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1335 entry if you want to know the details. The fix, however, neglected to
1336 consider the case when local delivery batching is involved. The test for
1337 "previously delivered" was not happening when checking to see if an
1338 address could be batched with a previous (undelivered) one; under
1339 certain circumstances this could lead to multiple deliveries to the same
1342 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1343 in its include files, and this causes problems building Exim.
1345 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1346 header_syntax) but Exim was just ignoring anything given after a slash.
1347 In particular, this caused confusion with an attempt to use "verify =
1348 reverse_host_lookup/defer_ok". An error is now given when options are
1349 supplied for verify items that do not have them. (Maybe reverse_host_
1350 lookup should have a defer_ok option, but that's a different point.)
1352 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1353 defined by RFC 821) to 2048, because there were problems with some AUTH
1354 commands, and RFC 1869 says the size should be increased for extended
1355 SMTP commands that take arguments.
1357 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1360 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1361 "unknown" error; now it says that the functionality isn't in the binary.
1363 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1364 an address' error message when a string expansion fails (syntax or
1365 whatever). Otherwise the password may appear in the log. Following change
1366 PH/42 below, there is no longer a chance of it appearing in a bounce
1369 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1371 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1372 of its addresses was ever tried. (Bugzilla bug #2.)
1374 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1375 the result incorrectly in the debug output. (It correctly added a newline
1376 to what was transported.)
1378 TF/01 Added $received_time.
1380 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1381 commented out examples of how to interface to a virus scanner and to
1382 SpamAssassin. Also added commented examples of av_scanner and
1383 spamd_address settings.
1385 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1386 and controls are allowed in which ACLs. There were a couple of minor
1387 errors. Some of the entries in the conditions table (which is a table of
1388 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1389 negation of where the condition IS allowed.
1391 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1393 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1394 header file does not have a version number, so I've had to invent a new
1395 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1396 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1397 radiusclient), but it was contributed by a Radius user.
1399 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1400 files or directories, for OpenSSL.
1402 PH/19 When an Exim process that is running as root has to create an Exim log
1403 file, it does so in a subprocess that runs as exim:exim so as to get the
1404 ownership right at creation (otherwise, other Exim processes might see
1405 the file with the wrong ownership). There was no test for failure of this
1406 fork() call, which would lead to the process getting stuck as it waited
1407 for a non-existent subprocess. Forks do occasionally fail when resources
1408 run out. I reviewed all the other calls to fork(); they all seem to check
1411 PH/20 When checking for unexpected SMTP input at connect time (before writing
1412 the banner), Exim was not dealing correctly with a non-positive return
1413 from the read() function. If the client had disconnected by this time,
1414 the result was a log entry for a synchronization error with an empty
1415 string after "input=" when read() returned zero. If read() returned -1
1416 (an event I could not check), uninitialized data bytes were printed.
1417 There were reports of junk text (parts of files, etc) appearing after
1420 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1422 PH/22 Added support for macro redefinition, and (re)definition in between
1423 driver and ACL definitions.
1425 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1426 forgetting to use the resulting value; it was using the unexpanded value.
1428 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1429 hadn't been configured. The fix is from Juergen Kreileder, who
1430 understands it better than I do:
1432 "Here's what I see happening with three configured cyrus_sasl
1433 authenticators configured (plain, login, cram-md5):
1435 On startup auth_cyrus_sasl_init() gets called for each of these.
1436 This means three calls to sasl_listmech() without a specified mech_list.
1437 => SASL tests which mechs of all available mechs actually work
1438 => three warnings about OTP not working
1439 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1441 With the patch, sasl_listmech() also gets called three times. But now
1442 SASL's mech_list option is set to the server_mech specified in the the
1443 authenticator. Or in other words, the answer from sasl_listmech()
1444 gets limited to just the mech you're testing for (which is different
1446 => the return list contains just 'plain' or 'login', 'cram-md5' or
1447 nothing depending on the value of ob->server_mech.
1449 I've just tested the patch: Authentication still works fine,
1450 unavailable mechs specified in the exim configuration are still
1451 caught, and the auth.log warnings about OTP are gone."
1453 PH/25 When debugging is enabled, the contents of the command line are added
1454 to the debugging output, even when log_selector=+arguments is not
1457 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1458 answer is "GNU", and only if the return is "GNU/something" is the answer
1461 PH/27 $acl_verify_message is now set immediately after the failure of a
1462 verification in an ACL, and so is available in subsequent modifiers. In
1463 particular, the message can be preserved by coding like this:
1465 warn !verify = sender
1466 set acl_m0 = $acl_verify_message
1468 Previously, $acl_verify_message was set only while expanding "message"
1469 and "log_message" when a very denied access.
1471 PH/28 Modified OS/os.c-Linux with
1473 -#ifndef OS_LOAD_AVERAGE
1474 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1476 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1477 nomenclature these days.)
1479 PH/29 Installed patch from the Sieve maintainer that adds the options
1480 sieve_useraddress and sieve_subaddress to the redirect router.
1482 PH/30 In these circumstances:
1483 . Two addresses routed to the same list of hosts;
1484 . First host does not offer TLS;
1485 . First host accepts first address;
1486 . First host gives temporary error to second address;
1487 . Second host offers TLS and a TLS session is established;
1488 . Second host accepts second address.
1489 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1490 and peerdn, if requested) that were in fact used only for the second
1493 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1494 attention to any local part prefix or suffix that was matched by the
1495 router that accepted the address. It now behaves in the same way as it
1496 does for delivery: the affixes are removed from the local part unless
1497 rcpt_include_affixes is set on the transport.
1499 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1500 timeout during the DATA phase of an incoming message.
1502 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1503 applied a patch sanctioned by the Sieve maintainer.
1505 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1506 the uid or gid is negative. A case of a negative gid caused this to be
1507 noticed. The fix allows for either to be negative.
1509 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1510 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1513 PH/36 The change PH/12 above was broken. Fixed it.
1515 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1516 the grounds that routing the same address twice would always produce the
1517 same answer. This might have been true once, but it is certainly no
1518 longer true now. Routing a child address may depend on the previous
1519 routing that produced that child. Some complicated redirection strategies
1520 went wrong when messages had multiple recipients, and made Exim's
1521 behaviour dependent on the order in which the addresses were given.
1523 I have moved the duplicate checking until after the routing is complete.
1524 Exim scans the addresses that are assigned to local and remote
1525 transports, and removes any duplicates. This means that more work will be
1526 done, as duplicates will always all be routed, but duplicates are
1527 presumably rare, so I don't expect this is of any significance.
1529 For deliveries to pipes, files, and autoreplies, the duplicate checking
1530 still happens during the routing process, since they are not going to be
1533 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1534 It corrects a timeout issue with spamd. This is Ian's comment: "The
1535 background is that sometimes spamd either never reads data from a
1536 connection it has accepted, or it never writes response data. The exiscan
1537 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1538 blindly assumes that writes won't block so it may never time out."
1540 PH/39 Allow G after quota size as well as K and M.
1542 PH/40 The value set for $authenticated_id in an authenticator may not contain
1543 binary zeroes or newlines because the value is written to log lines and
1544 to spool files. There was no check on this. Now the value is run through
1545 the string_printing() function so that such characters are converted to
1546 printable escape sequences.
1548 PH/41 $message_linecount is a new variable that contains the total number of
1549 lines in the message. Compare $body_linecount, which is the count for the
1552 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1553 bounce and delay warning messages, except in certain special cases, which
1556 (a) An SMTP error message from a remote host;
1557 (b) A message specified in a :fail: redirection;
1558 (c) A message specified in a "fail" command in a system filter;
1559 (d) A message specified in a FAIL return from the queryprogram router;
1560 (e) A message specified by the cannot_route_message router option.
1562 In these cases only, Exim does include the error details in bounce and
1563 warning messages. There are also a few cases where bland messages such
1564 as "unrouteable address" or "local delivery error" are given.
1566 PH/43 $value is now also set for the "else" part of a ${run expansion.
1568 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1569 being worked on, but at least Exim now implements the latest version to
1572 PH/45 In a pipe transport, although a timeout while waiting for the pipe
1573 process to complete was treated as a delivery failure, a timeout while
1574 writing the message to the pipe was logged, but erroneously treated as a
1575 successful delivery. Such timeouts include transport filter timeouts. For
1576 consistency with the overall process timeout, these timeouts are now
1577 treated as errors, giving rise to delivery failures by default. However,
1578 there is now a new Boolean option for the pipe transport called
1579 timeout_defer, which, if set TRUE, converts the failures into defers for
1580 both kinds of timeout. A transport filter timeout is now identified in
1583 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1584 systems where "make" and "gmake" are different, calling "gmake" at top
1585 level broke things. I've arranged for the value of $(MAKE) to be passed
1586 from the Makefile to this script so that it can call the same version of
1590 A note about Exim versions 4.44 and 4.50
1591 ----------------------------------------
1593 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1594 changes of various kinds. As a consequence, a big documentation update was
1595 needed. This delayed the release for rather longer than seemed good, especially
1596 in the light of a couple of (minor) security issues. Therefore, the changes
1597 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1598 release. So 4.44 and 4.50 are in effect two different branches that both start
1601 I have left the 4.50 change log unchanged; it contains all the changes since
1602 4.43. The change log for 4.44 is below; many of its items are identical to
1603 those for 4.50. This seems to be the most sensible way to preserve the
1604 historical information.
1610 1. Minor wording change to the doc/README.SIEVE file.
1612 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1613 computation of the current number of files was incorrect.
1615 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1616 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1619 4. Give more explanation in the error message when the command for a transport
1620 filter fails to execute.
1622 5. There are several places where Exim runs a non-Exim command in a
1623 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1624 was being done only for the command run by the queryprogram router. It is
1625 now done for all such subprocesses. The other cases are: ${run, transport
1626 filters, and the commands run by the lmtp and pipe transports.
1628 6. Added CONFIGURE_GROUP build-time option.
1630 7. Some older OS have a limit of 256 on the maximum number of file
1631 descriptors. Exim was using setrlimit() to set 1000 as a large value
1632 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1633 systems. I've change it so that if it can't get 1000, it tries for 256.
1635 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1636 was an oversight, and furthermore, ever since the addition of extra
1637 controls (e.g. 4.43/32), the checks on when to allow different forms of
1638 "control" were broken. There should now be diagnostics for all cases when a
1639 control that does not make sense is encountered.
1641 9. Added the /retain_sender option to "control=submission".
1643 10. $recipients is now available in the predata ACL (oversight).
1645 11. Tidy the search cache before the fork to do a delivery from a message
1646 received from the command line. Otherwise the child will trigger a lookup
1647 failure and thereby defer the delivery if it tries to use (for example) a
1648 cached ldap connection that the parent has called unbind on.
1650 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1651 of $address_data from the recipient verification was clobbered by the
1652 sender verification.
1654 13. The value of address_data from a sender verification is now available in
1655 $sender_address_data in subsequent conditions in the ACL statement.
1657 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1659 15. Added a new option "connect=<time>" to callout options, to set a different
1662 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1663 was its contents. (It was OK if the option was not defined at all.)
1665 17. A "Completed" log line is now written for messages that are removed from
1666 the spool by the -Mrm option.
1668 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1669 information about exactly what failed.
1671 19. Added -dd to debug only the daemon process.
1673 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1674 handles timeouts, both on the server side and network timeouts. Renamed the
1675 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1677 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1678 It is now set to "smtps".
1680 22. $host_address is now set to the target address during the checking of
1681 ignore_target_hosts.
1683 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1684 being passed; this would have caused $sender_host_name to have been used if
1685 matching the list had actually called for a host name (not very likely,
1686 since this list is usually IP addresses). A host name is now passed as
1689 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1690 code that specifies a non-restarting handler (typically sigaction() in
1691 modern systems) in an attempt to fix a rare and obscure crash bug.
1693 25. Narrowed the window for a race in the daemon that could cause it to ignore
1694 SIGCHLD signals. This is not a major problem, because they are used only to
1695 wake it up if nothing else does.
1697 26. A malformed maildirsize file could cause Exim to calculate negative values
1698 for the mailbox size or file count. Odd effects could occur as a result.
1699 The maildirsize information is now recalculated if the size or filecount
1702 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1703 support for a long time. Removed HAVE_SYS_VFS_H.
1705 28. Installed the latest version of exipick from John Jetmore.
1707 29. In an address list, if the pattern was not a regular expression, an empty
1708 subject address (from a bounce message) matched only if the pattern was an
1709 empty string. Non-empty patterns were not even tested. This was the wrong
1710 because it is perfectly reasonable to use an empty address as part of a
1711 database query. An empty address is now tested by patterns that are
1712 lookups. However, all the other forms of pattern expect the subject to
1713 contain a local part and a domain, and therefore, for them, an empty
1714 address still always fails if the pattern is not itself empty.
1716 30. Exim went into a mad DNS loop when attempting to do a callout where the
1717 host was specified on an smtp transport, and looking it up yielded more
1718 than one IP address.
1720 31. Re-factored the code for checking spool and log partition space into a
1721 function that finds that data and another that does the check. The former
1722 is then used to implement four new variables: $spool_space, $log_space,
1723 $spool_inodes, and $log_inodes.
1725 32. The RFC2047 encoding function was originally intended for short strings
1726 such as real names; it was not keeping to the 75-character limit for
1727 encoded words that the RFC imposes. It now respects the limit, and
1728 generates multiple encoded words if necessary. To be on the safe side, I
1729 have increased the buffer size for the ${rfc2047: expansion operator from
1732 33. It is now permitted to omit both strings after an "if" condition; if the
1733 condition is true, the result is "true". As before, when the second string
1734 is omitted, a false condition yields an empty string. This makes it less
1735 cumbersome to write custom ACL and router conditions.
1737 34. Failure to deliver a bounce message always caused it to be frozen, even if
1738 there was an errors_to setting on the router. The errors_to setting is now
1741 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1742 canonical form (fully expanded) before being placed in
1743 $sender_host_address.
1745 36. The table in the code that translates DNS record types into text (T_A to
1746 "A" for instance) was missing entries for NS and CNAME. It is just possible
1747 that this could have caused confusion if both these types were looked up
1748 for the same domain, because the text type is used as part of Exim's
1749 per-process caching. But the chance of anyone hitting this buglet seems
1752 37. The dnsdb lookup has been extended in a number of ways.
1754 (1) There is a new type, "zns", which walks up the domain tree until it
1755 finds some nameserver records. It should be used with care.
1757 (2) There is a new type, "mxh", which is like "mx" except that it returns
1758 just the host names, not the priorities.
1760 (3) It is now possible to give a list of domains (or IP addresses) to be
1761 looked up. The behaviour when one of the lookups defers can be
1762 controlled by a keyword.
1764 (4) It is now possible to specify the separator character for use when
1765 multiple records are returned.
1767 38. The dnslists ACL condition has been extended: it is now possible to supply
1768 a list of IP addresses and/or domains to be looked up in a particular DNS
1771 39. Added log_selector=+queue_time_overall.
1773 40. When running the queue in the test harness, wait just a tad after forking a
1774 delivery process, to get repeatability of debugging output.
1776 41. Include certificate and key file names in error message when GnuTLS fails
1777 to set them up, because the GnuTLS error message doesn't include the name
1778 of the failing file when there is a problem reading it.
1780 42. Allow both -bf and -bF in the same test run.
1782 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1784 44. The "Exiscan patch" is now merged into the mainline Exim source.
1786 45. Sometimes the final signoff response after QUIT could fail to get
1787 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1788 < 0 before doing a fflush(). This bug looks as though it goes back to the
1789 introduction of TLS in release 3.20, but "sometimes" must have been rare
1790 because the tests only now provoked it.
1792 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1793 (this can affect the format of dates).
1795 47. exim_tidydb, when checking for the continued existence of a message for
1796 which it has found a message-specific retry record, was not finding
1797 messages that were in split spool directories. Consequently, it was
1798 deleting retry records that should have stayed in existence.
1800 48. Steve fixed some bugs in eximstats.
1802 49. The SPA authentication driver was not abandoning authentication and moving
1803 on to the next authenticator when an expansion was forced to fail,
1804 contradicting the general specification for all authenticators. Instead it
1805 was generating a temporary error. It now behaves as specified.
1807 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1808 (the order specifies the preference for clients). The order is now AES256,
1809 AES128, 3DES, ARCFOUR128.
1811 51. Small patch to Sieve code - explicitly set From: when generating an
1814 52. Exim crashed if a remote delivery caused a very long error message to be
1815 recorded - for instance if somebody sent an entire SpamAssassin report back
1816 as a large number of 550 error lines. This bug was coincidentally fixed by
1817 increasing the size of one of Exim's internal buffers (big_buffer) that
1818 happened as part of the Exiscan merge. However, to be on the safe side, I
1819 have made the code more robust (and fixed the comments that describe what
1822 53. Now that there can be additional text after "Completed" in log lines (if
1823 the queue_time_overall log selector is set), a one-byte patch to exigrep
1824 was needed to allow it to recognize "Completed" as not the last thing in
1827 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1828 patch that reportedly fixes this has been added. I am not expert enough to
1829 create a test for it. This is what the patch creator wrote:
1831 "I found a little strange behaviour of ldap code when working with
1832 Windows 2003 AD Domain, where users was placed in more than one
1833 Organization Units. When I tried to give exim partial DN, the exit code
1834 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1835 But simultaneously result of request was absolutely normal ldap result,
1836 so I produce this patch..."
1838 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1839 REFERENCE, so I have modified the code to exclude the patch when that macro
1842 55. Some experimental protocols are using DNS PTR records for new purposes. The
1843 keys for these records are domain names, not reversed IP addresses. The
1844 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1845 leaves it alone. Component reversal etc. now happens only for IP addresses.
1848 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1850 57. Double the size of the debug message buffer (to 2048) so that more of very
1851 long debug lines gets shown.
1853 58. The exicyclog utility now does better if the number of log files to keep
1854 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1856 59. Two changes related to the smtp_active_hostname option:
1858 (1) $smtp_active_hostname is now available as a variable.
1859 (2) The default for smtp_banner uses $smtp_active_hostname instead
1860 of $primary_hostname.
1862 60. The host_aton() function is supposed to be passed a string that is known
1863 to be a valid IP address. However, in the case of IPv6 addresses, it was
1864 not checking this. This is a hostage to fortune. Exim now panics and dies
1865 if the condition is not met. A case was found where this could be provoked
1866 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1867 components; fortuitously, this particular loophole had already been fixed
1868 by change 4.50/55 above.
1870 If there are any other similar loopholes, the new check in host_aton()
1871 itself should stop them being exploited. The report I received stated that
1872 data on the command line could provoke the exploit when Exim was running as
1873 exim, but did not say which command line option was involved. All I could
1874 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1875 running as the user.
1878 61. There was a buffer overflow vulnerability in the SPA authentication code
1879 (which came originally from the Samba project). I have added a test to the
1880 spa_base64_to_bits() function which I hope fixes it.
1883 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1884 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1886 63. The daemon start-up calls getloadavg() while still root for those OS that
1887 need the first call to be done as root, but it missed one case: when
1888 deliver_queue_load_max is set with deliver_drop_privilege. This is
1889 necessary for the benefit of the queue runner, because there is no re-exec
1890 when deliver_drop_privilege is set.
1892 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1893 This has been fixed.
1895 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1896 was in use, was not putting the data itself into the right store pool;
1897 consequently, it could be overwritten for a subsequent message in the same
1898 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1901 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1903 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1904 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1905 the function were treating the return as a boolean value, which happened to
1906 work because 0=false and not-0=true, but is not correct code.
1908 68. The host_aton() function was not handling scoped IPv6 addresses (those
1909 with, for example, "%eth0" on the end) correctly.
1911 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1912 negated items (that is, ~something) in unsigned ints. Some compilers
1913 apparently mutter when there is no cast.
1915 70. If an address verification called from an ACL failed, and did not produce a
1916 user-specific message (i.e. there was only a "system" message), nothing was
1917 put in $acl_verify_message. In this situation, it now puts the system
1920 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1921 session; change 4.31/43 added the unwanted input to the log line - except
1922 that it did not do this in the start of session case. It now does.
1924 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1925 This is wrong and can cause the other end to generate a synchronization
1926 error if it is another Exim or anything else that does the synchronization
1927 check. A QUIT command is no longer sent after a timeout.
1929 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1930 during host lookups.
1932 74. The defer_ok option of callout verification was not working if it was used
1933 when verifying addresses in header lines, that is, for this case:
1935 verify = header_sender/callout=defer_ok
1937 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1938 those file descriptors could be used for SMTP connections. If anything
1939 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1940 could be sent to the SMTP client, causing chaos. The daemon now opens
1941 stdin, stdout, and stderr to /dev/null when it puts itself into the
1944 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1945 log by default. The user can override this with suitable Perl magic.
1947 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1948 the log message when discard triggers, was not working for the DATA ACL or
1949 for the non-SMTP ACL.
1951 78. Error message wording change in sieve.c.
1953 79. If smtp_accept_max_per_host was set, the number of connections could be
1954 restricted to fewer than expected, because the daemon was trying to set up
1955 a new connection before checking whether the processes handling previous
1956 connections had finished. The check for completed processes is now done
1957 earlier. On busy systems, this bug wouldn't be noticed because something
1958 else would have woken the daemon, and it would have reaped the completed
1961 80. If a message was submitted locally by a user whose login name contained one
1962 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1963 It caused a spool format error. I have fixed the spool reading code. A
1964 related problem was that the "from" clause in the Received: line became
1965 illegal because of the space(s). It is now covered by ${quote_local_part.
1967 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
1970 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
1972 83. Patch from Sieve maintainer to fix unterminated string problem in
1973 "vacation" handling.
1975 84. Some minor changes to the Linux configuration files to help with other
1976 OS variants using glibc.
1978 85. One more patch for Sieve to update vacation handling to latest spec.
1981 ----------------------------------------------------
1982 See the note above about the 4.44 and 4.50 releases.
1983 ----------------------------------------------------
1989 1. Change 4.43/35 introduced a bug that caused file counts to be
1990 incorrectly computed when quota_filecount was set in an appendfile
1993 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1994 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1997 3. Give more explanation in the error message when the command for a transport
1998 filter fails to execute.
2000 4. There are several places where Exim runs a non-Exim command in a
2001 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2002 was being done only for the command run by the queryprogram router. It is
2003 now done for all such subprocesses. The other cases are: ${run, transport
2004 filters, and the commands run by the lmtp and pipe transports.
2006 5. Some older OS have a limit of 256 on the maximum number of file
2007 descriptors. Exim was using setrlimit() to set 1000 as a large value
2008 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2009 systems. I've change it so that if it can't get 1000, it tries for 256.
2011 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2012 was an oversight, and furthermore, ever since the addition of extra
2013 controls (e.g. 4.43/32), the checks on when to allow different forms of
2014 "control" were broken. There should now be diagnostics for all cases when a
2015 control that does not make sense is encountered.
2017 7. $recipients is now available in the predata ACL (oversight).
2019 8. Tidy the search cache before the fork to do a delivery from a message
2020 received from the command line. Otherwise the child will trigger a lookup
2021 failure and thereby defer the delivery if it tries to use (for example) a
2022 cached ldap connection that the parent has called unbind on.
2024 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2025 of $address_data from the recipient verification was clobbered by the
2026 sender verification.
2028 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2029 was its contents. (It was OK if the option was not defined at all.)
2031 11. A "Completed" log line is now written for messages that are removed from
2032 the spool by the -Mrm option.
2034 12. $host_address is now set to the target address during the checking of
2035 ignore_target_hosts.
2037 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2038 being passed; this would have caused $sender_host_name to have been used if
2039 matching the list had actually called for a host name (not very likely,
2040 since this list is usually IP addresses). A host name is now passed as
2043 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2044 code that specifies a non-restarting handler (typically sigaction() in
2045 modern systems) in an attempt to fix a rare and obscure crash bug.
2047 15. Narrowed the window for a race in the daemon that could cause it to ignore
2048 SIGCHLD signals. This is not a major problem, because they are used only to
2049 wake it up if nothing else does.
2051 16. A malformed maildirsize file could cause Exim to calculate negative values
2052 for the mailbox size or file count. Odd effects could occur as a result.
2053 The maildirsize information is now recalculated if the size or filecount
2056 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2057 support for a long time. Removed HAVE_SYS_VFS_H.
2059 18. Updated exipick to current release from John Jetmore.
2061 19. Allow an empty sender to be matched against a lookup in an address list.
2062 Previously the only cases considered were a regular expression, or an
2065 20. Exim went into a mad DNS lookup loop when doing a callout where the
2066 host was specified on the transport, if the DNS lookup yielded more than
2069 21. The RFC2047 encoding function was originally intended for short strings
2070 such as real names; it was not keeping to the 75-character limit for
2071 encoded words that the RFC imposes. It now respects the limit, and
2072 generates multiple encoded words if necessary. To be on the safe side, I
2073 have increased the buffer size for the ${rfc2047: expansion operator from
2076 22. Failure to deliver a bounce message always caused it to be frozen, even if
2077 there was an errors_to setting on the router. The errors_to setting is now
2080 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2081 canonical form (fully expanded) before being placed in
2082 $sender_host_address.
2084 24. Updated eximstats to version 1.33
2086 25. Include certificate and key file names in error message when GnuTLS fails
2087 to set them up, because the GnuTLS error message doesn't include the name
2088 of the failing file when there is a problem reading it.
2090 26. Expand error message when OpenSSL has problems setting up cert/key files.
2093 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2094 (this can affect the format of dates).
2096 28. exim_tidydb, when checking for the continued existence of a message for
2097 which it has found a message-specific retry record, was not finding
2098 messages that were in split spool directories. Consequently, it was
2099 deleting retry records that should have stayed in existence.
2101 29. eximstats updated to version 1.35
2102 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2103 1.35 - bugfix such that pie charts by volume are generated correctly
2105 30. The SPA authentication driver was not abandoning authentication and moving
2106 on to the next authenticator when an expansion was forced to fail,
2107 contradicting the general specification for all authenticators. Instead it
2108 was generating a temporary error. It now behaves as specified.
2110 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2111 (the order specifies the preference for clients). The order is now AES256,
2112 AES128, 3DES, ARCFOUR128.
2114 31. Small patch to Sieve code - explicitly set From: when generating an
2117 32. Exim crashed if a remote delivery caused a very long error message to be
2118 recorded - for instance if somebody sent an entire SpamAssassin report back
2119 as a large number of 550 error lines. This bug was coincidentally fixed by
2120 increasing the size of one of Exim's internal buffers (big_buffer) that
2121 happened as part of the Exiscan merge. However, to be on the safe side, I
2122 have made the code more robust (and fixed the comments that describe what
2125 33. Some experimental protocols are using DNS PTR records for new purposes. The
2126 keys for these records are domain names, not reversed IP addresses. The
2127 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2128 leaves it alone. Component reversal etc. now happens only for IP addresses.
2131 34. The host_aton() function is supposed to be passed a string that is known
2132 to be a valid IP address. However, in the case of IPv6 addresses, it was
2133 not checking this. This is a hostage to fortune. Exim now panics and dies
2134 if the condition is not met. A case was found where this could be provoked
2135 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2136 components; fortuitously, this particular loophole had already been fixed
2137 by change 4.50/55 or 4.44/33 above.
2139 If there are any other similar loopholes, the new check in host_aton()
2140 itself should stop them being exploited. The report I received stated that
2141 data on the command line could provoke the exploit when Exim was running as
2142 exim, but did not say which command line option was involved. All I could
2143 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2144 running as the user.
2147 35. There was a buffer overflow vulnerability in the SPA authentication code
2148 (which came originally from the Samba project). I have added a test to the
2149 spa_base64_to_bits() function which I hope fixes it.
2152 36. The daemon start-up calls getloadavg() while still root for those OS that
2153 need the first call to be done as root, but it missed one case: when
2154 deliver_queue_load_max is set with deliver_drop_privilege. This is
2155 necessary for the benefit of the queue runner, because there is no re-exec
2156 when deliver_drop_privilege is set.
2158 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2159 was in use, was not putting the data itself into the right store pool;
2160 consequently, it could be overwritten for a subsequent message in the same
2161 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2164 38. Sometimes the final signoff response after QUIT could fail to get
2165 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2166 < 0 before doing a fflush(). This bug looks as though it goes back to the
2167 introduction of TLS in release 3.20, but "sometimes" must have been rare
2168 because the tests only now provoked it.
2174 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2175 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2176 Now it returns an integer. A number of calls were still expecting a T/F
2177 return. Fortuitously, in all cases, the tests worked in OK situations,
2178 which is the norm. However, things would have gone wrong on any write
2179 failures on the smtp file descriptor. This function is used when sending
2180 messages over SMTP and also when doing verify callouts.
2182 2. When Exim is called to do synchronous delivery of a locally submitted
2183 message (the -odf or -odi options), it no longer closes stderr before doing
2186 3. Implemented the mua_wrapper option.
2188 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2190 5. Implemented the functions header_remove(), header_testname(),
2191 header_add_at_position(), and receive_remove_recipient(), and exported them
2194 6. If an ACL "warn" statement specified the addition of headers, Exim already
2195 inserted X-ACL-Warn: at the start if there was no header name. However, it
2196 was not making this test for the second and subsequent header lines if
2197 there were newlines in the string. This meant that an invalid header could
2198 be inserted if Exim was badly configured.
2200 7. Allow an ACL "warn" statement to add header lines at the start or after all
2201 the Received: headers, as well as at the end.
2203 8. Added the rcpt_4xx retry error code.
2205 9. Added postmaster_mailfrom=xxx to callout verification option.
2207 10. Added mailfrom=xxxx to the callout verification option, for verify=
2210 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2211 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2213 12. Inserted some casts to stop certain compilers warning when using pointer
2214 differences as field lengths or precisions in printf-type calls (mostly
2215 affecting debugging statements).
2217 13. Added optional readline() support for -be (dynamically loaded).
2219 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2220 same clock tick as a message's arrival, so that its received time was the
2221 same as the "first fail" time on the retry record, and that message
2222 remained on the queue past the ultimate address timeout, every queue runner
2223 would try a delivery (because it was past the ultimate address timeout) but
2224 after another failure, the ultimate address timeout, which should have then
2225 bounced the address, did not kick in. This was a "< instead of <=" error;
2226 in most cases the first failure would have been in the next clock tick
2227 after the received time, and all would be well.
2229 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2230 being recognized when the domain list was tested by the match_domain
2231 condition in an expansion string.
2233 16. Added the ${str2b64: operator.
2235 17. Exim was always calling setrlimit() to set a large limit for the number of
2236 processes, without checking whether the existing limit was already
2237 adequate. (It did check for the limit on file descriptors.) Furthermore,
2238 errors from getrlimit() and setrlimit() were being ignored. Now they are
2239 logged to the main and panic logs, but Exim does carry on, to try to do its
2240 job under whatever limits there are.
2242 18. Imported PCRE 5.0.
2244 19. Trivial typo in log message " temporarily refused connection" (the leading
2247 20. If the log selector return_path_on_delivery was set and an address was
2248 redirected to /dev/null, the delivery process crashed because it assumed
2249 that a return path would always be set for a "successful" delivery. In this
2250 case, the whole delivery is bypassed as an optimization, and therefore no
2253 21. Internal re-arrangement: the function for sending a challenge and reading
2254 a response while authentication was assuming a zero-terminated challenge
2255 string. It's now changed to take a pointer and a length, to allow for
2256 binary data in such strings.
2258 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2260 23. Exim was not respecting finduser_retries when seeking the login of the
2261 uid under which it was called; it was always trying 10 times. (The default
2262 setting of finduser_retries is zero.) Also, it was sleeping after the final
2263 failure, which is pointless.
2265 24. Implemented tls_on_connect_ports.
2267 25. Implemented acl_smtp_predata.
2269 26. If the domain in control=submission is set empty, Exim assumes that the
2270 authenticated id is a complete email address when it generates From: or
2271 Sender: header lines.
2273 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2274 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2275 chown and chgrp in /bin and hostname in /usr/bin.
2277 28. Exim was keeping the "process log" file open after each use, just as it
2278 does for the main log. This opens the possibility of it remaining open for
2279 long periods when the USR1 signal hits a daemon. Occasional processlog
2280 errors were reported, that could have been caused by this. Anyway, it seems
2281 much more sensible not to leave this file open at all, so that is what now
2284 29. The long-running daemon process does not normally write to the log once it
2285 has entered its main loop, and it closes the log before doing so. This is
2286 so that log files can straightforwardly be renamed and moved. However,
2287 there are a couple of unusual error situations where the daemon does write
2288 log entries, and I had neglected to close the log afterwards.
2290 30. The text of an SMTP error response that was received during a remote
2291 delivery was being truncated at 512 bytes. This is too short for some of
2292 the long messages that one sometimes sees. I've increased the limit to
2295 31. It is now possible to make retry rules that apply only when a message has a
2296 specific sender, in particular, an empty sender.
2298 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2299 it possible to be selective about when SMTP synchronization is enforced.
2301 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2303 32. Implemented hosts_connection_nolog.
2305 33. Added an ACL for QUIT.
2307 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2310 35. Added mailbox_size and mailbox_filecount to appendfile.
2312 36. Added control = no_multiline_responses to ACLs.
2314 37. There was a bug in the logic of the code that waits for the clock to tick
2315 in the case where the clock went backwards by a substantial amount such
2316 that the microsecond fraction of "now" was more than the microsecond
2317 fraction of "then" (but the whole seconds number was less).
2319 38. Added support for the libradius Radius client library this is found on
2320 FreeBSD (previously only the radiusclient library was supported).
2326 1. When certain lookups returned multiple values in the form name=value, the
2327 quoting of the values was not always being done properly. Specifically:
2328 (a) If the value started with a double quote, but contained no whitespace,
2330 (b) If the value contained whitespace other than a space character (i.e.
2331 tabs or newlines or carriage returns) it was not quoted.
2332 This fix has been applied to the mysql and pgsql lookups by writing a
2333 separate quoting function and calling it from the lookup code. The fix
2334 should probably also be applied to nisplus, ibase and oracle lookups, but
2335 since I cannot test any of those, I have not disturbed their existing code.
2337 2. A hit in the callout cache for a specific address caused a log line with no
2338 reason for rejecting RCPT. Now it says "Previous (cached) callout
2339 verification failure".
2341 3. There was an off-by-one bug in the queryprogram router. An over-long
2342 return line was truncated at 256 instead of 255 characters, thereby
2343 overflowing its buffer with the terminating zero. As well as fixing this, I
2344 have increased the buffer size to 1024 (and made a note to document this).
2346 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2347 when Exim is waiting for an SMTP response from a remote server, Exim
2348 restarts its select() call on the socket, thereby resetting its timeout.
2349 This is not a problem when such interrupts are rare. Somebody set up a cron
2350 job to run exiwhat every 2 minutes, which is less than the normal select()
2351 timeout (5 or 10 minutes). This meant that the select() timeout never
2352 kicked in because it was always reset. I have fixed this by comparing the
2353 time when an interrupt arrives with the time at the start of the first call
2354 to select(). If more time than the timeout has elapsed, the interrupt is
2355 treated as a timeout.
2357 5. Some internal re-factoring in preparation for the addition of Sieve
2358 extensions (by MH). In particular, the "personal" test is moved to a
2359 separate function, and given an option for scanning Cc: and Bcc: (which is
2360 not set for Exim filters).
2362 6. When Exim created an email address using the login of the caller as the
2363 local part (e.g. when creating a From: or Sender: header line), it was not
2364 quoting the local part when it contained special characters such as @.
2366 7. Installed new OpenBSD configuration files.
2368 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2369 try to make them clearer.
2371 9. Callout options, other than the timeout value, were being ignored when
2372 verifying sender addresses in header lines. For example, when using
2374 verify = header_sender/callout=no_cache
2376 the cache was (incorrectly) being used.
2378 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2379 only the Cygwin environment.
2381 11. When return_path_on_delivery was set as a log selector, if different remote
2382 addresses in the same message used different return paths and parallel
2383 remote delivery occurred, the wrong values would sometimes be logged.
2384 (Whenever a remote delivery process finished, the return path value from
2385 the most recently started remote delivery process was logged.)
2387 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2388 lines when AUTH and/or TLS are in use. This is the "received protocol"
2389 field. Exim used to use "asmtp" for authenticated SMTP, without any
2390 indication (in the protocol name) for TLS use. Now it follows the RFC and
2391 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2392 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2393 names appear in log lines as well as in Received: header lines.
2395 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2396 extensions, and comparison tests, and to fix some bugs.
2398 14. Changes to the "personal" filter test:
2400 (1) The test was buggy in that it was just doing the equivalent of
2401 "contains" tests on header lines. For example, if a user's address was
2402 anne@some.where, the "personal" test would incorrectly be true for
2404 To: susanne@some.where
2406 This test is now done by extracting each address from the header in turn,
2407 and checking the entire address. Other tests that are part of "personal"
2408 are now done using regular expressions (for example, to check local parts
2409 of addresses in From: header lines).
2411 (2) The list of non-personal local parts in From: addresses has been
2412 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2413 taken from the Sieve specification recommendations.
2415 (3) If the message contains any header line starting with "List-" it is
2416 treated as non-personal.
2418 (4) The test for "circular" in the Subject: header line has been removed
2419 because it now seems ill-conceived.
2421 15. Minor typos in src/EDITME comments corrected.
2423 16. Installed latest exipick from John Jetmore.
2425 17. If headers_add on a router specified a text string that was too long for
2426 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2427 of string_sprintf() is now avoided.
2429 18. $message_body_size was not set (it was always zero) when running the DATA
2430 ACL and the local_scan() function.
2432 19. For the "mail" command in an Exim filter, no default was being set for
2433 the once_repeat time, causing a random time value to be used if "once" was
2434 specified. (If the value happened to be <= 0, no repeat happened.) The
2435 default is now 0s, meaning "never repeat". The "vacation" command was OK
2436 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2437 (I found it when inspecting the code).
2439 20. There is now an overall timeout for performing a callout verification. It
2440 defaults to 4 times the callout timeout, which applies to individual SMTP
2441 commands during the callout. The overall timeout applies when there is more
2442 than one host that can be tried. The timeout is checked before trying the
2443 next host. This prevents very long delays if there are a large number of
2444 hosts and all are timing out (e.g. when the network connections are timing
2445 out). The value of the overall timeout can be changed by specifying an
2446 additional sub-option for "callout", called "maxwait". For example:
2448 verify = sender/callout=5s,maxwait=20s
2450 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2451 to the end before writing, but this should make it even safer).
2453 22. Exim was forgetting that it had advertised PIPELINING for the second and
2454 subsequent messages on an SMTP connection. It was also not resetting its
2455 memory on STARTTLS and an internal HELO.
2457 23. When Exim logs an SMTP synchronization error within a session, it now
2458 records whether PIPELINING has been advertised or not.
2460 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2461 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2462 rather than long int.
2464 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2466 26. Added the never_mail option to autoreply.
2472 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2473 crash if the getsockname() call failed; this can happen if a connection is
2474 closed very soon after it is established. The problem was simply in the
2475 order in which certain operations were done, causing Exim to try to write
2476 to the SMTP stream before it had set up the file descriptor. The bug has
2477 been fixed by making things happen in the correct order.
2483 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2484 before the connection was closed, thus losing the rejection response.
2486 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2487 some early Solaris releases, but causes trouble in current releases where
2488 socklen_t is defined.
2490 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2493 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2496 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2497 that releasing the top of it at the end releases what was used for sub-
2498 expansions (unless the block got too big). However, discard this block if
2499 the first thing is a variable or header, so that we can use its block when
2500 it is dynamic (useful for very large $message_headers, for example).
2502 6. Lookups now cache *every* query, not just the most recent. A new, separate
2503 store pool is used for this. It can be recovered when all lookup caches are
2504 flushed. Lookups now release memory at the end of their result strings.
2505 This has involved some general refactoring of the lookup sources.
2507 7. Some code has been added to the store_xxx() functions to reduce the amount
2508 of flapping under certain conditions.
2510 8. log_incoming_interface used to affect only the <= reception log lines. Now
2511 it causes the local interface and port to be added to several more SMTP log
2512 lines, for example "SMTP connection from", and rejection lines.
2514 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2516 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2518 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2519 could be overwritten at the end of the current message (or the start of a
2520 new message if it was set in a HELO ACL). The value is now preserved for
2521 the duration of the SMTP connection.
2523 12. If a transport had a headers_rewrite setting, and a matching header line
2524 contained an unqualified address, that address was qualified, even if it
2525 did not match any rewriting rules. The underlying bug was that the values
2526 of the flags that permit the existence of unqualified sender and recipient
2527 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2528 non-local messages, and by -bnq for local messages) were not being
2529 preserved with the message after it was received.
2531 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2532 "next input=" as part of the text comprising the host identity instead of
2533 the correct text. The code was using the same buffer for two different
2534 strings. However, depending on which order the printing function evaluated
2535 its arguments, the bug did not always show up. Under Linux, for example, my
2536 test suite worked just fine.
2538 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2539 doesn't work with some older versions of Perl. It has been changed to "my",
2540 which in any case is probably the better facility to use.
2542 15. A really picky compiler found some instances of statements for creating
2543 error messages that either had too many or two few arguments for the format
2546 16. The size of the buffer for calls to the DNS resolver has been increased
2547 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2548 for addresses that have a lot of PTR records. This alleviates a problem; it
2549 does not fully solve it.
2551 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2552 buffer now truncates the list and logs the incident, which is the same
2553 action as happens when Exim is looking up a host name and its aliases.
2554 Previously in this situation something unpredictable would happen;
2555 sometimes it was "internal error: store_reset failed".
2557 18. If a server dropped the connection unexpectedly when an Exim client was
2558 using GnuTLS and trying to read a response, the client delivery process
2559 crashed while trying to generate an error log message.
2561 19. If a "warn" verb in an ACL added multiple headers to a message in a single
2562 string, for example:
2564 warn message = H1: something\nH2: something
2566 the text was added as a single header line from Exim's point of view
2567 though it ended up OK in the delivered message. However, searching for the
2568 second and subsequent header lines using $h_h2: did not work. This has been
2569 fixed. Similarly, if a system filter added multiple headers in this way,
2570 the routers could not see them.
2572 20. Expanded the error message when iplsearch is called with an invalid key to
2573 suggest using net-iplsearch in a host list.
2575 21. When running tests using -bh, any delays imposed by "delay" modifiers in
2576 ACLs are no longer actually imposed (and a message to that effect is
2579 22. If a "gecos" field in a passwd entry contained escaped characters, in
2580 particular, if it contained a \" sequence, Exim got it wrong when building
2581 a From: or a Sender: header from that name. A second bug also caused
2582 incorrect handling when an unquoted " was present following a character
2583 that needed quoting.
2585 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2586 was not being matched caselessly.
2588 24. Arranged for all hyphens in the exim.8 source to be escaped with
2591 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2592 properly. Recipient callout cache records were still being keyed to include
2593 the sender, even when use_sender was set false. This led to far more
2594 callouts that were necessary. The sender is no longer included in the key
2595 when use_sender is false.
2597 26. Added "control = submission" modifier to ACLs.
2599 27. Added the ${base62d: operator to decode base 62 numbers.
2601 28. dnsdb lookups can now access SRV records.
2603 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2604 the configuration file.
2606 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2607 (-v) mode. This makes the output for a verbose queue run more intelligible.
2609 31. Added a use_postmaster feature to recipient callouts.
2611 32. Added the $body_zerocount variable, containing the number of binary zero
2612 bytes in the message body.
2614 33. The time of last modification of the "new" subdirectory is now used as the
2615 "mailbox time last read" when there is a quota error for a maildir
2618 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2620 35. Added +ignore_unknown as a special item in host lists.
2622 36. Code for decoding IPv6 addresses in host lists is now included, even if
2623 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2624 address was recognized as an IP address, but was then not correctly decoded
2625 into binary, causing unexpected and incorrect effects when compared with
2632 1. Very minor rewording of debugging text in manualroute to say "list of
2633 hosts" instead of "hostlist".
2635 2. If verify=header_syntax was set, and a header line with an unqualified
2636 address (no domain) and a large number of spaces between the end of the
2637 name and the colon was received, the reception process suffered a buffer
2638 overflow, and (when I tested it) crashed. This was caused by some obsolete
2639 code that should have been removed. The fix is to remove it!
2641 3. When running in the test harness, delay a bit after writing a bounce
2642 message to get a bit more predictability in the log output.
2644 4. Added a call to search_tidyup() just before forking a reception process. In
2645 theory, someone could use a lookup in the expansion of smtp_accept_max_
2646 per_host which, without the tidyup, could leave open a database connection.
2648 5. Added the variables $recipient_data and $sender_data which get set from a
2649 lookup success in an ACL "recipients" or "senders" condition, or a router
2650 "senders" option, similar to $domain_data and $local_part_data.
2652 6. Moved the writing of debug_print from before to after the "senders" test
2655 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2656 problems for message scanning, either using a data ACL, or using
2657 local_scan() because the Received: header was not generated till after they
2658 were called (in order to set the time as the time of reception completion).
2659 I have revised the way this works. The header is now generated after the
2660 body is received, but before the ACL or local_scan() are called. After they
2661 are run, the timestamp in the header is updated.
2667 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2668 before starting a queue runner without re-exec. This happened only when
2669 deliver_drop_privilege was set or when the Exim user was set to root. The
2670 effect of the bug was that timeouts during subsequent deliveries caused
2671 crashes instead of being properly handled. The handler is now left at its
2672 default (and expected) setting.
2674 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2675 message, again when deliver_drop_privilege is set or Exim is run as root.
2676 The bug described in (1) was not present in this case, but the tidying up
2677 of the other signals was missing. I have made the two cases consistent.
2679 3. The ignore_target_hosts setting on a manualroute router was being ignored
2680 for hosts that were looked up using the /MX notation.
2682 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2685 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2686 operated on the sender address. After changing the $sender_address to <>
2687 for the sender address verify, Exim was re-instated it as the original
2688 (before rewriting) address, but remembering that it had rewritten it, so it
2689 wasn't rewriting it again. This bug also had the effect of breaking the
2690 sender address verification caching when the sender address was rewritten.
2692 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2693 This has been changed so that if the ip literal address matches
2694 ignore_target_hosts, the router declines.
2696 7. Added expansion conditions match_domain, match_address, and match_local_
2697 part (NOT match_host).
2699 8. The placeholder for the Received: header didn't have a length field set.
2701 9. Added code to Exim itself and to exim_lock to test for a specific race
2702 condition that could lead to file corruption when using MBX delivery. The
2703 issue is with the lockfile that is created in /tmp. If this file is removed
2704 after a process has opened it but before that process has acquired a lock,
2705 there is the potential for a second process to recreate the file and also
2706 acquire a lock. This could lead to two Exim processes writing to the file
2707 at the same time. The added code performs the same test as UW imapd; it
2708 checks after acquiring the lock that its file descriptor still refers to
2709 the same named file.
2711 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2712 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2715 11. Added the smtp_active_hostname option. If used, this will typically be made
2716 to depend on the incoming interface address. Because $interface_address is
2717 not set up until the daemon has forked a reception process, error responses
2718 that can happen earlier (such as "too many connections") no longer contain
2721 12. If an expansion in a condition on a "warn" statement fails because a lookup
2722 defers, the "warn" statement is abandoned, and the next ACL statement is
2723 processed. Previously this caused the whole ACL to be aborted.
2725 13. Added the iplsearch lookup type.
2727 14. Added ident_timeout as a log selector.
2729 15. Added tls_certificate_verified as a log selector.
2731 16. Added a global option tls_require_ciphers (compare the smtp transport
2732 option of the same name). This controls incoming TLS connections.
2734 17. I finally figured out how to make tls_require_ciphers do a similar thing
2735 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2736 before starting the TLS session.
2738 18. Tabs are now shown as \t in -bP output.
2740 19. If the log selector return_path_on_delivery was set, Exim crashed when
2741 bouncing a message because it had too many Received: header lines.
2743 20. If two routers both had headers_remove settings, and the first one included
2744 a superfluous trailing colon, the final name in the first list and the
2745 first name in the second list were incorrectly joined into one item (with a
2746 colon in the middle).
2752 1. Added -C and -D options to the exinext utility, mainly to make it easier
2753 to include in the automated testing, but these could be helpful when
2754 multiple configurations are in use.
2756 2. The exinext utility was not formatting the output nicely when there was
2757 an alternate port involved in the retry record key, nor when there was a
2758 message id as well (for retries that were specific to a specific message
2759 and a specific host). It was also confused by IPv6 addresses, because of
2760 the additional colons they contain. I have fixed the IPv4 problem, and
2761 patched it up to do a reasonable job for IPv6.
2763 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2764 delivery, the log line now contains "pipelined" if PIPELINING was used.
2766 4. An SMTP transport process used to panic and die if the bind() call to set
2767 an explicit outgoing interface failed. This has been changed; it is now
2768 treated in the same way as a connect() failure.
2770 5. A reference to $sender_host_name in the part of a conditional expansion
2771 that was being skipped was still causing a DNS lookup. This no longer
2774 6. The def: expansion condition was not recognizing references to header lines
2775 that used bh_ and bheader_.
2777 7. Added the _cache feature to named lists.
2779 8. The code for checking quota_filecount in the appendfile transport was
2780 allowing one more file than it should have been.
2782 9. For compatibility with Sendmail, the command line option
2790 and sets the incoming protocol and host name (for trusted callers). The
2791 host name and its colon can be omitted when only the protocol is to be set.
2792 Note the Exim already has two private options, -pd and -ps, that refer to
2793 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2794 "s", but I don't think that's a major issue.
2796 10. A number of refactoring changes to the code, none of which should affect
2799 (a) The number of logging options was getting close to filling up the
2800 32-bit word that was used as a bit map. I have split them into two classes:
2801 those that are passed in the argument to log_write(), and those that are
2802 only ever tested independently outside of that function. These are now in
2803 separate 32-bit words, so there is plenty of room for expansion again.
2804 There is no change in the user interface or the logging behaviour.
2806 (b) When building, for example, log lines, the code previously used a
2807 macro that called string_cat() twice, in order to add two strings. This is
2808 not really sufficiently general. Furthermore, there was one instance where
2809 it was actually wrong because one of the argument was used twice, and in
2810 one call a function was used. (As it happened, calling the function twice
2811 did not affect the overall behaviour.) The macro has been replaced by a
2812 function that can join an arbitrary number of extra strings onto a growing
2815 (c) The code for expansion conditions now uses a table and a binary chop
2816 instead of a serial search (which was left over from when there were very
2817 few conditions). Also, it now recognizes conditions like "pam" even when
2818 the relevant support is not compiled in: a suitably worded error message is
2819 given if an attempt is made to use such a condition.
2821 11. Added ${time_interval:xxxxx}.
2823 12. A bug was causing one of the ddress fields not to be passed back correctly
2824 from remote delivery subprocesses. The field in question was not being
2825 subsequently used, so this caused to problems in practice.
2827 13. Added new log selectors queue_time and deliver_time.
2829 14. Might have fixed a bug in maildirsizefile handling that threw up
2830 "unexpected character" debug warnings, and recalculated the data
2831 unnecessarily. In any case, I expanded the warning message to give more
2834 15. Added the message "Restricted characters in address" to the statements in
2835 the default ACL that block characters like @ and % in local parts.
2837 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2838 Three changes have been made:
2840 (a) There was a serious bug; a negative response to MAIL caused the whole
2841 recipient domain to be cached as invalid, thereby blocking all messages
2842 to all local parts at the same domain, from all senders. This bug has
2843 been fixed. The domain is no longer cached after a negative response to
2844 MAIL if the sender used is not empty.
2846 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2849 (c) A new callout option, "use_sender" has been added for people who want
2850 the modified behaviour.
2856 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2859 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2860 indeed breaks things for older releases.
2862 3. Added additional logging to the case where there is a problem reading data
2863 from a filter that is running in a subprocess using a pipe, in order to
2864 try to track down a specific problem.
2866 4. Testing facility fudge: when running in the test harness and attempting
2867 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2868 getting "No route to host". Convert this to a timeout.
2870 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2873 6. Some OS don't have socklen_t but use size_t instead. This affects the
2874 fifth argument of getsockopt() amongst other things. This is now
2875 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2876 can be set for individual OS. I have set it for SunOS5, OSF1, and
2877 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2878 some earlier ones do not.
2880 7. Change 4.30/15 was not doing the test caselessly.
2882 8. The standard form for an IPv6 address literal was being rejected by address
2883 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2884 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2885 this, as well as the form without the "IPv6" on the front (but only when
2886 address literals are enabled, of course).
2888 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2890 10. Exim crashed if a message with an empty sender address specified by -f
2891 encountered a router with an errors_to setting. This could be provoked only
2892 by a command such as
2896 where an empty string was supplied; "<>" did not hit this bug.
2898 11. Installed PCRE release 4.5.
2900 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2901 remained set. It is now erased.
2903 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2904 times from message ids (which are base 36 rather than the normal 62).
2906 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2907 were being counted as actual protocol errors, and logged if the log
2908 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2909 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2910 and DATA following a set of rejected RCPTs do not count as protocol errors.
2911 In other words, Exim assumes they were pipelined, though this may not
2912 actually be the case. Of course, in all cases the client gets an
2913 appropriate error code.
2915 15. If a lookup fails in an ACL condition, a message about the failure may
2916 be available; it is used if testing the ACL cannot continue, because most
2917 such messages specify what the cause of the deferral is. However, some
2918 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2919 that caused an old message to be retained and used if a later statement
2920 caused a defer, replacing the real cause of the deferral.
2922 16. If an IP address had so many PTR records that the DNS lookup buffer
2923 was not large enough to hold them, Exim could crash while trying to process
2924 the truncated data. It now detects and logs this case.
2926 17. Further to 4.21/58, another change has been made: if (and only if) the
2927 first line of a message (the first header line) ends with CRLF, a bare LF
2928 in a subsequent header line has a space inserted after it, so as not to
2929 terminate the header.
2931 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2932 unnecessarily, used atoi() instead of strtol(), and didn't check the
2933 termination when getting file sizes from file names by regex.
2935 19. Completely re-implemented the support for maildirsize files, in the light
2936 of a number of problems with the previous contributed implementation
2937 (4.30/29). In particular:
2939 . If the quota is zero, the maildirsize file is maintained, but no quota is
2942 . If the maildir directory does not exist, it is created before any attempt
2943 to write a maildirsize file.
2945 . The quota value in the file is just a cache; if the quota is changed in
2946 the transport, the new value overrides.
2948 . A regular expression is available for excluding directories from the
2951 20. The autoreply transport checks the characters in options that define the
2952 message's headers; it allows continued headers, but it was checking with
2953 isspace() after an embedded newline instead of explicitly looking for a
2956 21. If all the "regular" hosts to which an address was routed had passed their
2957 expiry times, and had not reached their retry times, the address was
2958 bounced, even if fallback hosts were defined. Now Exim should go on to try
2961 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2962 equivalent code in the SMTP transport. Some hosts send humungous responses
2963 to HELO/EHLO, more than 1024 it seems.
2965 23. Refactoring: code in filter.c used (void *) for "any old type" but this
2966 gives compiler warnings in some environments. I've now done it "properly",
2969 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2970 (because of problems with the built-in one) was declared to return uschar *
2971 instead of char *, causing compiler failure.
2973 25. Fixed a file descriptor leak when processing alias/forward files.
2975 26. Fixed a minor format string issue in dbfn.c.
2977 27. Typo in exim.c: ("dmbnz" for "dbmnz").
2979 28. If a filter file refered to $h_xxx or $message_headers, and the headers
2980 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2983 29. When a sender address is verified, it is cached, to save repeating the test
2984 when there is more than one recipient in a message. However, when the
2985 verification involves a callout, it is possible for different callout
2986 options to be set for different recipients. It is too complicated to keep
2987 track of this in the cache, so now Exim always runs a verification when a
2988 callout is required, relying on the callout cache for the optimization.
2989 The overhead is duplication of the address routing, but this should not be
2992 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
2993 to be verified with callout=postmaster, and the main callout worked but the
2994 postmaster check failed, the verification correctly failed. However, if a
2995 subsequent RCPT command asked for sender verification *without* the
2996 postmaster check, incorrect caching caused this verification also to fail,
2999 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3000 it was not caching the DNS options (qualify_single, search_parents) that
3001 were used when the lookup failed. A subsequent lookup with different
3002 options therefore always gave the same answer, though there were cases
3003 where it should not have. (Example: a "domains = !$mx_any" option on a
3004 dnslookup router: the "domains" option is always processed without any
3005 widening, but the router might have qualify_single set.) Now Exim uses the
3006 cached value only when the same options are set.
3008 32. Added John Jetmore's "exipick" utility to the distribution.
3010 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3011 than a timeout (e.g. a certificate is required, and is not provided), an
3012 Exim server now closes the connection immediately. Previously it waited for
3013 the client to close - but if the client is SSL, it seems that they each
3014 wait for each other, leading to a delay before one of them times out.
3016 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3017 maintained 0.8.x compatibility because I don't think many are using it, and
3018 it is clearly obsolete.
3020 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3023 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3024 client certificate was expired. A simple patch fixes this, though I don't
3025 understand the full logic of why the verify callback is called multiple
3028 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3029 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3030 which causes problems with some clients (such as the Certicom SSL Plus
3031 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3032 disables the coutermeasure allowing Eudora to connect."
3034 38. Exim was not checking that a write() to a log file succeeded. This could
3035 lead to Bad Things if a log got too big, in particular if it hit a file
3036 size limit. Exim now panics and dies if it cannot write to a log file, just
3037 as it does if it cannot open a log file.
3039 39. Modified OS/Makefile-Linux so that it now contains
3041 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3043 The two -D definitions ensure that Exim is compiled with large file
3044 support, which makes it possible to handle log files that are bigger than
3047 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3048 instance) a domain was checked against a named list that involved a lookup,
3049 causing $domain_data to be set, then another domain was checked against the
3050 same list, then the first domain was re-checked, the value of $domain_data
3051 after the final check could be wrong. In particular, if the second check
3052 failed, it could be set empty. This bug probably also applied to
3055 41. The strip_trailing_dot option was not being applied to the address given
3056 with the -f command-line option.
3058 42. The code for reading a message's header from the spool was incrementing
3059 $received_count, but never initializing it. This meant that the value was
3060 incorrect (doubled) while delivering a message in the same process in which
3061 it was received. In the most common configuration of Exim, this never
3062 happens - a fresh exec is done - but it can happen when
3063 deliver_drop_privilege is set.
3065 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3066 it now includes up to 150 characters of the unexpected data in the log
3069 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3070 and building data strings. The size of both of these buffers was 10 000
3071 bytes - far larger than anybody would *ever* want, thought I. Needless to
3072 say, somebody hit the limit. I have increased the maximum line length to
3073 20 000 and the maximum data length of concatenated lines to 100 000. I have
3074 also fixed two bugs, because there was no checking on these buffers. Tsk,
3075 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3076 buffer is too small.
3078 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3079 lsearch lookups. Now it does.
3081 46. When parsing a route_list item in a manualroute router, a fixed-length
3082 buffer was used for the list of hosts. I made this 1024 bytes long,
3083 thinking that nobody would ever have a list of hosts that long. Wrong.
3084 Somebody had a whole pile of complicated expansion conditions, and the
3085 string was silently truncated, leading to an expansion error. It turns out
3086 that it is easier to change to an unlimited length (owing to other changes
3087 that have happened since this code was originally written) than to build
3088 structure for giving a limitation error. The length of the item that
3089 expands into the list of hosts is now unlimited.
3091 47. The lsearch lookup could not handle data where the length of text line was
3092 more than 4095 characters. Such lines were truncated, leading to shortened
3093 data being returned. It should now handle lines of any length.
3095 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3096 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3099 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3100 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3101 no longer happens in comment lines. A list of replacements is now placed
3102 at the head of all of the source files, except those whose only change is
3103 to replace PERL_COMMAND in the very first #! line.
3105 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3106 messages on the queue, with a bottom-up merge sort, using code contributed
3107 by Michael Haardt. This should make operations like -bp somewhat faster on
3108 large queues. It won't affect queue runners, except when queue_run_in_order
3111 51. Installed eximstats 1.31 in the distribution.
3113 52. Added support for SRV lookups to the dnslookup router.
3115 53. If an ACL referred to $message_body or $message_body_end, the value was not
3116 reset for any messages that followed in the same SMTP session.
3118 54. The store-handling optimization for building very long strings was not
3119 differentiating between the different store pools. I don't think this
3120 actually made any difference in practice, but I've tidied it.
3122 55. While running the routers to verify a sender address, $sender_address
3123 was still set to the sender address. This is wrong, because when routing to
3124 send a bounce to the sender, it would be empty. Therefore, I have changed
3125 it so that, while verifying a sender address, $sender_address is set to <>.
3126 (There is no change to what happens when verifying a recipient address.)
3128 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3129 target A or AAAA records (if not already returned) without resetting the
3130 qualify_single or search_parents options of the DNS resolver. These are
3131 inappropriate in this case because the targets of MX and SRV records must
3132 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3133 target that, when qualified, matched something in the local domain. These
3134 two options are now turned off when doing these lookups.
3136 57. It seems that at least some releases of Reiserfs (which does not have the
3137 concept of a fixed number of inodes) returns zero and not -1 for the
3138 number of available inodes. This interacted badly with check_spool_inodes,
3139 which assumed that -1 was the "no such thing" setting. What I have done is
3140 to check that the total number of inodes is greater than zero before doing
3141 the test of how many are available.
3143 58. When a "warn" ACL statement has a log_message modifier, the message is
3144 remembered, and not repeated. This is to avoid a lot of repetition when a
3145 message has many recipients that cause the same warning to be written.
3146 Howewer, Exim was preserving the list of already written lines for an
3147 entire SMTP session, which doesn't seem right. The memory is now reset if a
3148 new message is started.
3150 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3151 debugging output unless log_rewrite was also set.
3153 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3154 of (int)(handle) when we know that handle contains (void *)(-1).
3156 61. The Exim daemon panic-logs an error return when it closes the incoming
3157 connection. However "connection reset by peer" seems to be common, and
3158 isn't really an error worthy of noting specially, so that particular error
3161 62. When Exim is trying to find all the local interfaces, it used to panic and
3162 die if the ioctl to get the interface flags failed. However, it seems that
3163 on at least one OS (Solaris 9) it is possible to have an interface that is
3164 included in the list of interfaces, but for which you get a failure error
3165 for this call. This happens when the interface is not "plumbed" into a
3166 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3167 failure of the "get flags" call assumes that the interface is down.
3169 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3170 makes life easier for people who are doing arithmetic on fields extracted
3171 from dates, where you often get leading zeros that should not be
3172 interpreted as octal.
3174 64. Added qualify_domain to the redirect router, to override the global
3177 65. If a pathologically long header line contained very many addresses (the
3178 report of this problem mentioned 10 000) and each of them was rewritten,
3179 Exim could use up a very large amount of memory. (It kept on making new
3180 copies of the header line as it rewrote, and never released the old ones.)
3181 At the expense of a bit more processing, the header rewriting function has
3182 been changed so that it no longer eats memory in this way.
3184 66. The generation of the Received: header has been moved from the time that a
3185 message starts to be received, to the time that it finishes. The timestamp
3186 in the Received: header should now be very close to that of the <= log
3187 line. There are two side-effects of this change:
3189 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3190 logged header lines no longer include the local Received: line, because
3191 it has not yet been created. The same applies to a copy of the message
3192 that is returned to a non-SMTP sender when a message is rejected.
3194 (b) When a filter file is tested using -bf, no additional Received: header
3195 is added to the test message. After some thought, I decided that this
3198 This change does not affect the value of $received_for. It is still set
3199 after address rewriting, but before local_scan() is called.
3201 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3203 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3204 gave an unhelpful panic error message, and a defer error. I have managed to
3205 change this behaviour so that it now rejects any supplied certificate,
3206 which seems right, as the list of acceptable certificates is empty.
3208 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3209 gave an unhelpful defer error. I have not managed to make this reject any
3210 supplied certificates, but the error message it gives is "no certificate
3211 supplied", which is not helpful.
3213 70. exigrep's output now also includes lines that are not associated with any
3214 message, but which match the given pattern. Implemented by a patch from
3215 Martin Sluka, which also tidied up the Perl a bit.
3217 71. Recipient callout verification, like sender verification, was using <> in
3218 the MAIL FROM command. This isn't really the right thing, since the actual
3219 sender may affect whether the remote host accepts the recipient or not. I
3220 have changed it to use the actual sender in the callout; this means that
3221 the cache record is now keyed on a recipient/sender pair, not just the
3222 recipient address. There doesn't seem to be a real danger of callout loops,
3223 since a callout by the remote host to check the sender would use <>.
3224 [SEE ABOVE: changed after hitting problems.]
3226 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3227 temporary errors. However, in the case of such a code being given after
3228 the end of a data transmission (i.e. after ".") Exim was failing to write
3229 a retry record for the message. (Yes, there was some broken host that was
3230 actually sending 8xx at this point.)
3232 73. An unknown lookup type in a host list could cause Exim to panic-die when
3233 the list was checked. (An example that provoked this was putting <; in the
3234 middle of a list instead of at the start.) If this happened during a DATA
3235 ACL check, a -D file could be left lying around. This kind of configuration
3236 error no longer causes Exim to die; instead it causes a defer errror. The
3237 incident is still logged to the main and panic logs.
3239 74. Buglet left over from Exim 3 conversion. The message "too many messages
3240 in one connection" was written to the rejectlog but not the mainlog, except
3241 when address rewriting (yes!) was being logged.
3243 75. Added write_rejectlog option.
3245 76. When a system filter was run not as root (that is, when system_filter_user
3246 was set), the values of the $n variables were not being returned to the
3247 main process; thus, they were not subsequently available in the $sn
3250 77. Added +return_path_on_delivery log selector.
3252 78. A connection timeout was being treated differently from recipients deferred
3253 when testing hosts_max_try with a message that was older than the host's
3254 retry timeout. (The host should not be counted, thus allowing all hosts to
3255 be tried at least once before bouncing.) This may have been the cause of an
3256 occasionally reported bug whereby a message would remain on the queue
3257 longer than the retry timeout, but would be bounced if a delivery was
3258 forced. I say "may" because I never totally pinned down the problem;
3259 setting up timeout/retry tests is difficult. See also the next item.
3261 79. The ultimate address timeout was not being applied to errors that involved
3262 a combination of host plus message (for example, a timeout on a MAIL
3263 command). When an address resolved to a number of possible hosts, and they
3264 were not all tried for each delivery (e.g. because of hosts_max_try), a
3265 message could remain on the queue longer than the retry timeout.
3267 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3270 81. Fixed an obscure SMTP outgoing bug which required at least the following
3271 conditions: (a) there was another message waiting for the same server;
3272 (b) the server returned 5xx to all RCPT commands in the first message so
3273 that the message was not completed; (c) the server dropped the connection
3274 or gave a negative response to the RSET that Exim sends to abort the
3275 transaction. The observed case was a dropped connection after DATA that had
3276 been sent in pipelining mode. That is, the server had advertised PIPELINING
3277 but was not implementing it correctly. The effect of the bug was incorrect
3278 behaviour, such as trying another host, and this could lead to a crash.
3284 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3285 and daemon.c were passed as pointers to ints; they should have been
3286 pointers to socklen_t variables (which are typically unsigned ints).
3288 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3291 3. Fixed a really odd bug that affected only the testing scheme; patching a
3292 certain fixed string in the binary changed the value of another string that
3293 happened to be identical to the end of the original first string.
3295 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3296 name", it returns that address as the IP address. On some operating
3297 systems (e.g. Solaris), it also passes back the IP address string as the
3298 "host name". However, on others (e.g. Linux), it passes back an empty
3299 string. Exim wasn't checking for this, and was changing the host name to an
3300 empty string, assuming it had been canonicized.
3302 5. Although rare, it is permitted to have more than one PTR record for a given
3303 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3304 all the names associated with an address, because they do in Solaris.
3305 However, it seems that they do not in Linux for data that comes from the
3306 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3307 I found this out when I moved to a new Linux workstation and tried to run
3308 the Exim test suite.
3310 To get round this problem I have changed the code so that it now does its
3311 own call to the DNS to look up PTR records when searching for a host name.
3312 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3313 addresses that are only in /etc/hosts are still found.
3315 This behaviour is, however, controlled by an option called host_lookup_
3316 order, which defaults to "bydns:byaddr". If people want to use the other
3317 order, or indeed, just use one or the other means of lookup, they can
3318 specify it in this variable.
3320 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3321 some operating systems, this comes back from gethostbyaddr() as an empty
3322 string, and this is what Exim used to test for. However, it seems that in
3323 other systems, "." is yielded. Exim now tests for this case too.
3325 7. The values of check_spool_space and check_log_space are now held internally
3326 as a number of kilobytes instead of an absolute number of bytes. If a
3327 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3328 kilobyte. This means that much larger values can be stored.
3330 8. Exim monitor: an attempt to get the action menu when not actually pointing
3331 at a message produces an empty menu entitled "No message selected". This
3332 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3333 no entries in it ("Shell widget menu has zero width and/or height"). So I
3334 have added a single, blank menu entry in this case.
3336 9. Added ${quote_local_part.
3338 10. MIME decoding is now applied to the contents of Subject: header lines when
3341 11. Now that a reference to $sender_host_address automatically causes a reverse
3342 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3343 host lookup before query-style lookups in lists that might use this
3344 variable. This has therefore been abolished, and the "net-" prefix is no
3345 longer necessary for query-style lookups.
3347 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3348 have been a typo for LFLAGS, so it has been changed.
3350 13. The install script calls Exim with "-C /dev/null" in order to find the
3351 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3352 to be output. Howeve, since Exim outputs its version number before the
3353 error, it didn't break the script. It just looked ugly. I fixed this by
3354 always allowing "-C /dev/null" if the caller is root.
3356 14. Ignore overlarge ACL variable number when reading spool file - insurance
3357 against a later release with more variables having written the file.
3359 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3360 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3361 this, as well as the form without the "IPv6" on the front.
3363 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3364 OS/Makefile-Darwin file.
3366 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3367 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3369 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3370 variable. However, some OS use ints for inodes. I've added cast to long int
3371 to get rid of the compiler warning.
3373 19. I had forgotten to lock out "/../" in configuration file names when
3374 ALT_CONFIG_PREFIX was set.
3376 20. Routers used for verification do not need to specify transports. However,
3377 if such a router generated a host list, and callout was configured, Exim
3378 crashed, because it could not find a port number from the (non-existent)
3379 transport. It now assumes port 25 in this circumstance.
3381 21. Added the -t option to exigrep.
3383 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3384 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3385 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3386 them is set, LOOKUP_LSEARCH is forced.
3388 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3390 24. Added sender and host information to the "rejected by local_scan()" log
3391 line; previously there was no indication of these.
3393 25. Added .include_if_exists.
3395 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3396 Linux. It turns out that not all file systems support this. Apparently some
3397 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3398 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3399 not supported on the file descriptor, is now ignored when Exim is trying to
3400 sync a directory. This applies only to Linux.
3402 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3404 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3405 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3408 29. Added support for maildirsize files from supplied patch (modified a bit).
3410 30. The use of :fail: followed by an empty string could lead Exim to respond to
3411 sender verification failures with (e.g.):
3413 550 Verification failed for <xxx>
3414 550 Sender verify failed
3416 where the first response line was missing the '-' that indicates it is not
3417 the final line of the response.
3419 31. The loop for finding the name of the user that called Exim had a hardwired
3420 limit of 10; it now uses the value of finduser_retries, which is used for
3421 all other user lookups.
3423 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3426 33. Exim was neglecting to zero errno before one call of strtol() when
3427 expanding a string and expecting an integer value. On some systems this
3428 resulted in spurious "integer overflow" errors. Also, it was casting the
3429 result into an int without checking.
3431 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3432 did not work. The code looks as if it has *never* worked, though it appears
3433 to have been documented since at least releast 1.62. I have made it work.
3435 35. The "timeout_DNS" error in retry rules, also documented since at least
3436 1.62, also never worked. As it isn't clear exactly what this means, and
3437 clearly it isn't a major issue, I have abolished the feature by treating it
3438 as "timeout", and writing a warning to the main and panic logs.
3440 36. The display of retry rules for -brt wasn't always showing the error code
3443 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3444 timeout_connect_A, timeout_connect_MX.
3446 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3447 to the empty sender.
3449 39. The daemon was not analysing the content of -oX till after it had closed
3450 stderr and disconnected from the controlling terminal. This meant that any
3451 syntax errors were only noted on the panic log, and the return code from
3452 the command was 0. By re-arranging the code a little, I've made the
3453 decoding happen first, so such errors now appear on stderr, and the return
3454 code is 1. However, the actual setting up of the sockets still happens in
3455 the disconnected process, so errors there are still only recorded on the
3458 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3459 connections (as happens on some IP stacks) was logged at start up time as
3460 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3461 it from "IPv6 and IPv4", which means that two separate sockets are being
3464 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3465 says whether AF_INET or AF_INET6 was passed as an argument.
3467 42. Exiwhat output was messed up when time zones were included in log
3470 43. Exiwhat now gives more information about the daemon's listening ports,
3471 and whether -tls-on-connect was used.
3473 44. The "port" option of the smtp transport is now expanded.
3475 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3476 silently ignored. Now an error message is written to the main and panic
3479 46. There's a new ACL modifier called "logwrite" which writes to a log file
3480 as soon as it is encountered.
3482 47. Added $local_user_uid and $local_user_gid at routing time.
3484 48. Exim crashed when trying to verify a sender address that was being
3487 49. Exim was recognizing only a space character after ".include". It now also
3488 recognizes a tab character.
3490 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3491 extracting the relevant information from the specification. The man page no
3492 longer contains scrambled data for the -d option, and I've added a section
3493 at the front about calling Exim under different names.
3495 51. Added "extra_headers" argument to the "mail" command in filter files.
3497 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3500 53. Installed eximstats 1.29.
3502 54. Added transport_filter_timeout as a generic transport option.
3504 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3505 Cc: header lines. This was required by RFC 822, but it not required by RFC
3508 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3509 incoming messages that did not have them. Now it does so only if the
3510 message originates locally, that is, if there is no associated remote host
3511 address. When Resent- header lines are present, this applies to the Resent-
3512 lines rather than the non-Resent- lines.
3514 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3515 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3517 58. Messages for configuration errors now include the name of the main
3518 configuration file - useful now that there may be more than one file in a
3519 list (.included file names were always shown).
3521 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3522 for those rare installations that do not start the daemon as root or run it
3523 setuid root. I've cut out the call to initgroups() if the daemon is not
3526 60. The Exim user and group can now be bound into the binary as text strings
3527 that are looked up at the start of Exim's processing.
3529 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3531 62. Added $mailstore_basename variable.
3533 63. Installed patch to sieve.c from Michael Haardt.
3535 64. When Exim failed to open the panic log after failing to open the main log,
3536 the original message it was trying to log was written to stderr and debug
3537 output, but if they were not available (the usual case in production), it
3538 was lost. Now it is written to syslog before the two lines that record the
3539 failures to open the logs.
3541 65. Users' Exim filters run in subprocesses under the user's uid. It is
3542 possible for a "deliver" command or an alias in a "personal" command to
3543 provoke an address rewrite. If logging of address rewriting is configured,
3544 this fails because the process is not running as root or exim. There may be
3545 a better way of dealing with this, but for the moment (because 4.30 needs
3546 to be released), I have disabled address rewrite logging when running a
3547 filter in a non-root, non-exim process.
3553 1. The buildconfig auxiliary program wasn't quoting the value set for
3554 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3555 not defined. This bug was masked in 4.22 by the effect that was fixed in
3558 2. Some messages that were rejected after a message id was allocated were
3559 shown as "incomplete" by exigrep. It no longer does this for messages that
3560 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3562 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3563 have allow_domain_literals set, the ID did not get logged in the <= line.
3564 Domain literals are now always recognized in Message-ID: header lines.
3566 4. The first argument for a ${extract expansion item is the key name or field
3567 number. Leading and trailing spaces in this item were not being ignored,
3568 causing some misleading effects.
3570 5. When deliver_drop_privilege was set, single queue runner processes started
3571 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3572 same command in the process it spins off) were not dropping privilege.
3574 6. When the daemon running as "exim" started a queue runner, it always
3575 re-executed Exim in the spun-off process. This is a waste of effort when
3576 deliver_drop_privilege is set. The new process now just calls the
3577 queue-runner function directly.
3583 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3586 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3587 ignored. Though the use of -oP was forcing the writing of a pid file, it
3588 was always written to the default place.
3590 3. If the message "no IP address found for host xxxx" is generated during
3591 incoming verification, it is now followed by identification of the incoming
3592 connection (so you can more easily find what provoked it).
3594 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3596 5. Added some features to "harden" Exim a bit more against certain attacks:
3598 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3599 be put in Local/Makefile. This is like the never_users runtime option,
3600 but it cannot be overridden. The default setting is "root".
3602 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3603 prefix string with which any file named in a -C command line option
3606 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3607 is retained for -C and -D only if the caller of Exim is root. Without
3608 it, the exim user may also use -C and -D and retain privilege.
3610 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3611 command line option is disabled.
3613 6. Macro names set by the -D option must start with an upper case letter, just
3614 like macro names defined in the configuration file.
3616 7. Added "dereference=" facility to LDAP.
3618 8. Two instances of the typo "uknown" in the source files are fixed.
3620 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3621 the Configure-Makefile script screwed up while processing it.
3623 10. Incorporated PCRE 4.4.
3625 11. The SMTP synchronization check was not operating right at the start of an
3626 SMTP session. For example, it could not catch a HELO sent before the client
3627 waited for the greeting. There is now a check for outstanding input at the
3628 point when the greeting is written. Because of the duplex, asynchronous
3629 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3630 way, but not yet received, when the check is performed.
3632 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3633 on TCP/IP sockets, because this apparently causes some broken clients to
3636 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3637 unchanged) from the Cygwin maintainer.
3639 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3640 was supported instead of testing for mbx. Effectively a typo.
3642 15. The spa authenticator server code was not checking that the input it
3643 received was valid base64.
3645 16. The debug output line for the "set" modifier in ACLs was not showing the
3646 name of the variable that was being set.
3648 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3650 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3651 lookup on its own. Something else was needed to trigger the lookup. For
3652 example, a match in host_lookup or the need for a host name in a host list.
3653 Now, if $sender_host_name is referenced and the host name has not yet been
3654 looked up, a lookup is performed. If the lookup fails, the variable remains
3655 empty, and $host_lookup_failed is set to "1".
3657 19. Added "eqi" as a case-independent comparison operator.
3659 20. The saslauthd authentication condition could segfault if neither service
3660 nor realm was specified.
3662 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3663 error message that was logged was misleading, and incoming SMTP
3664 connections were dropped. The message is now more accurate, and temporary
3665 errors are given to SMTP connections.
3667 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3668 (except RSET) with a 421 error, until QUIT is received. However, it was
3669 failing to send a response to QUIT.
3671 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3672 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3673 ACL is now run after the helo_try_verify_hosts code.
3675 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3676 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3677 case-independent, so other case variants are also recognized). Apparently
3678 some systems use these upper case variants.
3680 25. If more than two messages were waiting for the same host, and a transport
3681 filter was specified for the transport, Exim sent two messages over the
3682 same TCP/IP connection, and then failed with "socket operation on non-
3683 socket" when it tried to send the third.
3685 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3687 27. The extern definition of crypt16() in expand.c was not being excluded when
3688 the OS had its own crypt16() function.
3690 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3691 as a preferred synonym for return_size_limit, both as an option and as an
3694 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3696 30. Changed the default configuration ACL to relax the local part checking rule
3697 for addresses that are not in any local domains. For these addresses,
3698 slashes and pipe symbols are allowed within local parts, but the sequence
3699 /../ is explicitly forbidden.
3701 31. SPA server authentication was not clearing the challenge buffer before
3704 32. log_message in a "warn" ACL statement was writing to the reject log as
3705 well as to the main log, which contradicts the documentation and doesn't
3706 seem right (because no rejection is happening). So I have stopped it.
3708 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3709 However, I am unable to do any testing of this.
3711 34. Fixed an infelicity in the appendfile transport. When checking directories
3712 for a mailbox, to see if any needed to be created, it was accidentally
3713 using path names with one or more superfluous leading slashes; tracing
3714 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3716 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3717 contents are added to the log line that is written for every discarded
3718 recipient. (Previously a log_message setting was ignored.)
3720 36. The ${quote: operator now quotes the string if it is empty.
3722 37. The install script runs exim in order to find its version number. If for
3723 some reason other than non-existence or emptiness, which it checks, it
3724 could not run './exim', it was installing it with an empty version number,
3725 i.e. as "exim-". This error state is now caught, and the installation is
3728 38. An argument was missing from the function that creates an error message
3729 when Exim fails to connect to the socket for saslauthd authentication.
3730 This could cause Exim to crash, or give a corrupted message.
3732 39. Added isip, isip4, and isip6 to ${if conditions.
3734 40. The ACL variables $acl_xx are now saved with the message, and can be
3735 accessed later in routers, transports, and filters.
3737 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3738 strings in the file are not string-expanded.
3740 42. If a MAIL command specified a SIZE value that was too large to fit into an
3741 int variable, the check against message_size_limit failed. Such values are
3742 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3743 day this will have to be increased, but I don't think I want to be around
3744 when emails are that large.
3751 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3752 iconv() is not standard in FreeBSD.
3754 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3755 IPv6 enabled. The observed symptom was a segmentation fault on return from
3756 the function os_common_find_running_interfaces() in src/os.c.
3758 3. In the check_special_case() function in daemon.c I had used "errno" as an
3759 argument name, which causes warnings on some systems. This was basically a
3760 typo, since it was named "eno" in the comments!
3762 4. The code that waits for the clock to tick (at a resolution of some fraction
3763 of a second) so as to ensure message-id uniqueness was always waiting for
3764 at least one whole tick, when it could have waited for less. [This is
3765 almost certainly not relevant at current processor speeds, where it is
3766 unlikely to ever wait at all. But we try to future-proof.]
3768 5. The function that sleeps for a time interval that includes fractions of a
3769 second contained a race. It did not block SIGALRM between setting the
3770 timer, and suspending (a couple of lines later). If the interval was short
3771 and the sigsuspend() was delayed until after it had expired, the suspension
3772 never ended. On busy systems this could lead to processes getting stuck for
3775 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3776 process, before it forks any delivery processes. The open lookup caching
3777 mechanism meant that the open file or database connection was passed into
3778 the delivery process. The problem was that delivery processes always tidy
3779 up cached lookup data. This could cause a problem for the next delivery
3780 process started by the queue runner, because the external queue runner
3781 process does not know about the closure. So the next delivery process
3782 still has data in the lookup cache. In the case of a file lookup, there was
3783 no problem because closing a file descriptor in a subprocess doesn't affect
3784 the parent. However, if the lookup was caching a connection to a database,
3785 the connection was closed, and the second delivery process was likely to
3786 see errors such as "PGSQL: query failed: server closed the connection
3787 unexpectedly". The problem has been fixed by closing all cached lookups
3788 in a queue runner before running a delivery process.
3790 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3791 seem to have the "const" qualifier which it has on other OS. I've
3794 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3795 *of the same type* (client or server) with the same public name that an
3796 error should be diagnosed.
3798 9. When Exim looked up a host name for an IP address, but failed to find the
3799 original IP address when looking up the host name (a safety check), it
3800 output the message "<ip address> does not match any IP for NULL", which was
3801 confusing, to say the least. The bug was that the host name should have
3802 appeared instead of "NULL".
3804 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3805 user that is built into the binary, and the -C or -D options is used, root
3806 privilege is dropped before the configuration file is read. In addition,
3807 logging is switched to stderr instead of the normal log files. If the
3808 configuration then re-defines the Exim user, the unprivileged environment
3809 is probably not what is expected, so Exim logs a panic warning message (but
3812 However, if deliver_drop_privilege is set, the unprivileged state may well
3813 be exactly what is intended, so the warning has been cut out in that case,
3814 and Exim is allowed to try to write to its normal log files.
3820 1. smtp_return_error_details was not giving details for temporary sender
3821 or receiver verification errors.
3823 2. Diagnose a configuration error if two authenticators have the same public
3826 3. Exim used not to create the message log file for a message until the first
3827 delivery attempt. This could be confusing when incoming messages were held
3828 for policy or load reasons. The message log file is now created at the time
3829 the message is received, and an initial "Received" line is written to it.
3831 4. The automatically generated man page for command line options had a minor
3832 bug that caused no ill effects; however, a more serious problem was that
3833 the procedure for building the man page automatically didn't always
3834 operate. Consequently, release 4.20 contains an out-of-date version. This
3835 shouldn't happen again.
3837 5. When building Exim with embedded Perl support, the script that builds the
3838 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3839 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3841 6. The freeze_tell option was not being used for messages that were frozen on
3842 arrival, either by an ACL or by local_scan().
3844 7. Added the smtp_incomplete_transaction log selector.
3846 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3847 was accepting AUTH without a new EHLO.
3849 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3850 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3853 10. Logging of TCP/IP connections (when configured) now happens in the main
3854 daemon process instead of the child process, so that the TCP/IP connection
3855 count is more accurate (but it can never be perfect).
3857 11. The use of "drop" in a nested ACL was not being handled correctly in the
3858 outer ACL. Now, if condition failure induced by the nested "drop" causes
3859 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3860 or "require"), the connection is dropped.
3862 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3863 that yield "discard" can now be used with an "accept" or a "discard" verb,
3864 but an error is generated for any others (because I can't see a useful way
3865 to define what should happen).
3867 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3868 are now processed in the same way as lines in the Exim configuration file.
3869 In particular, continuation lines are supported.
3871 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3873 15. Added -ti meaning -t -i.
3875 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3876 domains, and warn by logging if others are found.
3878 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3879 returned from GIFCONF when Exim is trying to find the list of interfaces on
3880 a host. The code in os.c has been modified to copy each ifreq to an aligned
3881 structure in all cases.
3883 Also, in some cases, the returned ifreq's were being copied to a 'struct
3884 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3885 means the last couple of bytes of an IPv6 address could be chopped if the
3886 ifreq contained only a normal sockaddr (14 bytes storage).
3888 18. Named domain lists were not supported in the hosts_treat_as_local option.
3889 An entry such as +xxxx was not recognized, and was treated as a literal
3892 19. Ensure that header lines added by a DATA ACL are included in the reject log
3893 if the ACL subsequently rejects the message.
3895 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3896 MD5 (which is deprecated).
3898 21. When testing a filter file using -bf, Exim was writing a message when it
3899 took the sender from a "From " line in the message, but it was not doing so
3900 when it took $return_path from a Return-Path: header line. It now does.
3902 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3903 with a valid header line field name (a series of printing characters
3904 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3906 23. Changed "disc" in the source to "disk" to conform to the documentation and
3907 the book and for uniformity.
3909 24. Ignore Sendmail's -Ooption=value command line item.
3911 25. When execve() failed while trying to run a command in a pipe transport,
3912 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3913 could be confused with a return value of 69 from the command itself. This
3914 has been changed to 127, the value the shell returns if it is asked to run
3915 a non-existent command. The wording for the related log line suggests a
3916 non-existent command as the problem.
3918 26. If received_header_text expands to an empty string, do not add a Received:
3919 header line to the message. (Well, it adds a token one on the spool, but
3920 marks it "old" so that it doesn't get used or transmitted.)
3922 27. Installed eximstats 1.28 (addition of -nt option).
3924 28. There was no check for failure on the call to getsockname() in the daemon
3925 code. This can fail if there is a shortage of resources on the system, with
3926 ENOMEM, for example. A temporary error is now given on failure.
3928 29. Contrary to the C standard, it seems that in some environments, the
3929 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3930 program. Exim now does this explicitly; it affects the formatting of
3931 timestamps using strftime().
3933 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3934 complaints. I've now initialized all the variables, to avoid this.
3936 32. Header lines added by a system filter were not being "seen" during
3937 transport-time rewrites.
3939 33. The info_callback() function passed to OpenSSL is set up with type void
3940 (*)(SSL *, int, int), as described somewhere. However, when calling the
3941 function (actually a macro) that sets it up, the type void(*)() is
3942 expected. I've put in a cast to prevent warnings from picky compilers.
3944 34. If a DNS black list lookup found a CNAME record, but there were no A
3945 records associated with the domain it pointed at, Exim crashed.
3947 35. If a DNS black list lookup returned more than one A record, Exim ignored
3948 all but the first. It now scans all returned addresses if a particular IP
3949 value is being sought. In this situation, the contents of the
3950 $dnslist_value variable are a list of all the addresses, separated by a
3953 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3954 to accept a host name and all its aliases if the forward lookup for any of
3955 them yielded the IP address of the incoming connection. Now it accepts only
3956 those names whose forward lookup yields the correct IP address. Any other
3957 names are discarded. This closes a loophole whereby a rogue DNS
3958 administrator could create reverse DNS records to break through a
3959 wildcarded host restriction in an ACL.
3961 37. If a user filter or a system filter that ran in a subprocess used any of
3962 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3963 the wrong values were passed to the pipe command ($thisaddress had the
3964 value of $0, $0 had the value of $1, etc). This bug was introduced by
3965 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3967 38. Improved the line breaking for long SMTP error messages from ACLs.
3968 Previously, if there was no break point between 40 and 75 characters, Exim
3969 left the rest of the message alone. Two changes have been made: (a) I've
3970 reduced the minimum length to 35 characters; (b) if it can't find a break
3971 point between 35 and 75 characters, it looks ahead and uses the first one
3972 that it finds. This may give the occasional overlong line, but at least the
3973 remaining text gets split now.
3975 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3976 file descriptors might be low, and that setting 1000 would always raise it.
3977 It turns out that in some environments, the limit is already over 1000 and
3978 that lowering it causes trouble. So now Exim takes care not to decrease it.
3980 40. When delivering a message, the value of $return_path is set to $sender_
3981 address at the start of routing (routers may change the value). By an
3982 oversight, this default was not being set up when an address was tested by
3983 -bt or -bv, which affected the outcome if any router or filter referred to
3986 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
3987 log only when "message" or "log_message" are set. However, if one of the
3988 conditions was an address verification, or a call to a nested ACL, the
3989 messages generated by the underlying test were being passed through. This
3990 no longer happens. The underlying message is available in $acl_verify_
3991 message for both "message" and "log_message" expansions, so it can be
3992 passed through if needed.
3994 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3995 new expansion $bh_ to give the encoded byte string without charset
3996 translation. Translation happens only if iconv() is available; HAVE_ICONV
3997 indicates this at build time. HEADERS_CHARSET gives the charset to
3998 translate to; headers_charset can change it in the configuration, and
3999 "headers charset" can change it in an individual filter file.
4001 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4002 that creates RFC 2047 encoded "words" labels them as that charset instead
4003 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4004 expansion operator; (ii) when Exim creates a From: line for a local
4005 message; (iii) when a header line is rewritten to include a "phrase" part.
4007 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4008 buggy, causing it to skip the first lines of messages whose message ID
4009 ended in 'D'. This would not have bitten before Exim release 4.14, because
4010 message IDs were unlikely to end in 'D' before then. The effect was to have
4011 incorrect size information for certain domains.
4013 45. #include "config.h" was missing at the start of the crypt16.c module. This
4014 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4017 46. If there was a timeout during a "random" callout check, Exim treated it as
4018 a failure of the random address, and carried on sending RSET and the real
4019 address. If the delay was just some slowness somewhere, the response to the
4020 original RCPT would be taken as a response to RSET and so on, causing
4021 mayhem of various kinds.
4023 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4024 when I implemented it. It didn't allow for the fact that some option values
4025 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4026 the right test for positive values.
4028 48. Domain names in DNS records are case-independent. Exim always looks them up
4029 in lower case. Some resolvers return domain names in exactly the case they
4030 appear in the zone file, that is, they may contain uppercase letters. Not
4031 all resolvers do this - some return always lower case. Exim was treating a
4032 change of case by a resolver as a change of domain, similar to a widening
4033 of a domain abbreviation. This triggered its re-routing code and so it was
4034 trying to route what was effectively the same domain again. This normally
4035 caused routing to fail (because the router wouldn't handle the domain
4036 twice). Now Exim checks for this case specially, and just changes the
4037 casing of the domain that it ultimately uses when it transmits the message
4040 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4043 50. If a filter generated a file delivery with a non-absolute name (possible if
4044 no home directory exists for the router), the forbid_file option was not
4047 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4048 the existing equality matching.
4050 52. Exim was using ints instead of ino_t variables in some places where it was
4051 dealing with inode numbers.
4053 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4054 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4055 called TMPDIR, and if it finds it is different, it changes its value.
4057 54. The smtp_printf() function is now made available to local_scan() so
4058 additional output lines can be written before returning. There is also an
4059 smtp_fflush() function to enable the detection of a dropped connection.
4060 The variables smtp_input and smtp_batched_input are exported to
4063 55. Changed the default runtime configuration: the message "Unknown user"
4064 has been removed from the ACL, and instead placed on the localuser router,
4065 using the cannot_route_message feature. This means that any verification
4066 failures that generate their own messages won't get overridden. Similarly,
4067 the "Unrouteable address" message that was in the ACL for unverifiable
4068 relay addresses has also been removed.
4070 56. Added hosts_avoid_esmtp to the smtp transport.
4072 57. The exicyclog script was not checking for the esoteric option
4073 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4074 will work only if exicyclog is run under the appropriate euid.
4076 58. Following a discussion on the list, the rules by which Exim recognises line
4077 endings on incoming messages have been changed. The -dropcr and drop_cr
4078 options are now no-ops, retained only for backwards compatibility. The
4079 following line terminators are recognized: LF CRLF CR. However, special
4080 processing applies to CR:
4082 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4083 nor a local message in the state where . is a terminator.
4085 (ii) If a bare CR is encountered in a header line, an extra space is added
4086 after the line terminator so as not to end the header. The reasoning
4087 behind this is that bare CRs in header lines are most likely either
4088 to be mistakes, or people trying to play silly games.
4090 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4091 was being incorrectly given as 18 bytes larger than it should have been.
4092 This is a VOB (very old bug).
4094 60. This may never have affected anything current, but just in case it has:
4095 When the local host is found other than at the start of a list of hosts,
4096 the local host, those with the same MX, and any that follow, are discarded.
4097 When the list in question was part of a longer list of hosts, the following
4098 hosts (not currently being processed) were also being discarded. This no
4099 longer happens. I'm not sure if this situation could ever has previously
4102 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4105 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4106 header. This is something that is recommended in a new Internet Draft, and
4107 is something that is documented as being done by Sendmail. There are two
4108 possible values. For messages generated by the autoreply transport, Exim
4111 Auto-Submitted: auto-replied
4113 whereas for all other generated messages (e.g. bounces) it adds
4115 Auto-Submitted: auto-generated
4117 63. The "personal" condition in filters now includes a test for the
4118 Auto-Submitted: header. If it contains the string "auto-" the message it
4119 not considered personal.
4121 64. Added rcpt_include_affixes as a generic transport option.
4123 65. Added queue_only_override (default true).
4125 66. Added the syslog_duplication option.
4127 67. If what should have been the first header line of a message consisted of
4128 a space followed by a colon, Exim was mis-interpreting it as a header line.
4129 It isn't of course - it is syntactically invalid and should therefore be
4130 treated as the start of the message body. The misbehaviour could have
4131 caused a number of strange effects, including loss of data in subsequent
4132 header lines, and spool format errors.
4134 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4135 client host had authenticated. This control can now be exercised by an ACL
4136 for more flexibility.
4138 69. By default, callouts do not happen when testing with -bh. There is now a
4139 variant, -bhc, which does actually run the callout code, including
4140 consulting and updating the callout cache.
4142 70. Added support for saslauthd authentication, courtesy of Alexander
4145 71. If statvfs() failed on the spool or log directories while checking their
4146 size for availability, Exim confusingly gave the error "space shortage".
4147 Furthermore, in debugging mode it crashed with a floating point exception.
4148 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4149 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4150 serious problem, Exim now writes to the main and panic logs when this
4151 happens, with details of the failure. It then refuses to accept the
4152 incoming message, giving the message "spool directory problem" or "log
4153 directory problem" with a 421 code for SMTP messages.
4155 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4156 0, 1, and 2 exist, because some OS complain for execs without them (see
4157 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4158 descriptors. However, the code omitted to check that the open succeeded,
4159 causing mysterious errors if for some reason the permissions on /dev/null
4160 got screwed. Now Exim writes a message to the main and panic logs, and
4161 bombs out if it can't open /dev/null.
4163 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4164 interact so that it is all more flexible. It is supposed to remain
4165 backwards compatible. Also added extra_local_interfaces.
4167 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4168 to bomb out with an assertion failure - to the client this appears as a
4169 connection drop. This problem occurs in the part of the code that was taken
4170 from the Samba project. Fortunately, the assertion is in a very simple
4171 function, so I have fixed this by reproducing the function inline in the
4172 one place where it is called, and arranging for authentication to fail
4173 instead of killing the process with assert().
4175 75. The SPA client code was not working when the server requested OEM rather
4176 than Unicode encoding.
4178 76. Added code to make require_files with a specific uid setting more usable in
4179 the case where statting the file as root fails - usually a non-root-mounted
4180 NFS file system. When this happens and the failure is EACCES, Exim now
4181 forks a subprocess and does the per-uid checking as the relevant uid.
4183 77. Added process_log_path.
4185 78. If log_file_path was not explicitly set, a setting of check_log_space or
4186 check_log_inodes was ignored.
4188 79. If a space check for the spool or log partitions fails, the incident is now
4189 logged. Of course, in the latter case the data may get lost...
4191 80. Added the %p formatting code to string_format() so that it can be used to
4192 print addresses in debug_print(). Adjusted all the address printing in the
4193 debugging in store.c to use %p rather than %d.
4195 81. There was a concern that a line of code in smtp_in.c could overflow a
4196 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4197 initially expressed, the concern was not well-founded, because trailing
4198 spaces are removed early. However, if the trailing spaces were followed by
4199 a NULL, they did not get removed, so the overflow was possible. Two fixes
4202 (a) I re-wrote the offending code in a cleaner fashion.
4203 (b) If an incoming SMTP command contains a NULL character, it is rejected
4206 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4207 runs initgroups(), so that if the Exim user is in any additional groups,
4208 they will be used during message reception.
4214 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob