1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.381 2006/07/27 10:13:52 ph10 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 SC/01 Use a glob alias rather than an array ref in eximstats generated
10 parser. This improves both readability and performance.
12 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
13 Don't display local sender or destination tables in eximstats unless
14 there is data to show.
15 Added average volumes into the eximstats top table text output.
17 SC/03 Collect data on the number of addresses (recipients) as well
18 as the number of messages in eximstats.
20 TF/01 Correct an error in the documentation for the redirect router. Exim
21 does (usually) call initgroups() when daemonizing.
23 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
24 with consistent privilege compared to when running as a daemon.
26 TF/03 Note in the spec that $authenticated_id is not set for local
27 submissions from trusted users.
29 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
30 Thanks to Dean Brooks <dean@iglou.com> for the patch.
32 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
33 by adding some example configuration directives to the default
34 configuration file. A little bit of work is required to uncomment the
35 directives and define how usernames and passwords are checked, but
36 there is now a framework to start from.
38 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
39 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
40 without this. I don't know how relevant this is to other LDAP libraries.
42 PH/02 Add the verb name to the "unknown ACL verb" error.
44 PH/03 Magnus Holmgren's patch for filter_prepend_home.
46 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
48 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
49 directory not expanded when it should be if an expanded home directory
50 was set for the address (which is overridden by the transport).
52 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
55 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
56 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
57 because it is too late at that time, and has no effect.
59 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
60 security issue with \' (bugzilla #107). I could not use the
61 PQescapeStringConn() function, because it needs a PGconn value as one of
64 PH/08 When testing addresses using -bt, indicate those final addresses that
65 are duplicates that would not cause an additional delivery. At least one
66 person was confused, thinking that -bt output corresponded to deliveries.
67 (Suppressing duplicates isn't a good idea as you lose the information
68 about possibly different redirections that led to the duplicates.)
70 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
71 systems where poll() doesn't work, in particular OS X.
73 PH/10 Added more information to debugging output for retry time not reached.
75 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
76 operations in malware.c.
78 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
81 PH/13 If write_rejectlog was set false when logging was sent to syslog with
82 syslog_duplication set false, log lines that would normally be written
83 both the the main log and to the reject log were not written to syslog at
86 PH/14 In the default configuration, change the use of "message" in ACL warn
87 statements to "add_header".
89 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
90 not followed by a command (e.g. "seen endif").
92 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
93 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
96 PH/17 Added an extra condition to the default value of delay_warning_condition
100 { match{$h_precedence:}{(?i)bulk|list|junk} }
101 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} }
104 The Auto-Submitted: header is standardised, whereas I don't think
105 Precedence: ever was.
107 PH/18 Refactored debugging code in route_finduser() to show more information,
108 in particular, the error code if getpwnam() issues one.
110 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
111 This is apparently needed in addition to the PH/07 change above to avoid
112 any possible encoding problems.
114 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
115 but not after initializing Perl.
117 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
118 output them only if debugging. By default they are written stderr,
119 apparently, which is not desirable.
121 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
124 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
127 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
129 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
130 authenticated or an ident call has been made. Suppress the default
131 values for $authenticated_id and $authenticated_sender (but permit -oMai
132 and -oMas) when testing with -bh.
138 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
139 other effects) broke the use of negated acl sub-conditions.
141 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
144 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
145 "Deny" causes Exim to reject the incoming connection with a 554 error.
146 Unfortunately, if there is a major crisis, such as a disk failure,
147 tcp-wrappers gives "deny", whereas what one would like would be some
148 kind of temporary error. A kludge has been added to help with this.
149 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
150 554 error is used if errno is still zero or contains ENOENT (which occurs
151 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
154 PH/03 Add -lutil to the default FreeBSD LIBS setting.
156 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
157 errors. Otherwise a message that provokes a temporary error (when other
158 messages do not) can cause a whole host to time out.
160 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
161 addresses were routed directly to files or pipes from a redirect router.
162 File deliveries just didn't batch; pipe deliveries might have suffered
165 PH/06 A failure to get a lock for a hints database would erroneously always say
166 "Failed to get write lock", even when it was really a read lock.
168 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
169 of 0600. This has been changed to use the value of the lockfile_mode
170 option (which defaults to 0600).
172 PH/08 Applied small patch from the Sieve maintainer.
174 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
175 folder from quota calculations, a direct delivery into this folder messed
176 up the contents of the maildirsize file. This was because the regex was
177 used only to exclude .Trash (or whatever) when the size of the mailbox
178 was calculated. There was no check that a delivery was happening into an
179 excluded directory. This bug has been fixed by ignoring all quota
180 processing for deliveries into excluded directories.
182 PH/10 Added the maildirfolder_create_regex option to appendfile.
188 PH/01 The code for finding all the local interface addresses on a FreeBSD
189 system running IPv6 was broken. This may well have applied to all BSD
190 systems, as well as to others that have similar system calls. The broken
191 code found IPv4 interfaces correctly, but gave incorrect values for the
192 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
193 that it would not match correctly against @[] and not recognize the IPv6
196 PH/02 The ipliteral router was not recognizing addresses of the form user@
197 [ipv6:....] because it didn't know about the "ipv6:" prefix.
199 PH/03 Added disable_ipv6.
201 PH/04 Changed $reply_address to use the raw form of the headers instead of the
202 decoded form, because it is most often used to construct To: headers
203 lines in autoreplies, and the decoded form may well be syntactically
204 invalid. However, $reply_address has leading white space removed, and all
205 newlines turned into spaces so that the autoreply transport does not
208 PH/05 If group was specified without a user on a router, and no group or user
209 was specified on a transport, the group from the router was ignored.
211 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
212 for visible compile-time settings that can be used to change these
213 numbers, for those that want even more. Backwards compatibility with old
214 spool files has been maintained. However, going back to a previous Exim
215 release will lost any variables that are in spool files.
217 PH/07 Two small changes when running in the test harness: increase delay when
218 passing a TCP/IP connection to a new process, in case the original
219 process has to generate a bounce, and remove special handling of
220 127.0.0.2 (sic), which is no longer necessary.
222 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
223 be the same on different OS.
225 PH/09 Moved a debug statement in filter processing to avoid a race problem when
228 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
229 whether --show-vars was specified or not
231 JJ/02 exipick: Added support for new ACL variable spool format introduced
234 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
235 syntactically invalid From: or Reply-to: line, and a filter used this to
236 generate an autoreply, and therefore failed to obtain an address for the
237 autoreply, Exim could try to deliver to a non-existent relative file
238 name, causing unrelated and misleading errors. What now happens is that
239 it logs this as a hard delivery error, but does not attempt to create a
242 PH/11 The exinext utility has a -C option for testing purposes, but although
243 the given file was scanned by exinext itself; it wasn't being passed on
244 when Exim was called.
246 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
247 an end-of-file indication when reading a command response.
249 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
250 compiled. In many other places in Exim, IPv6 addresses are always
251 recognized, so I have changed this. It also means that IPv4 domain
252 literals of the form [IPV4:n.n.n.n] are now always recognized.
254 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
255 used if the router is not running as root, for example, when verifying at
256 ACL time, or when using -bh. The debugging output from this situation was
257 non-existent - all you got was a failure to exec. I have made two
260 (a) Failures to set uid/gid, the current directory, or a process leader
261 in a subprocess such as that created by queryprogram now generate
262 suitable debugging ouput when -d is set.
264 (b) The queryprogram router detects when it is not running as root,
265 outputs suitable debugging information if -d is set, and then runs
266 the subprocess without attempting to change uid/gid.
268 PH/15 Minor change to Makefile for building test_host (undocumented testing
271 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
272 additional section of a DNS packet that returns MX or SRV records.
273 Instead, it always explicitly searches for A/AAAA records. This avoids
274 major problems that occur when a DNS server includes only records of one
275 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
276 fixed another bug: if SRV records were looked up and the corresponding
277 address records were *not* found in the additional section, the port
278 values from the SRV records were lost.
280 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
281 using the correct key (the original address) when searching the retry
282 rules in order to find which one to use for generating the retry hint.
284 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
285 adding the default one. Similarly, if it contains a Reply-To: header, the
286 errors_reply_to option, if set, is not used.
288 PH/19 When calculating a retry time, Exim used to measure the "time since
289 failure" by looking at the "first failed" field in the retry record. Now
290 it does not use this if it is later than than the arrival time of the
291 message. Instead it uses the arrival time. This makes for better
292 behaviour in cases where some deliveries succeed, thus re-setting the
293 "first failed" field. An example is a quota failure for a huge message
294 when small messages continue to be delivered. Without this change, the
295 "time since failure" will always be short, possible causing more frequent
296 delivery attempts for the huge message than are intended.
297 [Note: This change was subsequently modified - see PH/04 for 4.62.]
299 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
300 $1, $2, $3) because the numerical variables can be reset during some
301 expansion items (e.g. "match"), thereby losing the authentication data.
303 PH/21 Make -bV show the size of off_t variables so that the test suite can
304 decide whether to run tests for quotas > 2G.
306 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
307 mailbox_size, and mailbox_filecount in the appendfile transport. If a
308 filecount value is greater than 2G or if a quota value is greater than 2G
309 on a system where the size of off_t is not greater than 4, a panic error
312 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
313 never match. The debug and -bh output now contains an explicit error
314 message indicating a malformed IPv4 address or mask.
316 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
317 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
320 PH/25 Do not write to syslog when running in the test harness. The only
321 occasion when this arises is a failure to open the main or panic logs
322 (for which there is an explicit test).
324 PH/26 Added the /no_tell option to "control=freeze".
326 PH/27 If a host name lookup failed very early in a connection, for example, if
327 the IP address matched host_lookup and the reverse lookup yielded a name
328 that did not have a forward lookup, an error message of the form "no IP
329 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
330 could be logged. Now it outputs the IP address instead of "NULL".
332 PH/28 An enabling patch from MH: add new function child_open_exim2() which
333 allows the sender and the authenticated sender to be set when
334 submitting a message from within Exim. Since child_open_exim() is
335 documented for local_scan(), the new function should be too.
337 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
338 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
339 results in an empty string is now treated as unset.
341 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
343 PH/31 Added sender_verify_fail logging option.
345 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
346 needed by RFC 822 but not by RFC 2822 was commented out. I have now
347 tidied the source and removed it altogether.
349 PH/33 When a queue run was abandoned because the load average was too high, a
350 log line was always written; now it is written only if the queue_run log
351 selector is set. In addition, the log line for abandonment now contains
352 information about the queue run such as the pid. This is always present
353 in "start" and "stop" lines but was omitted from the "abandon" line.
355 PH/34 Omit spaces between a header name and the colon in the error message that
356 is given when verify = headers_syntax fails (if there are lots of them,
357 the message gets confusing).
359 PH/35 Change the default for dns_check_names_pattern to allow slashes within
360 names, as there are now some PTR records that contain slashes. This check
361 is only to protect against broken name servers that fall over on strange
362 characters, so the fact that it applies to all lookups doesn't matter.
364 PH/36 Now that the new test suite is complete, we can remove some of the
365 special code in Exim that was needed for the old test suite. For example,
366 sorting DNS records because real resolvers return them in an arbitrary
367 order. The new test suite's fake resolver always returns records in the
370 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
371 bounces) except when queue_only is set, to avoid logging races between
372 the different processes.
374 PH/38 Panic-die if .include specifies a non-absolute path.
376 PH/39 A tweak to the "H" retry rule from its user.
378 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
379 a label. They prevented compilation on older perls.
381 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
382 a warning to be raised on newish perls.
384 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
385 on queue. Changes to match documented behaviour of showing count of
386 messages matching specified criteria.
388 PH/40 Changed the default ident timeout from 30s to 5s.
390 PH/41 Added support for the use of login_cap features, on those BSD systems
391 that have them, for controlling the resources used by pipe deliveries.
393 PH/42 The content-scanning code uses fopen() to create files in which to put
394 message data. Previously it was not paying any attention to the mode of
395 the files. Exim runs with umask(0) because the rest of the code creates
396 files with open(), and sets the required mode explicitly. Thus, these
397 files were ending up world-writeable. This was not a big issue, because,
398 being within the spool directory, they were not world-accessible. I have
399 created a function called modefopen, which takes an additional mode
400 argument. It sets umask(777), creates the file, chmods it to the required
401 mode, then resets the umask. All the relevant calls to fopen() in the
402 content scanning code have been changed to use this function.
404 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
405 to 24 hours. This avoids potential overflow problems when processing G
406 and H retry rules. I suspect nobody ever tinkers with this value.
408 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
410 PH/45 When the plaintext authenticator is running as a client, the server's
411 challenges are checked to ensure they are valid base64 strings. By
412 default, the authentication attempt is cancelled if an invalid string is
413 received. Setting client_ignore_invalid_base64 true ignores these errors.
414 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
415 they are received. Thus, the responses can be made to depend on the
416 challenges. If an invalid string is ignored, an empty string is placed in
419 PH/46 Messages that are created by the autoreply transport now contains a
420 References: header, in accordance with RFCs 2822 and 3834.
422 PH/47 Added authenticated_sender_force to the smtp transport.
424 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
426 PH/49 Installed latest patch from the Sieve maintainer.
428 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
429 was also set, the appendfile transport was unnecessarily scanning a
430 directory of message files (e.g. for maildir delivery) to find the count
431 of files (along with the size), even though it did not need this
432 information. It now does the scan only if it needs to find either the
433 size of the count of files.
435 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
437 PH/52 Two bugs concerned with error handling when the smtp transport is
440 (i) Exim was not creating retry information for temporary errors given
441 for individual recipients after the DATA command when the smtp transport
442 was used in LMTP mode. This meant that they could be retried too
443 frequently, and not timed out correctly.
445 (ii) Exim was setting the flag that allows error details to be returned
446 for LMTP errors on RCPT commands, but not for LMTP errors for individual
447 recipients that were returned after the DATA command.
449 PH/53 This is related to PH/52, but is more general: for any failing address,
450 when detailed error information was permitted to be returned to the
451 sender, but the error was temporary, then after the final timeout, only
452 "retry timeout exceeded" was returned. Now it returns the full error as
453 well as "retry timeout exceeded".
455 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
456 do this, and (what is worse) MTAs that accept it.
458 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
459 will now be deprecated.
461 PH/56 New os.c-cygwin from the Cygwin maintainer.
463 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
464 formats (previously only available in exim formats via -bpr, -bpru,
465 and -bpra. Now also available in native and exiqgrep formats)
467 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
468 with very large, slow to parse queues
470 JJ/08 exipick: added ! as generic prefix to negate any criteria format
472 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
474 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
475 responses to authentication challenges, though it was showing the
476 challenges; (ii) I've removed the CR characters from the debug output for
479 PH/58 Allow for the insertion of a newline as well as a space when a string
480 is turned into more than one encoded-word during RFC 2047 encoding. The
481 Sieve code now uses this.
483 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
484 data_4xx, lost_connection, tls_required.
486 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
487 message was being sent as an SMTP response.
489 PH/61 Add -l and -k options to exicyclog.
491 PH/62 When verifying, if an address was redirected to one new address, so that
492 verification continued, and the new address failed or deferred after
493 having set something in $address_data, the value of $address_data was not
494 passed back to the ACL. This was different to the case when no
495 redirection occurred. The value is now passed back in both cases.
497 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
498 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
499 use login_cap.h, so on its own it isn't the distinguishing feature. The
500 new name refers directly to the setclassresources() function.
502 PH/65 Added configuration files for NetBSD3.
504 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
506 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
507 is preferred over IPv4.
509 PH/68 The bounce_return_message and bounce_return_body options were not being
510 honoured for bounces generated during the reception of non-SMTP messages.
511 In particular, this applied to messages rejected by the ACL. This bug has
512 been fixed. However, if bounce_return_message is true and bounce_return_
513 body is false, the headers that are returned for a non-SMTP message
514 include only those that have been read before the error was detected.
515 (In the case of an ACL rejection, they have all been read.)
517 PH/69 The HTML version of the specification is now built in a directory called
518 spec_html instead of spec.html, because the latter looks like a path with
519 a MIME-type, and this confuses some software.
521 PH/70 Catch two compiler warnings in sieve.c.
523 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
524 function verify_get_ident() calls ip_connect() to connect a socket, but
525 if the "connect()" function timed out, ip_connect() used to close the
526 socket. However, verify_get_ident() also closes the socket later, and in
527 between Exim writes to the log, which may get opened at this point. When
528 the socket was closed in ip_connect(), the log could get the same file
529 descriptor number as the socket. This naturally causes chaos. The fix is
530 not to close the socket in ip_connect(); the socket should be closed by
531 the function that creates it. There was only one place in the code where
532 this was missing, in the iplookup router, which I don't think anybody now
533 uses, but I've fixed it anyway.
535 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
536 well as to direct DNS lookups. Otherwise the handling of names in host
537 lists is inconsistent and therefore confusing.
543 PH/01 Two changes to the default runtime configuration:
545 (1) Move the checks for relay_from_hosts and authenticated clients from
546 after to before the (commented out) DNS black list checks.
548 (2) Add control=submission to the relay_from_hosts and authenticated
549 clients checks, on the grounds that messages accepted by these
550 statements are most likely to be submissions.
552 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
554 (1) Generate an error if the third argument for the ${prvs expansion is
557 (2) Treat a missing third argument of ${prvscheck as if it were an empty
560 (3) Reset the variables that are obtained from the first argument of
561 ${prvscheck and used in the second argument before leaving the code,
562 because their memory is reclaimed, so using them afterwards may do
565 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
566 one (it's much easier than Tom thought :-).
568 (5) Because of (4), we can now allow for the use of $prvscheck_result
569 inside the third argument.
571 PH/03 For some reason, the default setting of PATH when running a command from
572 a pipe transport was just "/usr/bin". I have changed it to
575 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
576 anything to be listed in the output from -bV.
578 PH/05 When a filter generated an autoreply, the entire To: header line was
579 quoted in the delivery log line, like this:
581 => >A.N.Other <ano@some.domain> <original@ddress> ...
583 This has been changed so that it extracts the operative address. There
584 may be more than one such address. If so, they are comma-separated, like
587 => >ano@some.domain,ona@other.domain <original@ddress> ...
589 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
590 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
591 not being looked up in the rDNS to get a host name, Exim was showing the
592 IP address twice in Received: lines, even though the IP addresses were
593 identical. For example:
595 Received: from [1.2.3.4] (helo=[1.2.3.4])
597 However, if the real host name was known, it was omitting the HELO data
598 if it matched the actual IP address. This has been tidied up so that it
599 doesn't show the same IP address twice.
601 PH/07 When both +timestamp and +memory debugging was on, the value given by
602 $tod_xxx expansions could be wrong, because the tod_stamp() function was
603 called by the debug printing, thereby overwriting the timestamp buffer.
604 Debugging no longer uses the tod_stamp() function when +timestamp is set.
606 PH/08 When the original message was included in an autoreply transport, it
607 always said "this is a copy of the message, including all the headers",
608 even if body_only or headers_only was set. It now gives an appropriate
611 PH/09 Applied a patch from the Sieve maintainer which:
613 o fixes some comments
614 o adds the (disabled) notify extension core
615 o adds some debug output for the result of if/elsif tests
616 o points to the current vacation draft in the documentation
617 and documents the missing references header update
621 o fixes a bug in processing the envelope test (when testing
622 multiple envelope elements, the last element determinted the
625 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
626 Electronic Mail") by including:
628 Auto-submitted: auto-generated
630 in the messages that it generates (bounce messages and others, such as
631 warnings). In the case of bounce messages for non-SMTP mesages, there was
632 also a typo: it was using "Auto_submitted" (underscore instead of
633 hyphen). Since every message generated by Exim is necessarily in response
634 to another message, thes have all been changed to:
636 Auto-Submitted: auto-replied
638 in accordance with these statements in the RFC:
640 The auto-replied keyword:
642 - SHOULD be used on messages sent in direct response to another
643 message by an automatic process,
645 - MUST NOT be used on manually-generated messages,
647 - MAY be used on Delivery Status Notifications (DSNs) and Message
648 Disposition Notifications (MDNs),
650 - MUST NOT be used on messages generated by automatic or periodic
651 processes, except for messages which are automatic responses to
654 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
655 to the default Received: header definition.
657 PH/12 Added log selector acl_warn_skipped (default on).
659 PH/13 After a successful wildlsearch lookup, discard the values of numeric
660 variables because (a) they are in the wrong storage pool and (b) even if
661 they were copied, it wouldn't work properly because of the caching.
663 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
664 checking when decoding. Apparently there are clients that generate
665 overlong encoded strings. Why am I not surprised?
667 PH/15 If the first argument of "${if match_address" was not empty, but did not
668 contain an "@" character, Exim crashed. Now it writes a panic log message
669 and treats the condition as false.
671 PH/16 In autoreply, treat an empty string for "once" the same as unset.
673 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
674 extension "envelope-auth". The code is finished and in agreement with
675 other implementations, but there is no documentation so far and in fact,
676 nobody wrote the draft yet. This extension is currently #undef'ed, thus
677 not changing the active code.
679 Print executed "if" and "elsif" statements when debugging is used. This
680 helps a great deal to understand what a filter does.
682 Document more things not specified clearly in RFC3028. I had all this
683 sorted out, when out of a sudden new issues came to my mind. Oops."
685 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
688 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
689 canonical form (as documented). However, after a host name lookup from
690 the IP address, check_host() was doing a simple string comparison with
691 addresses acquired from the DNS when checking that the found name did
692 have the original IP as one of its addresses. Since any found IPv6
693 addresses are likely to be in abbreviated form, the comparison could
694 fail. Luckily, there already exists a function for doing the comparison
695 by converting both addresses to binary, so now that is used instead of
698 PH/20 There was another similar case to PH/19, when a complete host name was
699 given in a host list; looking up its IP address could give an abbreviated
700 form, whereas the current host's name might or might not be abbreviated.
701 The same fix has been applied.
707 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
708 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
711 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
712 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
714 PH/03 Typo: missing ".o" in src/pcre/Makefile.
716 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
717 header line, restrict the check to what is listed in RFCs 2369 and 2929.
718 Also, for "Auto-Submitted", treat anything other than "no" as
719 non-personal, in accordance with RFC 3834. (Previously it treated
720 anything starting "auto-" as non-personal.)
722 TF/01 The control=submission/name=... option had a problem with syntax
723 errors if the name included a slash character. The /name= option
724 now slurps the rest of the string, so it can include any characters
725 but it must come last in the list of options (after /sender_retain
728 PH/05 Some modifications to the interface to the fake nameserver for the new
736 TK/01 Added the "success_on_redirect" address verification option. See
737 NewStuff for rationale and an example.
739 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
741 PH/02 Patch to exigrep to allow it to work on syslog lines.
743 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
744 fread() to skip over the body file's header line, because in Cygwin the
745 header line is locked and is inaccessible.
747 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
748 co-exist for some time) to make it clear that it is the Exim ID that is
749 referenced, not the Message-ID: header line.
751 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
752 string_format() function, because snprintf() does not exist on all
755 PH/06 The use of forbid_filter_existstest now also locks out the use of the
756 ${stat: expansion item.
758 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
759 protocol synchronization error", to keep the pedants happy.
761 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
762 well as for IRIX systems, when gcc is being used. See the host.c source
765 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
767 PH/10 Named domain lists were not working if used in a queue_smtp_domains
770 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
771 transport and to the smtp transport in LMTP mode.
773 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
775 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
776 run a filter in a subprocess. This could lead to confusion in subsequent
777 lookups in the parent process. There should also be a search_tidyup() at
778 the end of the subprocess.
780 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
781 only if the host matched helo_try_verify_hosts, which caused the
782 verification to occur when the EHLO/HELO command was issued. The ACL just
783 tested the remembered result. Now, if a previous verification attempt has
784 not happened, "verify = helo" does it there and then.
786 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
788 TK/03 Fix log output including CR from clamd.
790 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
791 exist provoked a memory error which could cause a segfault.
793 PH/15 Installed PCRE 6.2
795 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
797 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
798 of the problem. Specifically, suggested +O2 rather than +O1 for the
801 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
803 PH/20 If a delivery was routed to a non-standard port by means of an SRV
804 record, the port was not correctly logged when the outgoing_port log
805 selector was set (it logged the transort's default port).
807 PH/21 Added support for host-specific ports to manualroute, queryprogram,
808 fallback_hosts, and "hosts" in the smtp transport.
810 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
811 host errors such as "Connection refused".
813 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
814 authentication with radiusclient 0.4.9:
816 - Error returned from rc_read_config was caught wrongly
817 - Username/password not passed on to radius server due to wrong length.
819 The presumption is that some radiusclient API changes for 4.51/PH/17
820 were not taken care of correctly. The code is still untested by me (my
821 Linux distribution still has 0.3.2 of radiusclient), but it was
822 contributed by a Radius user.
824 PH/24 When doing a callout, the value of $domain wasn't set correctly when
825 expanding the "port" option of the smtp transport.
827 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
828 while reading a MIME header. Thanks to Tom Hughes for a patch.
830 PH/24 Include config.h inside local_scan.h so that configuration settings are
833 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
834 that in an ACL for RCPT (for example), you can examine exactly what was
837 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
838 commands, but it was not correctly comparing the address with the actual
839 client host address. Thus, it would show the EHLO address in Received:
840 header lines when this was not necessary.
842 PH/27 Added the % operator to ${eval:}.
844 PH/28 Exim tries to create and chdir to its spool directory when it starts;
845 it should be ignoring failures (because with -C, for example, it has lost
846 privilege). It wasn't ignoring creation failures other than "already
849 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
852 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
853 because an input line was too long, either on its own, or by virtue of
854 too many continuations, the temporary file was not being removed, and the
855 return code was incorrect.
857 PH/31 Missing "BOOL" in function definition in filtertest.c.
859 PH/32 Applied Sieve patches from the maintainer.
861 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
863 PH/33 Added "verify = not_blind".
865 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
866 Local/Makefile (with some defaults set). These are used in built scripts
867 such as exicyclog, but they have never been used in the exim_install
868 script (though there are many overriding facilities there). I have
869 arranged that the exim_install script now takes note of these two
872 PH/35 Installed configuration files for Dragonfly.
874 PH/36 When a locally submitted message by a trusted user did not contain a
875 From: header, and the sender address was obtained from -f or from an SMTP
876 MAIL command, and the trusted user did not use -F to supply a sender
877 name, $originator_name was incorrectly used when constructing a From:
878 header. Furthermore, $originator_name was used for submission mode
879 messages from external hosts without From: headers in a similar way,
880 which is clearly wrong.
882 PH/37 Added control=suppress_local_fixups.
884 PH/38 When log_selector = +received_sender was set, and the addition of the
885 sender made the log line's construction buffer exactly full, or one byte
886 less than full, an overflow happened when the terminating "\n" was
889 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
890 when the result of a list match is failure because a DNS lookup failed.
892 PH/40 RM_COMMAND is now used in the building process.
894 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
895 the "build-* directories that it finds.
897 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
898 address in a domain literal was a prefix of an interface address.
900 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
901 when verifying a sender address, unless rewrite_headers is false.
903 PH/44 Wrote a long comment about why errors_to addresses are verified as
904 recipients, not senders.
906 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
907 the ratelimit ACL was added.
909 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
911 PH/46 Added extra information about PostgreSQL errors to the error string.
913 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
914 avoiding the need to install special zones in a real server. This is
915 backwards compatible; if it can't find the fake resolver, it drops back.
916 Thus, both old and new test suites can be run.
918 TF/02 Added util/ratelimit.pl
920 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
921 clock is set back in time.
923 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
924 Candler <B.Candler@pobox.com>.
926 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
927 OK for addresses that are the result of redirections.
929 PH/48 A number of further additions for the benefit of the new test suite,
930 including a fake gethostbyname() that interfaces to the fake DNS resolver
933 TF/06 The fix for widen_domains has also been applied to qualify_single and
934 search_parents which are the other dnslookup options that can cause
937 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
940 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
942 TF/07 Exim produced the error message "an SRV record indicated no SMTP
943 service" if it encountered an MX record with an empty target hostname.
944 The message is now "an MX or SRV record indicated no SMTP service".
946 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
947 if the DNS of the sending site is misconfigured. This is quite a
948 common situation. This change restores the behaviour of treating a
949 helo verification defer as a failure.
951 PH/51 If self=fail was set on a router, the bounce message did not include the
952 actual error message.
958 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
960 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
961 command itself ended in error, the underlying message about the transport
962 filter timeout was being overwritten with the pipe command error. Now the
963 underlying error message should be appended to the second error message.
965 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
967 PH/02 Reduce the amount of output that "make" produces by default. Full output
968 can still be requested.
970 PH/03 The warning log line about a condition test deferring for a "warn" verb
971 was being output only once per connection, rather than after each
972 occurrence (because it was using the same function as for successful
973 "warn" verbs). This seems wrong, so I have changed it.
975 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
976 it should not have, which might have caused a crash in the right
977 circumstances, but probably never did.
979 PH/04 Installed a modified version of Tony Finch's patch to make submission
980 mode fix the return path as well as the Sender: header line, and to
981 add a /name= option so that you can make the user's friendly name appear
984 TF/03 Added the control = fakedefer ACL modifier.
986 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
987 Mark Lowes for thorough testing.
989 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
991 TK/03 Merged latest SRS patch from Miles Wilton.
993 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
994 with the definition in sysexits.h (which is #included earlier).
995 Fortunately, Exim does not actually use EX_OK. The code used to try to
996 preserve the sysexits.h value, by assumimg that macro definitions were
997 scanned for macro replacements. I have been disabused of this notion,
998 so now the code just undefines EX_OK before #including unistd.h.
1000 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1001 in the smtp transport. When a block could not be written in a single
1002 write() function, the timeout was being re-applied to each part-write.
1003 This seems wrong - if the receiver was accepting one byte at a time it
1004 would take for ever. The timeout is now adjusted when this happens. It
1005 doesn't have to be particularly precise.
1007 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1008 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1010 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1011 without a domain if the check to <postmaster@domain> fails.
1013 SC/01 Eximstats: added -xls and the ability to specify output files
1014 (patch written by Frank Heydlauf).
1016 SC/02 Eximstats: use FileHandles for outputing results.
1018 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1020 SC/04 Eximstats: fixed display of large numbers with -nvr option
1022 SC/05 Eximstats: fixed merging of reports with empty tables.
1024 SC/06 Eximstats: added the -include_original_destination flag
1026 SC/07 Eximstats: removed tabs and trailing whitespace.
1028 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1030 TK/06 MBOX spool code: Add real "From " MBOX separator line
1031 so the .eml file is really in mbox format (even though
1032 most programs do not really care). Patch from Alex Miller.
1034 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1035 The latter is generated from $received_to and is only set if the
1036 message has one envelope recipient. SA can use these headers,
1037 obviously out-of-the-box. Patch from Alex Miller.
1039 PH/08 The ${def test on a variable was returning false if the variable's
1040 value was "0", contrary to what the specification has always said!
1041 The result should be true unless the variable is empty.
1043 PH/09 The syntax error of a character other than { following "${if
1044 def:variable_name" (after optional whitespace) was not being diagnosed.
1045 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1046 accidental colon was present, for example, could give incorrect results.
1048 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1049 result is used (not including appendfile, where other changes are about
1052 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1053 This involved changing a lot of size variables from int to off_t. It
1054 should work with maildirs and everything.
1056 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1057 spamd dying while we are connected to it.
1059 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1062 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1063 function for BDB 4.3.
1065 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1067 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1068 the sources for PCRE have been re-arranged and more files are now
1071 PH/15 The code I had for printing potentially long long variables in PH/11
1072 above was not the best (it lost precision). The length of off_t variables
1073 is now inspected at build time, and an appropriate printing format (%ld
1074 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1075 to be "long long int" or "long int". This is needed for the internal
1076 formatting function string_vformat().
1078 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1079 the configuration file to be ":syslog", then the script "guesses" where
1080 the logs files are, rather than using the compiled in default. In our
1081 case the guess is not the same as the compiled default, so the script
1082 suddenly stopped working when I started to use syslog. The patch checks
1083 to see if log_file_path is "". If so, it attempts to read it from exim
1084 with no configuration file to get the compiled in version, before it
1085 falls back to the previous guessing code."
1087 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1088 implementing BATV in an Exim configuration. See NewStuff for the gory
1091 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1092 Makefile that are specific to HP-UX.
1094 PH/18 If the "use_postmaster" option was set for a recipient callout together
1095 with the "random" option, the postmaster address was used as the MAIL
1096 FROM address for the random test, but not for the subsequent recipient
1097 test. It is now used for both.
1099 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1100 patch removes a few documentation additions to RFC 3028, because the
1101 latest draft now contains them. It adds the new en;ascii-case comparator
1102 and a new error check for 8bit text in MIME parts. Comparator and
1103 require names are now matched exactly. I enabled the subaddress
1104 extension, but it is not well tested yet (read: it works for me)."
1106 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1107 rework some of the code of TK/09 above to avoid the hardwired use of
1108 "%lld" and "long long". Replaced the call to snprintf() with a call to
1111 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1112 records point to non-existent hosts", "retry timeout exceeded", and
1113 "retry time not reached for any host after a long failure period".
1115 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1116 experimental DomainKeys support:
1118 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1119 (2) On an error such as an illegally used "control", the wrong name for
1120 the control was given.
1122 These problems did NOT occur unless DomainKeys support was compiled.
1124 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1126 PH/24 Added ${if match_ip condition.
1128 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1129 databases so that it will be absolutely obvious if a crash occurs in the
1130 DB library. This is a regular occurrence (often caused by mis-matched
1133 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1134 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1135 picked up on a user's system that detects such things. There doesn't seem
1136 to be a gcc warning option for this - only an attribute that has to be
1137 put on the function's prototype. It seems that in Fedora Core 4 they have
1138 set this on a number of new functions. No doubt there will be more in due
1141 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1142 specify a transport. However, if an address that was verified by such a
1143 router was the subject of a callout, Exim crashed because it tried to
1144 read the rcpt_include_affixes from the non-existent transport. Now it
1145 just assumes that the setting of that option is false. This bug was
1146 introduced by 4.51/PH/31.
1148 PH/28 Changed -d+all to exclude +memory, because that information is very
1149 rarely of interest, but it makes the output a lot bigger. People tend to
1150 do -d+all out of habit.
1152 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1153 code in os-type was giving problems when libc.so lives in lib64, like on
1156 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1157 aren't the modern standard, and it seems that some systems' include files
1158 don't always have them. Exim was already checking for some of the newer
1159 ones like T_AAAA, and defining it itself. I've added checks for all the
1160 record types that Exim uses.
1162 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1163 not automatically generating a new one, as it is supposed to. This
1164 prevented TLS from working. If the file did exist, but contained invalid
1165 data, a new version was generated, as expected. It was only the case of a
1166 non-existent file that was broken.
1168 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1169 with a change in libdomainkeys > 0.64.
1171 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1172 from DNS. If the selector record carries the flag, it now has
1173 precedence over the domain-wide flag.
1175 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1177 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1178 the use of a port name that isn't defined in /etc/services) occurred, the
1179 message was deferred as in a normal delivery, and thus remained on the
1180 spool, instead of being failed because of the mua_wrapper setting. This
1181 is now fixed, and I tidied up some of the mua_wrapper messages at the
1184 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1185 the messages in a hash of arrays rather than using individual hashes.
1186 This is a bit cleaner and results in dramatic memory savings, albeit
1187 at a slight CPU cost.
1189 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1190 as requested by Marc Sherman.
1192 SC/10 Eximstats: added histograms for user specified patterns as requested
1195 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1197 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1198 fopen() in the content-scanning modules that did not already have it.
1204 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1205 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1207 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1209 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1211 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1212 to test Sieve filters that use "vacation".
1214 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1215 that changes the way the GnuTLS parameters are stored in the cache file.
1216 The new format can be generated externally. For backward compatibility,
1217 if the data in the cache doesn't make sense, Exim assumes it has read an
1218 old-format file, and it generates new data and writes a new file. This
1219 means that you can't go back to an older release without removing the
1222 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1223 work if there are any delivery delays because "one_time" forces the
1224 parent to be marked "delivered", so its unseen clone is never tried
1225 again. For this reason, Exim now forbids the simultaneous setting of
1228 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1229 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1230 entry if you want to know the details. The fix, however, neglected to
1231 consider the case when local delivery batching is involved. The test for
1232 "previously delivered" was not happening when checking to see if an
1233 address could be batched with a previous (undelivered) one; under
1234 certain circumstances this could lead to multiple deliveries to the same
1237 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1238 in its include files, and this causes problems building Exim.
1240 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1241 header_syntax) but Exim was just ignoring anything given after a slash.
1242 In particular, this caused confusion with an attempt to use "verify =
1243 reverse_host_lookup/defer_ok". An error is now given when options are
1244 supplied for verify items that do not have them. (Maybe reverse_host_
1245 lookup should have a defer_ok option, but that's a different point.)
1247 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1248 defined by RFC 821) to 2048, because there were problems with some AUTH
1249 commands, and RFC 1869 says the size should be increased for extended
1250 SMTP commands that take arguments.
1252 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1255 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1256 "unknown" error; now it says that the functionality isn't in the binary.
1258 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1259 an address' error message when a string expansion fails (syntax or
1260 whatever). Otherwise the password may appear in the log. Following change
1261 PH/42 below, there is no longer a chance of it appearing in a bounce
1264 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1266 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1267 of its addresses was ever tried. (Bugzilla bug #2.)
1269 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1270 the result incorrectly in the debug output. (It correctly added a newline
1271 to what was transported.)
1273 TF/01 Added $received_time.
1275 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1276 commented out examples of how to interface to a virus scanner and to
1277 SpamAssassin. Also added commented examples of av_scanner and
1278 spamd_address settings.
1280 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1281 and controls are allowed in which ACLs. There were a couple of minor
1282 errors. Some of the entries in the conditions table (which is a table of
1283 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1284 negation of where the condition IS allowed.
1286 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1288 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1289 header file does not have a version number, so I've had to invent a new
1290 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1291 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1292 radiusclient), but it was contributed by a Radius user.
1294 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1295 files or directories, for OpenSSL.
1297 PH/19 When an Exim process that is running as root has to create an Exim log
1298 file, it does so in a subprocess that runs as exim:exim so as to get the
1299 ownership right at creation (otherwise, other Exim processes might see
1300 the file with the wrong ownership). There was no test for failure of this
1301 fork() call, which would lead to the process getting stuck as it waited
1302 for a non-existent subprocess. Forks do occasionally fail when resources
1303 run out. I reviewed all the other calls to fork(); they all seem to check
1306 PH/20 When checking for unexpected SMTP input at connect time (before writing
1307 the banner), Exim was not dealing correctly with a non-positive return
1308 from the read() function. If the client had disconnected by this time,
1309 the result was a log entry for a synchronization error with an empty
1310 string after "input=" when read() returned zero. If read() returned -1
1311 (an event I could not check), uninitialized data bytes were printed.
1312 There were reports of junk text (parts of files, etc) appearing after
1315 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1317 PH/22 Added support for macro redefinition, and (re)definition in between
1318 driver and ACL definitions.
1320 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1321 forgetting to use the resulting value; it was using the unexpanded value.
1323 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1324 hadn't been configured. The fix is from Juergen Kreileder, who
1325 understands it better than I do:
1327 "Here's what I see happening with three configured cyrus_sasl
1328 authenticators configured (plain, login, cram-md5):
1330 On startup auth_cyrus_sasl_init() gets called for each of these.
1331 This means three calls to sasl_listmech() without a specified mech_list.
1332 => SASL tests which mechs of all available mechs actually work
1333 => three warnings about OTP not working
1334 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1336 With the patch, sasl_listmech() also gets called three times. But now
1337 SASL's mech_list option is set to the server_mech specified in the the
1338 authenticator. Or in other words, the answer from sasl_listmech()
1339 gets limited to just the mech you're testing for (which is different
1341 => the return list contains just 'plain' or 'login', 'cram-md5' or
1342 nothing depending on the value of ob->server_mech.
1344 I've just tested the patch: Authentication still works fine,
1345 unavailable mechs specified in the exim configuration are still
1346 caught, and the auth.log warnings about OTP are gone."
1348 PH/25 When debugging is enabled, the contents of the command line are added
1349 to the debugging output, even when log_selector=+arguments is not
1352 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1353 answer is "GNU", and only if the return is "GNU/something" is the answer
1356 PH/27 $acl_verify_message is now set immediately after the failure of a
1357 verification in an ACL, and so is available in subsequent modifiers. In
1358 particular, the message can be preserved by coding like this:
1360 warn !verify = sender
1361 set acl_m0 = $acl_verify_message
1363 Previously, $acl_verify_message was set only while expanding "message"
1364 and "log_message" when a very denied access.
1366 PH/28 Modified OS/os.c-Linux with
1368 -#ifndef OS_LOAD_AVERAGE
1369 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1371 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1372 nomenclature these days.)
1374 PH/29 Installed patch from the Sieve maintainer that adds the options
1375 sieve_useraddress and sieve_subaddress to the redirect router.
1377 PH/30 In these circumstances:
1378 . Two addresses routed to the same list of hosts;
1379 . First host does not offer TLS;
1380 . First host accepts first address;
1381 . First host gives temporary error to second address;
1382 . Second host offers TLS and a TLS session is established;
1383 . Second host accepts second address.
1384 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1385 and peerdn, if requested) that were in fact used only for the second
1388 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1389 attention to any local part prefix or suffix that was matched by the
1390 router that accepted the address. It now behaves in the same way as it
1391 does for delivery: the affixes are removed from the local part unless
1392 rcpt_include_affixes is set on the transport.
1394 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1395 timeout during the DATA phase of an incoming message.
1397 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1398 applied a patch sanctioned by the Sieve maintainer.
1400 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1401 the uid or gid is negative. A case of a negative gid caused this to be
1402 noticed. The fix allows for either to be negative.
1404 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1405 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1408 PH/36 The change PH/12 above was broken. Fixed it.
1410 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1411 the grounds that routing the same address twice would always produce the
1412 same answer. This might have been true once, but it is certainly no
1413 longer true now. Routing a child address may depend on the previous
1414 routing that produced that child. Some complicated redirection strategies
1415 went wrong when messages had multiple recipients, and made Exim's
1416 behaviour dependent on the order in which the addresses were given.
1418 I have moved the duplicate checking until after the routing is complete.
1419 Exim scans the addresses that are assigned to local and remote
1420 transports, and removes any duplicates. This means that more work will be
1421 done, as duplicates will always all be routed, but duplicates are
1422 presumably rare, so I don't expect this is of any significance.
1424 For deliveries to pipes, files, and autoreplies, the duplicate checking
1425 still happens during the routing process, since they are not going to be
1428 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1429 It corrects a timeout issue with spamd. This is Ian's comment: "The
1430 background is that sometimes spamd either never reads data from a
1431 connection it has accepted, or it never writes response data. The exiscan
1432 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1433 blindly assumes that writes won't block so it may never time out."
1435 PH/39 Allow G after quota size as well as K and M.
1437 PH/40 The value set for $authenticated_id in an authenticator may not contain
1438 binary zeroes or newlines because the value is written to log lines and
1439 to spool files. There was no check on this. Now the value is run through
1440 the string_printing() function so that such characters are converted to
1441 printable escape sequences.
1443 PH/41 $message_linecount is a new variable that contains the total number of
1444 lines in the message. Compare $body_linecount, which is the count for the
1447 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1448 bounce and delay warning messages, except in certain special cases, which
1451 (a) An SMTP error message from a remote host;
1452 (b) A message specified in a :fail: redirection;
1453 (c) A message specified in a "fail" command in a system filter;
1454 (d) A message specified in a FAIL return from the queryprogram router;
1455 (e) A message specified by the cannot_route_message router option.
1457 In these cases only, Exim does include the error details in bounce and
1458 warning messages. There are also a few cases where bland messages such
1459 as "unrouteable address" or "local delivery error" are given.
1461 PH/43 $value is now also set for the "else" part of a ${run expansion.
1463 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1464 being worked on, but at least Exim now implements the latest version to
1467 PH/45 In a pipe transport, although a timeout while waiting for the pipe
1468 process to complete was treated as a delivery failure, a timeout while
1469 writing the message to the pipe was logged, but erroneously treated as a
1470 successful delivery. Such timeouts include transport filter timeouts. For
1471 consistency with the overall process timeout, these timeouts are now
1472 treated as errors, giving rise to delivery failures by default. However,
1473 there is now a new Boolean option for the pipe transport called
1474 timeout_defer, which, if set TRUE, converts the failures into defers for
1475 both kinds of timeout. A transport filter timeout is now identified in
1478 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1479 systems where "make" and "gmake" are different, calling "gmake" at top
1480 level broke things. I've arranged for the value of $(MAKE) to be passed
1481 from the Makefile to this script so that it can call the same version of
1485 A note about Exim versions 4.44 and 4.50
1486 ----------------------------------------
1488 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1489 changes of various kinds. As a consequence, a big documentation update was
1490 needed. This delayed the release for rather longer than seemed good, especially
1491 in the light of a couple of (minor) security issues. Therefore, the changes
1492 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1493 release. So 4.44 and 4.50 are in effect two different branches that both start
1496 I have left the 4.50 change log unchanged; it contains all the changes since
1497 4.43. The change log for 4.44 is below; many of its items are identical to
1498 those for 4.50. This seems to be the most sensible way to preserve the
1499 historical information.
1505 1. Minor wording change to the doc/README.SIEVE file.
1507 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1508 computation of the current number of files was incorrect.
1510 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1511 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1514 4. Give more explanation in the error message when the command for a transport
1515 filter fails to execute.
1517 5. There are several places where Exim runs a non-Exim command in a
1518 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1519 was being done only for the command run by the queryprogram router. It is
1520 now done for all such subprocesses. The other cases are: ${run, transport
1521 filters, and the commands run by the lmtp and pipe transports.
1523 6. Added CONFIGURE_GROUP build-time option.
1525 7. Some older OS have a limit of 256 on the maximum number of file
1526 descriptors. Exim was using setrlimit() to set 1000 as a large value
1527 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1528 systems. I've change it so that if it can't get 1000, it tries for 256.
1530 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1531 was an oversight, and furthermore, ever since the addition of extra
1532 controls (e.g. 4.43/32), the checks on when to allow different forms of
1533 "control" were broken. There should now be diagnostics for all cases when a
1534 control that does not make sense is encountered.
1536 9. Added the /retain_sender option to "control=submission".
1538 10. $recipients is now available in the predata ACL (oversight).
1540 11. Tidy the search cache before the fork to do a delivery from a message
1541 received from the command line. Otherwise the child will trigger a lookup
1542 failure and thereby defer the delivery if it tries to use (for example) a
1543 cached ldap connection that the parent has called unbind on.
1545 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1546 of $address_data from the recipient verification was clobbered by the
1547 sender verification.
1549 13. The value of address_data from a sender verification is now available in
1550 $sender_address_data in subsequent conditions in the ACL statement.
1552 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1554 15. Added a new option "connect=<time>" to callout options, to set a different
1557 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1558 was its contents. (It was OK if the option was not defined at all.)
1560 17. A "Completed" log line is now written for messages that are removed from
1561 the spool by the -Mrm option.
1563 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1564 information about exactly what failed.
1566 19. Added -dd to debug only the daemon process.
1568 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1569 handles timeouts, both on the server side and network timeouts. Renamed the
1570 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1572 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1573 It is now set to "smtps".
1575 22. $host_address is now set to the target address during the checking of
1576 ignore_target_hosts.
1578 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1579 being passed; this would have caused $sender_host_name to have been used if
1580 matching the list had actually called for a host name (not very likely,
1581 since this list is usually IP addresses). A host name is now passed as
1584 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1585 code that specifies a non-restarting handler (typically sigaction() in
1586 modern systems) in an attempt to fix a rare and obscure crash bug.
1588 25. Narrowed the window for a race in the daemon that could cause it to ignore
1589 SIGCHLD signals. This is not a major problem, because they are used only to
1590 wake it up if nothing else does.
1592 26. A malformed maildirsize file could cause Exim to calculate negative values
1593 for the mailbox size or file count. Odd effects could occur as a result.
1594 The maildirsize information is now recalculated if the size or filecount
1597 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1598 support for a long time. Removed HAVE_SYS_VFS_H.
1600 28. Installed the latest version of exipick from John Jetmore.
1602 29. In an address list, if the pattern was not a regular expression, an empty
1603 subject address (from a bounce message) matched only if the pattern was an
1604 empty string. Non-empty patterns were not even tested. This was the wrong
1605 because it is perfectly reasonable to use an empty address as part of a
1606 database query. An empty address is now tested by patterns that are
1607 lookups. However, all the other forms of pattern expect the subject to
1608 contain a local part and a domain, and therefore, for them, an empty
1609 address still always fails if the pattern is not itself empty.
1611 30. Exim went into a mad DNS loop when attempting to do a callout where the
1612 host was specified on an smtp transport, and looking it up yielded more
1613 than one IP address.
1615 31. Re-factored the code for checking spool and log partition space into a
1616 function that finds that data and another that does the check. The former
1617 is then used to implement four new variables: $spool_space, $log_space,
1618 $spool_inodes, and $log_inodes.
1620 32. The RFC2047 encoding function was originally intended for short strings
1621 such as real names; it was not keeping to the 75-character limit for
1622 encoded words that the RFC imposes. It now respects the limit, and
1623 generates multiple encoded words if necessary. To be on the safe side, I
1624 have increased the buffer size for the ${rfc2047: expansion operator from
1627 33. It is now permitted to omit both strings after an "if" condition; if the
1628 condition is true, the result is "true". As before, when the second string
1629 is omitted, a false condition yields an empty string. This makes it less
1630 cumbersome to write custom ACL and router conditions.
1632 34. Failure to deliver a bounce message always caused it to be frozen, even if
1633 there was an errors_to setting on the router. The errors_to setting is now
1636 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1637 canonical form (fully expanded) before being placed in
1638 $sender_host_address.
1640 36. The table in the code that translates DNS record types into text (T_A to
1641 "A" for instance) was missing entries for NS and CNAME. It is just possible
1642 that this could have caused confusion if both these types were looked up
1643 for the same domain, because the text type is used as part of Exim's
1644 per-process caching. But the chance of anyone hitting this buglet seems
1647 37. The dnsdb lookup has been extended in a number of ways.
1649 (1) There is a new type, "zns", which walks up the domain tree until it
1650 finds some nameserver records. It should be used with care.
1652 (2) There is a new type, "mxh", which is like "mx" except that it returns
1653 just the host names, not the priorities.
1655 (3) It is now possible to give a list of domains (or IP addresses) to be
1656 looked up. The behaviour when one of the lookups defers can be
1657 controlled by a keyword.
1659 (4) It is now possible to specify the separator character for use when
1660 multiple records are returned.
1662 38. The dnslists ACL condition has been extended: it is now possible to supply
1663 a list of IP addresses and/or domains to be looked up in a particular DNS
1666 39. Added log_selector=+queue_time_overall.
1668 40. When running the queue in the test harness, wait just a tad after forking a
1669 delivery process, to get repeatability of debugging output.
1671 41. Include certificate and key file names in error message when GnuTLS fails
1672 to set them up, because the GnuTLS error message doesn't include the name
1673 of the failing file when there is a problem reading it.
1675 42. Allow both -bf and -bF in the same test run.
1677 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1679 44. The "Exiscan patch" is now merged into the mainline Exim source.
1681 45. Sometimes the final signoff response after QUIT could fail to get
1682 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1683 < 0 before doing a fflush(). This bug looks as though it goes back to the
1684 introduction of TLS in release 3.20, but "sometimes" must have been rare
1685 because the tests only now provoked it.
1687 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1688 (this can affect the format of dates).
1690 47. exim_tidydb, when checking for the continued existence of a message for
1691 which it has found a message-specific retry record, was not finding
1692 messages that were in split spool directories. Consequently, it was
1693 deleting retry records that should have stayed in existence.
1695 48. Steve fixed some bugs in eximstats.
1697 49. The SPA authentication driver was not abandoning authentication and moving
1698 on to the next authenticator when an expansion was forced to fail,
1699 contradicting the general specification for all authenticators. Instead it
1700 was generating a temporary error. It now behaves as specified.
1702 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1703 (the order specifies the preference for clients). The order is now AES256,
1704 AES128, 3DES, ARCFOUR128.
1706 51. Small patch to Sieve code - explicitly set From: when generating an
1709 52. Exim crashed if a remote delivery caused a very long error message to be
1710 recorded - for instance if somebody sent an entire SpamAssassin report back
1711 as a large number of 550 error lines. This bug was coincidentally fixed by
1712 increasing the size of one of Exim's internal buffers (big_buffer) that
1713 happened as part of the Exiscan merge. However, to be on the safe side, I
1714 have made the code more robust (and fixed the comments that describe what
1717 53. Now that there can be additional text after "Completed" in log lines (if
1718 the queue_time_overall log selector is set), a one-byte patch to exigrep
1719 was needed to allow it to recognize "Completed" as not the last thing in
1722 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1723 patch that reportedly fixes this has been added. I am not expert enough to
1724 create a test for it. This is what the patch creator wrote:
1726 "I found a little strange behaviour of ldap code when working with
1727 Windows 2003 AD Domain, where users was placed in more than one
1728 Organization Units. When I tried to give exim partial DN, the exit code
1729 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1730 But simultaneously result of request was absolutely normal ldap result,
1731 so I produce this patch..."
1733 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1734 REFERENCE, so I have modified the code to exclude the patch when that macro
1737 55. Some experimental protocols are using DNS PTR records for new purposes. The
1738 keys for these records are domain names, not reversed IP addresses. The
1739 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1740 leaves it alone. Component reversal etc. now happens only for IP addresses.
1743 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1745 57. Double the size of the debug message buffer (to 2048) so that more of very
1746 long debug lines gets shown.
1748 58. The exicyclog utility now does better if the number of log files to keep
1749 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1751 59. Two changes related to the smtp_active_hostname option:
1753 (1) $smtp_active_hostname is now available as a variable.
1754 (2) The default for smtp_banner uses $smtp_active_hostname instead
1755 of $primary_hostname.
1757 60. The host_aton() function is supposed to be passed a string that is known
1758 to be a valid IP address. However, in the case of IPv6 addresses, it was
1759 not checking this. This is a hostage to fortune. Exim now panics and dies
1760 if the condition is not met. A case was found where this could be provoked
1761 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1762 components; fortuitously, this particular loophole had already been fixed
1763 by change 4.50/55 above.
1765 If there are any other similar loopholes, the new check in host_aton()
1766 itself should stop them being exploited. The report I received stated that
1767 data on the command line could provoke the exploit when Exim was running as
1768 exim, but did not say which command line option was involved. All I could
1769 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1770 running as the user.
1773 61. There was a buffer overflow vulnerability in the SPA authentication code
1774 (which came originally from the Samba project). I have added a test to the
1775 spa_base64_to_bits() function which I hope fixes it.
1778 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1779 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1781 63. The daemon start-up calls getloadavg() while still root for those OS that
1782 need the first call to be done as root, but it missed one case: when
1783 deliver_queue_load_max is set with deliver_drop_privilege. This is
1784 necessary for the benefit of the queue runner, because there is no re-exec
1785 when deliver_drop_privilege is set.
1787 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1788 This has been fixed.
1790 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1791 was in use, was not putting the data itself into the right store pool;
1792 consequently, it could be overwritten for a subsequent message in the same
1793 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1796 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1798 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1799 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1800 the function were treating the return as a boolean value, which happened to
1801 work because 0=false and not-0=true, but is not correct code.
1803 68. The host_aton() function was not handling scoped IPv6 addresses (those
1804 with, for example, "%eth0" on the end) correctly.
1806 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1807 negated items (that is, ~something) in unsigned ints. Some compilers
1808 apparently mutter when there is no cast.
1810 70. If an address verification called from an ACL failed, and did not produce a
1811 user-specific message (i.e. there was only a "system" message), nothing was
1812 put in $acl_verify_message. In this situation, it now puts the system
1815 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1816 session; change 4.31/43 added the unwanted input to the log line - except
1817 that it did not do this in the start of session case. It now does.
1819 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1820 This is wrong and can cause the other end to generate a synchronization
1821 error if it is another Exim or anything else that does the synchronization
1822 check. A QUIT command is no longer sent after a timeout.
1824 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1825 during host lookups.
1827 74. The defer_ok option of callout verification was not working if it was used
1828 when verifying addresses in header lines, that is, for this case:
1830 verify = header_sender/callout=defer_ok
1832 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1833 those file descriptors could be used for SMTP connections. If anything
1834 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1835 could be sent to the SMTP client, causing chaos. The daemon now opens
1836 stdin, stdout, and stderr to /dev/null when it puts itself into the
1839 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1840 log by default. The user can override this with suitable Perl magic.
1842 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1843 the log message when discard triggers, was not working for the DATA ACL or
1844 for the non-SMTP ACL.
1846 78. Error message wording change in sieve.c.
1848 79. If smtp_accept_max_per_host was set, the number of connections could be
1849 restricted to fewer than expected, because the daemon was trying to set up
1850 a new connection before checking whether the processes handling previous
1851 connections had finished. The check for completed processes is now done
1852 earlier. On busy systems, this bug wouldn't be noticed because something
1853 else would have woken the daemon, and it would have reaped the completed
1856 80. If a message was submitted locally by a user whose login name contained one
1857 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1858 It caused a spool format error. I have fixed the spool reading code. A
1859 related problem was that the "from" clause in the Received: line became
1860 illegal because of the space(s). It is now covered by ${quote_local_part.
1862 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
1865 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
1867 83. Patch from Sieve maintainer to fix unterminated string problem in
1868 "vacation" handling.
1870 84. Some minor changes to the Linux configuration files to help with other
1871 OS variants using glibc.
1873 85. One more patch for Sieve to update vacation handling to latest spec.
1876 ----------------------------------------------------
1877 See the note above about the 4.44 and 4.50 releases.
1878 ----------------------------------------------------
1884 1. Change 4.43/35 introduced a bug that caused file counts to be
1885 incorrectly computed when quota_filecount was set in an appendfile
1888 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1889 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1892 3. Give more explanation in the error message when the command for a transport
1893 filter fails to execute.
1895 4. There are several places where Exim runs a non-Exim command in a
1896 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1897 was being done only for the command run by the queryprogram router. It is
1898 now done for all such subprocesses. The other cases are: ${run, transport
1899 filters, and the commands run by the lmtp and pipe transports.
1901 5. Some older OS have a limit of 256 on the maximum number of file
1902 descriptors. Exim was using setrlimit() to set 1000 as a large value
1903 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1904 systems. I've change it so that if it can't get 1000, it tries for 256.
1906 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1907 was an oversight, and furthermore, ever since the addition of extra
1908 controls (e.g. 4.43/32), the checks on when to allow different forms of
1909 "control" were broken. There should now be diagnostics for all cases when a
1910 control that does not make sense is encountered.
1912 7. $recipients is now available in the predata ACL (oversight).
1914 8. Tidy the search cache before the fork to do a delivery from a message
1915 received from the command line. Otherwise the child will trigger a lookup
1916 failure and thereby defer the delivery if it tries to use (for example) a
1917 cached ldap connection that the parent has called unbind on.
1919 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1920 of $address_data from the recipient verification was clobbered by the
1921 sender verification.
1923 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1924 was its contents. (It was OK if the option was not defined at all.)
1926 11. A "Completed" log line is now written for messages that are removed from
1927 the spool by the -Mrm option.
1929 12. $host_address is now set to the target address during the checking of
1930 ignore_target_hosts.
1932 13. When checking ignore_target_hosts for an ipliteral router, no host name was
1933 being passed; this would have caused $sender_host_name to have been used if
1934 matching the list had actually called for a host name (not very likely,
1935 since this list is usually IP addresses). A host name is now passed as
1938 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1939 code that specifies a non-restarting handler (typically sigaction() in
1940 modern systems) in an attempt to fix a rare and obscure crash bug.
1942 15. Narrowed the window for a race in the daemon that could cause it to ignore
1943 SIGCHLD signals. This is not a major problem, because they are used only to
1944 wake it up if nothing else does.
1946 16. A malformed maildirsize file could cause Exim to calculate negative values
1947 for the mailbox size or file count. Odd effects could occur as a result.
1948 The maildirsize information is now recalculated if the size or filecount
1951 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1952 support for a long time. Removed HAVE_SYS_VFS_H.
1954 18. Updated exipick to current release from John Jetmore.
1956 19. Allow an empty sender to be matched against a lookup in an address list.
1957 Previously the only cases considered were a regular expression, or an
1960 20. Exim went into a mad DNS lookup loop when doing a callout where the
1961 host was specified on the transport, if the DNS lookup yielded more than
1964 21. The RFC2047 encoding function was originally intended for short strings
1965 such as real names; it was not keeping to the 75-character limit for
1966 encoded words that the RFC imposes. It now respects the limit, and
1967 generates multiple encoded words if necessary. To be on the safe side, I
1968 have increased the buffer size for the ${rfc2047: expansion operator from
1971 22. Failure to deliver a bounce message always caused it to be frozen, even if
1972 there was an errors_to setting on the router. The errors_to setting is now
1975 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1976 canonical form (fully expanded) before being placed in
1977 $sender_host_address.
1979 24. Updated eximstats to version 1.33
1981 25. Include certificate and key file names in error message when GnuTLS fails
1982 to set them up, because the GnuTLS error message doesn't include the name
1983 of the failing file when there is a problem reading it.
1985 26. Expand error message when OpenSSL has problems setting up cert/key files.
1988 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
1989 (this can affect the format of dates).
1991 28. exim_tidydb, when checking for the continued existence of a message for
1992 which it has found a message-specific retry record, was not finding
1993 messages that were in split spool directories. Consequently, it was
1994 deleting retry records that should have stayed in existence.
1996 29. eximstats updated to version 1.35
1997 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1998 1.35 - bugfix such that pie charts by volume are generated correctly
2000 30. The SPA authentication driver was not abandoning authentication and moving
2001 on to the next authenticator when an expansion was forced to fail,
2002 contradicting the general specification for all authenticators. Instead it
2003 was generating a temporary error. It now behaves as specified.
2005 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2006 (the order specifies the preference for clients). The order is now AES256,
2007 AES128, 3DES, ARCFOUR128.
2009 31. Small patch to Sieve code - explicitly set From: when generating an
2012 32. Exim crashed if a remote delivery caused a very long error message to be
2013 recorded - for instance if somebody sent an entire SpamAssassin report back
2014 as a large number of 550 error lines. This bug was coincidentally fixed by
2015 increasing the size of one of Exim's internal buffers (big_buffer) that
2016 happened as part of the Exiscan merge. However, to be on the safe side, I
2017 have made the code more robust (and fixed the comments that describe what
2020 33. Some experimental protocols are using DNS PTR records for new purposes. The
2021 keys for these records are domain names, not reversed IP addresses. The
2022 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2023 leaves it alone. Component reversal etc. now happens only for IP addresses.
2026 34. The host_aton() function is supposed to be passed a string that is known
2027 to be a valid IP address. However, in the case of IPv6 addresses, it was
2028 not checking this. This is a hostage to fortune. Exim now panics and dies
2029 if the condition is not met. A case was found where this could be provoked
2030 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2031 components; fortuitously, this particular loophole had already been fixed
2032 by change 4.50/55 or 4.44/33 above.
2034 If there are any other similar loopholes, the new check in host_aton()
2035 itself should stop them being exploited. The report I received stated that
2036 data on the command line could provoke the exploit when Exim was running as
2037 exim, but did not say which command line option was involved. All I could
2038 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2039 running as the user.
2042 35. There was a buffer overflow vulnerability in the SPA authentication code
2043 (which came originally from the Samba project). I have added a test to the
2044 spa_base64_to_bits() function which I hope fixes it.
2047 36. The daemon start-up calls getloadavg() while still root for those OS that
2048 need the first call to be done as root, but it missed one case: when
2049 deliver_queue_load_max is set with deliver_drop_privilege. This is
2050 necessary for the benefit of the queue runner, because there is no re-exec
2051 when deliver_drop_privilege is set.
2053 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2054 was in use, was not putting the data itself into the right store pool;
2055 consequently, it could be overwritten for a subsequent message in the same
2056 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2059 38. Sometimes the final signoff response after QUIT could fail to get
2060 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2061 < 0 before doing a fflush(). This bug looks as though it goes back to the
2062 introduction of TLS in release 3.20, but "sometimes" must have been rare
2063 because the tests only now provoked it.
2069 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2070 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2071 Now it returns an integer. A number of calls were still expecting a T/F
2072 return. Fortuitously, in all cases, the tests worked in OK situations,
2073 which is the norm. However, things would have gone wrong on any write
2074 failures on the smtp file descriptor. This function is used when sending
2075 messages over SMTP and also when doing verify callouts.
2077 2. When Exim is called to do synchronous delivery of a locally submitted
2078 message (the -odf or -odi options), it no longer closes stderr before doing
2081 3. Implemented the mua_wrapper option.
2083 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2085 5. Implemented the functions header_remove(), header_testname(),
2086 header_add_at_position(), and receive_remove_recipient(), and exported them
2089 6. If an ACL "warn" statement specified the addition of headers, Exim already
2090 inserted X-ACL-Warn: at the start if there was no header name. However, it
2091 was not making this test for the second and subsequent header lines if
2092 there were newlines in the string. This meant that an invalid header could
2093 be inserted if Exim was badly configured.
2095 7. Allow an ACL "warn" statement to add header lines at the start or after all
2096 the Received: headers, as well as at the end.
2098 8. Added the rcpt_4xx retry error code.
2100 9. Added postmaster_mailfrom=xxx to callout verification option.
2102 10. Added mailfrom=xxxx to the callout verification option, for verify=
2105 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2106 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2108 12. Inserted some casts to stop certain compilers warning when using pointer
2109 differences as field lengths or precisions in printf-type calls (mostly
2110 affecting debugging statements).
2112 13. Added optional readline() support for -be (dynamically loaded).
2114 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2115 same clock tick as a message's arrival, so that its received time was the
2116 same as the "first fail" time on the retry record, and that message
2117 remained on the queue past the ultimate address timeout, every queue runner
2118 would try a delivery (because it was past the ultimate address timeout) but
2119 after another failure, the ultimate address timeout, which should have then
2120 bounced the address, did not kick in. This was a "< instead of <=" error;
2121 in most cases the first failure would have been in the next clock tick
2122 after the received time, and all would be well.
2124 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2125 being recognized when the domain list was tested by the match_domain
2126 condition in an expansion string.
2128 16. Added the ${str2b64: operator.
2130 17. Exim was always calling setrlimit() to set a large limit for the number of
2131 processes, without checking whether the existing limit was already
2132 adequate. (It did check for the limit on file descriptors.) Furthermore,
2133 errors from getrlimit() and setrlimit() were being ignored. Now they are
2134 logged to the main and panic logs, but Exim does carry on, to try to do its
2135 job under whatever limits there are.
2137 18. Imported PCRE 5.0.
2139 19. Trivial typo in log message " temporarily refused connection" (the leading
2142 20. If the log selector return_path_on_delivery was set and an address was
2143 redirected to /dev/null, the delivery process crashed because it assumed
2144 that a return path would always be set for a "successful" delivery. In this
2145 case, the whole delivery is bypassed as an optimization, and therefore no
2148 21. Internal re-arrangement: the function for sending a challenge and reading
2149 a response while authentication was assuming a zero-terminated challenge
2150 string. It's now changed to take a pointer and a length, to allow for
2151 binary data in such strings.
2153 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2155 23. Exim was not respecting finduser_retries when seeking the login of the
2156 uid under which it was called; it was always trying 10 times. (The default
2157 setting of finduser_retries is zero.) Also, it was sleeping after the final
2158 failure, which is pointless.
2160 24. Implemented tls_on_connect_ports.
2162 25. Implemented acl_smtp_predata.
2164 26. If the domain in control=submission is set empty, Exim assumes that the
2165 authenticated id is a complete email address when it generates From: or
2166 Sender: header lines.
2168 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2169 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2170 chown and chgrp in /bin and hostname in /usr/bin.
2172 28. Exim was keeping the "process log" file open after each use, just as it
2173 does for the main log. This opens the possibility of it remaining open for
2174 long periods when the USR1 signal hits a daemon. Occasional processlog
2175 errors were reported, that could have been caused by this. Anyway, it seems
2176 much more sensible not to leave this file open at all, so that is what now
2179 29. The long-running daemon process does not normally write to the log once it
2180 has entered its main loop, and it closes the log before doing so. This is
2181 so that log files can straightforwardly be renamed and moved. However,
2182 there are a couple of unusual error situations where the daemon does write
2183 log entries, and I had neglected to close the log afterwards.
2185 30. The text of an SMTP error response that was received during a remote
2186 delivery was being truncated at 512 bytes. This is too short for some of
2187 the long messages that one sometimes sees. I've increased the limit to
2190 31. It is now possible to make retry rules that apply only when a message has a
2191 specific sender, in particular, an empty sender.
2193 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2194 it possible to be selective about when SMTP synchronization is enforced.
2196 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2198 32. Implemented hosts_connection_nolog.
2200 33. Added an ACL for QUIT.
2202 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2205 35. Added mailbox_size and mailbox_filecount to appendfile.
2207 36. Added control = no_multiline_responses to ACLs.
2209 37. There was a bug in the logic of the code that waits for the clock to tick
2210 in the case where the clock went backwards by a substantial amount such
2211 that the microsecond fraction of "now" was more than the microsecond
2212 fraction of "then" (but the whole seconds number was less).
2214 38. Added support for the libradius Radius client library this is found on
2215 FreeBSD (previously only the radiusclient library was supported).
2221 1. When certain lookups returned multiple values in the form name=value, the
2222 quoting of the values was not always being done properly. Specifically:
2223 (a) If the value started with a double quote, but contained no whitespace,
2225 (b) If the value contained whitespace other than a space character (i.e.
2226 tabs or newlines or carriage returns) it was not quoted.
2227 This fix has been applied to the mysql and pgsql lookups by writing a
2228 separate quoting function and calling it from the lookup code. The fix
2229 should probably also be applied to nisplus, ibase and oracle lookups, but
2230 since I cannot test any of those, I have not disturbed their existing code.
2232 2. A hit in the callout cache for a specific address caused a log line with no
2233 reason for rejecting RCPT. Now it says "Previous (cached) callout
2234 verification failure".
2236 3. There was an off-by-one bug in the queryprogram router. An over-long
2237 return line was truncated at 256 instead of 255 characters, thereby
2238 overflowing its buffer with the terminating zero. As well as fixing this, I
2239 have increased the buffer size to 1024 (and made a note to document this).
2241 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2242 when Exim is waiting for an SMTP response from a remote server, Exim
2243 restarts its select() call on the socket, thereby resetting its timeout.
2244 This is not a problem when such interrupts are rare. Somebody set up a cron
2245 job to run exiwhat every 2 minutes, which is less than the normal select()
2246 timeout (5 or 10 minutes). This meant that the select() timeout never
2247 kicked in because it was always reset. I have fixed this by comparing the
2248 time when an interrupt arrives with the time at the start of the first call
2249 to select(). If more time than the timeout has elapsed, the interrupt is
2250 treated as a timeout.
2252 5. Some internal re-factoring in preparation for the addition of Sieve
2253 extensions (by MH). In particular, the "personal" test is moved to a
2254 separate function, and given an option for scanning Cc: and Bcc: (which is
2255 not set for Exim filters).
2257 6. When Exim created an email address using the login of the caller as the
2258 local part (e.g. when creating a From: or Sender: header line), it was not
2259 quoting the local part when it contained special characters such as @.
2261 7. Installed new OpenBSD configuration files.
2263 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2264 try to make them clearer.
2266 9. Callout options, other than the timeout value, were being ignored when
2267 verifying sender addresses in header lines. For example, when using
2269 verify = header_sender/callout=no_cache
2271 the cache was (incorrectly) being used.
2273 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2274 only the Cygwin environment.
2276 11. When return_path_on_delivery was set as a log selector, if different remote
2277 addresses in the same message used different return paths and parallel
2278 remote delivery occurred, the wrong values would sometimes be logged.
2279 (Whenever a remote delivery process finished, the return path value from
2280 the most recently started remote delivery process was logged.)
2282 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2283 lines when AUTH and/or TLS are in use. This is the "received protocol"
2284 field. Exim used to use "asmtp" for authenticated SMTP, without any
2285 indication (in the protocol name) for TLS use. Now it follows the RFC and
2286 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2287 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2288 names appear in log lines as well as in Received: header lines.
2290 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2291 extensions, and comparison tests, and to fix some bugs.
2293 14. Changes to the "personal" filter test:
2295 (1) The test was buggy in that it was just doing the equivalent of
2296 "contains" tests on header lines. For example, if a user's address was
2297 anne@some.where, the "personal" test would incorrectly be true for
2299 To: susanne@some.where
2301 This test is now done by extracting each address from the header in turn,
2302 and checking the entire address. Other tests that are part of "personal"
2303 are now done using regular expressions (for example, to check local parts
2304 of addresses in From: header lines).
2306 (2) The list of non-personal local parts in From: addresses has been
2307 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2308 taken from the Sieve specification recommendations.
2310 (3) If the message contains any header line starting with "List-" it is
2311 treated as non-personal.
2313 (4) The test for "circular" in the Subject: header line has been removed
2314 because it now seems ill-conceived.
2316 15. Minor typos in src/EDITME comments corrected.
2318 16. Installed latest exipick from John Jetmore.
2320 17. If headers_add on a router specified a text string that was too long for
2321 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2322 of string_sprintf() is now avoided.
2324 18. $message_body_size was not set (it was always zero) when running the DATA
2325 ACL and the local_scan() function.
2327 19. For the "mail" command in an Exim filter, no default was being set for
2328 the once_repeat time, causing a random time value to be used if "once" was
2329 specified. (If the value happened to be <= 0, no repeat happened.) The
2330 default is now 0s, meaning "never repeat". The "vacation" command was OK
2331 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2332 (I found it when inspecting the code).
2334 20. There is now an overall timeout for performing a callout verification. It
2335 defaults to 4 times the callout timeout, which applies to individual SMTP
2336 commands during the callout. The overall timeout applies when there is more
2337 than one host that can be tried. The timeout is checked before trying the
2338 next host. This prevents very long delays if there are a large number of
2339 hosts and all are timing out (e.g. when the network connections are timing
2340 out). The value of the overall timeout can be changed by specifying an
2341 additional sub-option for "callout", called "maxwait". For example:
2343 verify = sender/callout=5s,maxwait=20s
2345 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2346 to the end before writing, but this should make it even safer).
2348 22. Exim was forgetting that it had advertised PIPELINING for the second and
2349 subsequent messages on an SMTP connection. It was also not resetting its
2350 memory on STARTTLS and an internal HELO.
2352 23. When Exim logs an SMTP synchronization error within a session, it now
2353 records whether PIPELINING has been advertised or not.
2355 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2356 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2357 rather than long int.
2359 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2361 26. Added the never_mail option to autoreply.
2367 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2368 crash if the getsockname() call failed; this can happen if a connection is
2369 closed very soon after it is established. The problem was simply in the
2370 order in which certain operations were done, causing Exim to try to write
2371 to the SMTP stream before it had set up the file descriptor. The bug has
2372 been fixed by making things happen in the correct order.
2378 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2379 before the connection was closed, thus losing the rejection response.
2381 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2382 some early Solaris releases, but causes trouble in current releases where
2383 socklen_t is defined.
2385 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2388 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2391 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2392 that releasing the top of it at the end releases what was used for sub-
2393 expansions (unless the block got too big). However, discard this block if
2394 the first thing is a variable or header, so that we can use its block when
2395 it is dynamic (useful for very large $message_headers, for example).
2397 6. Lookups now cache *every* query, not just the most recent. A new, separate
2398 store pool is used for this. It can be recovered when all lookup caches are
2399 flushed. Lookups now release memory at the end of their result strings.
2400 This has involved some general refactoring of the lookup sources.
2402 7. Some code has been added to the store_xxx() functions to reduce the amount
2403 of flapping under certain conditions.
2405 8. log_incoming_interface used to affect only the <= reception log lines. Now
2406 it causes the local interface and port to be added to several more SMTP log
2407 lines, for example "SMTP connection from", and rejection lines.
2409 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2411 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2413 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2414 could be overwritten at the end of the current message (or the start of a
2415 new message if it was set in a HELO ACL). The value is now preserved for
2416 the duration of the SMTP connection.
2418 12. If a transport had a headers_rewrite setting, and a matching header line
2419 contained an unqualified address, that address was qualified, even if it
2420 did not match any rewriting rules. The underlying bug was that the values
2421 of the flags that permit the existence of unqualified sender and recipient
2422 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2423 non-local messages, and by -bnq for local messages) were not being
2424 preserved with the message after it was received.
2426 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2427 "next input=" as part of the text comprising the host identity instead of
2428 the correct text. The code was using the same buffer for two different
2429 strings. However, depending on which order the printing function evaluated
2430 its arguments, the bug did not always show up. Under Linux, for example, my
2431 test suite worked just fine.
2433 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2434 doesn't work with some older versions of Perl. It has been changed to "my",
2435 which in any case is probably the better facility to use.
2437 15. A really picky compiler found some instances of statements for creating
2438 error messages that either had too many or two few arguments for the format
2441 16. The size of the buffer for calls to the DNS resolver has been increased
2442 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2443 for addresses that have a lot of PTR records. This alleviates a problem; it
2444 does not fully solve it.
2446 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2447 buffer now truncates the list and logs the incident, which is the same
2448 action as happens when Exim is looking up a host name and its aliases.
2449 Previously in this situation something unpredictable would happen;
2450 sometimes it was "internal error: store_reset failed".
2452 18. If a server dropped the connection unexpectedly when an Exim client was
2453 using GnuTLS and trying to read a response, the client delivery process
2454 crashed while trying to generate an error log message.
2456 19. If a "warn" verb in an ACL added multiple headers to a message in a single
2457 string, for example:
2459 warn message = H1: something\nH2: something
2461 the text was added as a single header line from Exim's point of view
2462 though it ended up OK in the delivered message. However, searching for the
2463 second and subsequent header lines using $h_h2: did not work. This has been
2464 fixed. Similarly, if a system filter added multiple headers in this way,
2465 the routers could not see them.
2467 20. Expanded the error message when iplsearch is called with an invalid key to
2468 suggest using net-iplsearch in a host list.
2470 21. When running tests using -bh, any delays imposed by "delay" modifiers in
2471 ACLs are no longer actually imposed (and a message to that effect is
2474 22. If a "gecos" field in a passwd entry contained escaped characters, in
2475 particular, if it contained a \" sequence, Exim got it wrong when building
2476 a From: or a Sender: header from that name. A second bug also caused
2477 incorrect handling when an unquoted " was present following a character
2478 that needed quoting.
2480 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2481 was not being matched caselessly.
2483 24. Arranged for all hyphens in the exim.8 source to be escaped with
2486 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2487 properly. Recipient callout cache records were still being keyed to include
2488 the sender, even when use_sender was set false. This led to far more
2489 callouts that were necessary. The sender is no longer included in the key
2490 when use_sender is false.
2492 26. Added "control = submission" modifier to ACLs.
2494 27. Added the ${base62d: operator to decode base 62 numbers.
2496 28. dnsdb lookups can now access SRV records.
2498 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2499 the configuration file.
2501 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2502 (-v) mode. This makes the output for a verbose queue run more intelligible.
2504 31. Added a use_postmaster feature to recipient callouts.
2506 32. Added the $body_zerocount variable, containing the number of binary zero
2507 bytes in the message body.
2509 33. The time of last modification of the "new" subdirectory is now used as the
2510 "mailbox time last read" when there is a quota error for a maildir
2513 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2515 35. Added +ignore_unknown as a special item in host lists.
2517 36. Code for decoding IPv6 addresses in host lists is now included, even if
2518 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2519 address was recognized as an IP address, but was then not correctly decoded
2520 into binary, causing unexpected and incorrect effects when compared with
2527 1. Very minor rewording of debugging text in manualroute to say "list of
2528 hosts" instead of "hostlist".
2530 2. If verify=header_syntax was set, and a header line with an unqualified
2531 address (no domain) and a large number of spaces between the end of the
2532 name and the colon was received, the reception process suffered a buffer
2533 overflow, and (when I tested it) crashed. This was caused by some obsolete
2534 code that should have been removed. The fix is to remove it!
2536 3. When running in the test harness, delay a bit after writing a bounce
2537 message to get a bit more predictability in the log output.
2539 4. Added a call to search_tidyup() just before forking a reception process. In
2540 theory, someone could use a lookup in the expansion of smtp_accept_max_
2541 per_host which, without the tidyup, could leave open a database connection.
2543 5. Added the variables $recipient_data and $sender_data which get set from a
2544 lookup success in an ACL "recipients" or "senders" condition, or a router
2545 "senders" option, similar to $domain_data and $local_part_data.
2547 6. Moved the writing of debug_print from before to after the "senders" test
2550 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2551 problems for message scanning, either using a data ACL, or using
2552 local_scan() because the Received: header was not generated till after they
2553 were called (in order to set the time as the time of reception completion).
2554 I have revised the way this works. The header is now generated after the
2555 body is received, but before the ACL or local_scan() are called. After they
2556 are run, the timestamp in the header is updated.
2562 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2563 before starting a queue runner without re-exec. This happened only when
2564 deliver_drop_privilege was set or when the Exim user was set to root. The
2565 effect of the bug was that timeouts during subsequent deliveries caused
2566 crashes instead of being properly handled. The handler is now left at its
2567 default (and expected) setting.
2569 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2570 message, again when deliver_drop_privilege is set or Exim is run as root.
2571 The bug described in (1) was not present in this case, but the tidying up
2572 of the other signals was missing. I have made the two cases consistent.
2574 3. The ignore_target_hosts setting on a manualroute router was being ignored
2575 for hosts that were looked up using the /MX notation.
2577 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2580 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2581 operated on the sender address. After changing the $sender_address to <>
2582 for the sender address verify, Exim was re-instated it as the original
2583 (before rewriting) address, but remembering that it had rewritten it, so it
2584 wasn't rewriting it again. This bug also had the effect of breaking the
2585 sender address verification caching when the sender address was rewritten.
2587 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2588 This has been changed so that if the ip literal address matches
2589 ignore_target_hosts, the router declines.
2591 7. Added expansion conditions match_domain, match_address, and match_local_
2592 part (NOT match_host).
2594 8. The placeholder for the Received: header didn't have a length field set.
2596 9. Added code to Exim itself and to exim_lock to test for a specific race
2597 condition that could lead to file corruption when using MBX delivery. The
2598 issue is with the lockfile that is created in /tmp. If this file is removed
2599 after a process has opened it but before that process has acquired a lock,
2600 there is the potential for a second process to recreate the file and also
2601 acquire a lock. This could lead to two Exim processes writing to the file
2602 at the same time. The added code performs the same test as UW imapd; it
2603 checks after acquiring the lock that its file descriptor still refers to
2604 the same named file.
2606 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2607 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2610 11. Added the smtp_active_hostname option. If used, this will typically be made
2611 to depend on the incoming interface address. Because $interface_address is
2612 not set up until the daemon has forked a reception process, error responses
2613 that can happen earlier (such as "too many connections") no longer contain
2616 12. If an expansion in a condition on a "warn" statement fails because a lookup
2617 defers, the "warn" statement is abandoned, and the next ACL statement is
2618 processed. Previously this caused the whole ACL to be aborted.
2620 13. Added the iplsearch lookup type.
2622 14. Added ident_timeout as a log selector.
2624 15. Added tls_certificate_verified as a log selector.
2626 16. Added a global option tls_require_ciphers (compare the smtp transport
2627 option of the same name). This controls incoming TLS connections.
2629 17. I finally figured out how to make tls_require_ciphers do a similar thing
2630 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2631 before starting the TLS session.
2633 18. Tabs are now shown as \t in -bP output.
2635 19. If the log selector return_path_on_delivery was set, Exim crashed when
2636 bouncing a message because it had too many Received: header lines.
2638 20. If two routers both had headers_remove settings, and the first one included
2639 a superfluous trailing colon, the final name in the first list and the
2640 first name in the second list were incorrectly joined into one item (with a
2641 colon in the middle).
2647 1. Added -C and -D options to the exinext utility, mainly to make it easier
2648 to include in the automated testing, but these could be helpful when
2649 multiple configurations are in use.
2651 2. The exinext utility was not formatting the output nicely when there was
2652 an alternate port involved in the retry record key, nor when there was a
2653 message id as well (for retries that were specific to a specific message
2654 and a specific host). It was also confused by IPv6 addresses, because of
2655 the additional colons they contain. I have fixed the IPv4 problem, and
2656 patched it up to do a reasonable job for IPv6.
2658 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2659 delivery, the log line now contains "pipelined" if PIPELINING was used.
2661 4. An SMTP transport process used to panic and die if the bind() call to set
2662 an explicit outgoing interface failed. This has been changed; it is now
2663 treated in the same way as a connect() failure.
2665 5. A reference to $sender_host_name in the part of a conditional expansion
2666 that was being skipped was still causing a DNS lookup. This no longer
2669 6. The def: expansion condition was not recognizing references to header lines
2670 that used bh_ and bheader_.
2672 7. Added the _cache feature to named lists.
2674 8. The code for checking quota_filecount in the appendfile transport was
2675 allowing one more file than it should have been.
2677 9. For compatibility with Sendmail, the command line option
2685 and sets the incoming protocol and host name (for trusted callers). The
2686 host name and its colon can be omitted when only the protocol is to be set.
2687 Note the Exim already has two private options, -pd and -ps, that refer to
2688 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2689 "s", but I don't think that's a major issue.
2691 10. A number of refactoring changes to the code, none of which should affect
2694 (a) The number of logging options was getting close to filling up the
2695 32-bit word that was used as a bit map. I have split them into two classes:
2696 those that are passed in the argument to log_write(), and those that are
2697 only ever tested independently outside of that function. These are now in
2698 separate 32-bit words, so there is plenty of room for expansion again.
2699 There is no change in the user interface or the logging behaviour.
2701 (b) When building, for example, log lines, the code previously used a
2702 macro that called string_cat() twice, in order to add two strings. This is
2703 not really sufficiently general. Furthermore, there was one instance where
2704 it was actually wrong because one of the argument was used twice, and in
2705 one call a function was used. (As it happened, calling the function twice
2706 did not affect the overall behaviour.) The macro has been replaced by a
2707 function that can join an arbitrary number of extra strings onto a growing
2710 (c) The code for expansion conditions now uses a table and a binary chop
2711 instead of a serial search (which was left over from when there were very
2712 few conditions). Also, it now recognizes conditions like "pam" even when
2713 the relevant support is not compiled in: a suitably worded error message is
2714 given if an attempt is made to use such a condition.
2716 11. Added ${time_interval:xxxxx}.
2718 12. A bug was causing one of the ddress fields not to be passed back correctly
2719 from remote delivery subprocesses. The field in question was not being
2720 subsequently used, so this caused to problems in practice.
2722 13. Added new log selectors queue_time and deliver_time.
2724 14. Might have fixed a bug in maildirsizefile handling that threw up
2725 "unexpected character" debug warnings, and recalculated the data
2726 unnecessarily. In any case, I expanded the warning message to give more
2729 15. Added the message "Restricted characters in address" to the statements in
2730 the default ACL that block characters like @ and % in local parts.
2732 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2733 Three changes have been made:
2735 (a) There was a serious bug; a negative response to MAIL caused the whole
2736 recipient domain to be cached as invalid, thereby blocking all messages
2737 to all local parts at the same domain, from all senders. This bug has
2738 been fixed. The domain is no longer cached after a negative response to
2739 MAIL if the sender used is not empty.
2741 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2744 (c) A new callout option, "use_sender" has been added for people who want
2745 the modified behaviour.
2751 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2754 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2755 indeed breaks things for older releases.
2757 3. Added additional logging to the case where there is a problem reading data
2758 from a filter that is running in a subprocess using a pipe, in order to
2759 try to track down a specific problem.
2761 4. Testing facility fudge: when running in the test harness and attempting
2762 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2763 getting "No route to host". Convert this to a timeout.
2765 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2768 6. Some OS don't have socklen_t but use size_t instead. This affects the
2769 fifth argument of getsockopt() amongst other things. This is now
2770 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2771 can be set for individual OS. I have set it for SunOS5, OSF1, and
2772 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2773 some earlier ones do not.
2775 7. Change 4.30/15 was not doing the test caselessly.
2777 8. The standard form for an IPv6 address literal was being rejected by address
2778 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2779 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2780 this, as well as the form without the "IPv6" on the front (but only when
2781 address literals are enabled, of course).
2783 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2785 10. Exim crashed if a message with an empty sender address specified by -f
2786 encountered a router with an errors_to setting. This could be provoked only
2787 by a command such as
2791 where an empty string was supplied; "<>" did not hit this bug.
2793 11. Installed PCRE release 4.5.
2795 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2796 remained set. It is now erased.
2798 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2799 times from message ids (which are base 36 rather than the normal 62).
2801 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2802 were being counted as actual protocol errors, and logged if the log
2803 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2804 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2805 and DATA following a set of rejected RCPTs do not count as protocol errors.
2806 In other words, Exim assumes they were pipelined, though this may not
2807 actually be the case. Of course, in all cases the client gets an
2808 appropriate error code.
2810 15. If a lookup fails in an ACL condition, a message about the failure may
2811 be available; it is used if testing the ACL cannot continue, because most
2812 such messages specify what the cause of the deferral is. However, some
2813 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2814 that caused an old message to be retained and used if a later statement
2815 caused a defer, replacing the real cause of the deferral.
2817 16. If an IP address had so many PTR records that the DNS lookup buffer
2818 was not large enough to hold them, Exim could crash while trying to process
2819 the truncated data. It now detects and logs this case.
2821 17. Further to 4.21/58, another change has been made: if (and only if) the
2822 first line of a message (the first header line) ends with CRLF, a bare LF
2823 in a subsequent header line has a space inserted after it, so as not to
2824 terminate the header.
2826 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2827 unnecessarily, used atoi() instead of strtol(), and didn't check the
2828 termination when getting file sizes from file names by regex.
2830 19. Completely re-implemented the support for maildirsize files, in the light
2831 of a number of problems with the previous contributed implementation
2832 (4.30/29). In particular:
2834 . If the quota is zero, the maildirsize file is maintained, but no quota is
2837 . If the maildir directory does not exist, it is created before any attempt
2838 to write a maildirsize file.
2840 . The quota value in the file is just a cache; if the quota is changed in
2841 the transport, the new value overrides.
2843 . A regular expression is available for excluding directories from the
2846 20. The autoreply transport checks the characters in options that define the
2847 message's headers; it allows continued headers, but it was checking with
2848 isspace() after an embedded newline instead of explicitly looking for a
2851 21. If all the "regular" hosts to which an address was routed had passed their
2852 expiry times, and had not reached their retry times, the address was
2853 bounced, even if fallback hosts were defined. Now Exim should go on to try
2856 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2857 equivalent code in the SMTP transport. Some hosts send humungous responses
2858 to HELO/EHLO, more than 1024 it seems.
2860 23. Refactoring: code in filter.c used (void *) for "any old type" but this
2861 gives compiler warnings in some environments. I've now done it "properly",
2864 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2865 (because of problems with the built-in one) was declared to return uschar *
2866 instead of char *, causing compiler failure.
2868 25. Fixed a file descriptor leak when processing alias/forward files.
2870 26. Fixed a minor format string issue in dbfn.c.
2872 27. Typo in exim.c: ("dmbnz" for "dbmnz").
2874 28. If a filter file refered to $h_xxx or $message_headers, and the headers
2875 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2878 29. When a sender address is verified, it is cached, to save repeating the test
2879 when there is more than one recipient in a message. However, when the
2880 verification involves a callout, it is possible for different callout
2881 options to be set for different recipients. It is too complicated to keep
2882 track of this in the cache, so now Exim always runs a verification when a
2883 callout is required, relying on the callout cache for the optimization.
2884 The overhead is duplication of the address routing, but this should not be
2887 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
2888 to be verified with callout=postmaster, and the main callout worked but the
2889 postmaster check failed, the verification correctly failed. However, if a
2890 subsequent RCPT command asked for sender verification *without* the
2891 postmaster check, incorrect caching caused this verification also to fail,
2894 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2895 it was not caching the DNS options (qualify_single, search_parents) that
2896 were used when the lookup failed. A subsequent lookup with different
2897 options therefore always gave the same answer, though there were cases
2898 where it should not have. (Example: a "domains = !$mx_any" option on a
2899 dnslookup router: the "domains" option is always processed without any
2900 widening, but the router might have qualify_single set.) Now Exim uses the
2901 cached value only when the same options are set.
2903 32. Added John Jetmore's "exipick" utility to the distribution.
2905 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
2906 than a timeout (e.g. a certificate is required, and is not provided), an
2907 Exim server now closes the connection immediately. Previously it waited for
2908 the client to close - but if the client is SSL, it seems that they each
2909 wait for each other, leading to a delay before one of them times out.
2911 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2912 maintained 0.8.x compatibility because I don't think many are using it, and
2913 it is clearly obsolete.
2915 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2918 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2919 client certificate was expired. A simple patch fixes this, though I don't
2920 understand the full logic of why the verify callback is called multiple
2923 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2924 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2925 which causes problems with some clients (such as the Certicom SSL Plus
2926 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2927 disables the coutermeasure allowing Eudora to connect."
2929 38. Exim was not checking that a write() to a log file succeeded. This could
2930 lead to Bad Things if a log got too big, in particular if it hit a file
2931 size limit. Exim now panics and dies if it cannot write to a log file, just
2932 as it does if it cannot open a log file.
2934 39. Modified OS/Makefile-Linux so that it now contains
2936 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2938 The two -D definitions ensure that Exim is compiled with large file
2939 support, which makes it possible to handle log files that are bigger than
2942 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2943 instance) a domain was checked against a named list that involved a lookup,
2944 causing $domain_data to be set, then another domain was checked against the
2945 same list, then the first domain was re-checked, the value of $domain_data
2946 after the final check could be wrong. In particular, if the second check
2947 failed, it could be set empty. This bug probably also applied to
2950 41. The strip_trailing_dot option was not being applied to the address given
2951 with the -f command-line option.
2953 42. The code for reading a message's header from the spool was incrementing
2954 $received_count, but never initializing it. This meant that the value was
2955 incorrect (doubled) while delivering a message in the same process in which
2956 it was received. In the most common configuration of Exim, this never
2957 happens - a fresh exec is done - but it can happen when
2958 deliver_drop_privilege is set.
2960 43. When Exim logs an SMTP synchronization error - client data sent too soon -
2961 it now includes up to 150 characters of the unexpected data in the log
2964 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2965 and building data strings. The size of both of these buffers was 10 000
2966 bytes - far larger than anybody would *ever* want, thought I. Needless to
2967 say, somebody hit the limit. I have increased the maximum line length to
2968 20 000 and the maximum data length of concatenated lines to 100 000. I have
2969 also fixed two bugs, because there was no checking on these buffers. Tsk,
2970 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2971 buffer is too small.
2973 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2974 lsearch lookups. Now it does.
2976 46. When parsing a route_list item in a manualroute router, a fixed-length
2977 buffer was used for the list of hosts. I made this 1024 bytes long,
2978 thinking that nobody would ever have a list of hosts that long. Wrong.
2979 Somebody had a whole pile of complicated expansion conditions, and the
2980 string was silently truncated, leading to an expansion error. It turns out
2981 that it is easier to change to an unlimited length (owing to other changes
2982 that have happened since this code was originally written) than to build
2983 structure for giving a limitation error. The length of the item that
2984 expands into the list of hosts is now unlimited.
2986 47. The lsearch lookup could not handle data where the length of text line was
2987 more than 4095 characters. Such lines were truncated, leading to shortened
2988 data being returned. It should now handle lines of any length.
2990 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2991 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2994 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
2995 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2996 no longer happens in comment lines. A list of replacements is now placed
2997 at the head of all of the source files, except those whose only change is
2998 to replace PERL_COMMAND in the very first #! line.
3000 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3001 messages on the queue, with a bottom-up merge sort, using code contributed
3002 by Michael Haardt. This should make operations like -bp somewhat faster on
3003 large queues. It won't affect queue runners, except when queue_run_in_order
3006 51. Installed eximstats 1.31 in the distribution.
3008 52. Added support for SRV lookups to the dnslookup router.
3010 53. If an ACL referred to $message_body or $message_body_end, the value was not
3011 reset for any messages that followed in the same SMTP session.
3013 54. The store-handling optimization for building very long strings was not
3014 differentiating between the different store pools. I don't think this
3015 actually made any difference in practice, but I've tidied it.
3017 55. While running the routers to verify a sender address, $sender_address
3018 was still set to the sender address. This is wrong, because when routing to
3019 send a bounce to the sender, it would be empty. Therefore, I have changed
3020 it so that, while verifying a sender address, $sender_address is set to <>.
3021 (There is no change to what happens when verifying a recipient address.)
3023 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3024 target A or AAAA records (if not already returned) without resetting the
3025 qualify_single or search_parents options of the DNS resolver. These are
3026 inappropriate in this case because the targets of MX and SRV records must
3027 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3028 target that, when qualified, matched something in the local domain. These
3029 two options are now turned off when doing these lookups.
3031 57. It seems that at least some releases of Reiserfs (which does not have the
3032 concept of a fixed number of inodes) returns zero and not -1 for the
3033 number of available inodes. This interacted badly with check_spool_inodes,
3034 which assumed that -1 was the "no such thing" setting. What I have done is
3035 to check that the total number of inodes is greater than zero before doing
3036 the test of how many are available.
3038 58. When a "warn" ACL statement has a log_message modifier, the message is
3039 remembered, and not repeated. This is to avoid a lot of repetition when a
3040 message has many recipients that cause the same warning to be written.
3041 Howewer, Exim was preserving the list of already written lines for an
3042 entire SMTP session, which doesn't seem right. The memory is now reset if a
3043 new message is started.
3045 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3046 debugging output unless log_rewrite was also set.
3048 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3049 of (int)(handle) when we know that handle contains (void *)(-1).
3051 61. The Exim daemon panic-logs an error return when it closes the incoming
3052 connection. However "connection reset by peer" seems to be common, and
3053 isn't really an error worthy of noting specially, so that particular error
3056 62. When Exim is trying to find all the local interfaces, it used to panic and
3057 die if the ioctl to get the interface flags failed. However, it seems that
3058 on at least one OS (Solaris 9) it is possible to have an interface that is
3059 included in the list of interfaces, but for which you get a failure error
3060 for this call. This happens when the interface is not "plumbed" into a
3061 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3062 failure of the "get flags" call assumes that the interface is down.
3064 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3065 makes life easier for people who are doing arithmetic on fields extracted
3066 from dates, where you often get leading zeros that should not be
3067 interpreted as octal.
3069 64. Added qualify_domain to the redirect router, to override the global
3072 65. If a pathologically long header line contained very many addresses (the
3073 report of this problem mentioned 10 000) and each of them was rewritten,
3074 Exim could use up a very large amount of memory. (It kept on making new
3075 copies of the header line as it rewrote, and never released the old ones.)
3076 At the expense of a bit more processing, the header rewriting function has
3077 been changed so that it no longer eats memory in this way.
3079 66. The generation of the Received: header has been moved from the time that a
3080 message starts to be received, to the time that it finishes. The timestamp
3081 in the Received: header should now be very close to that of the <= log
3082 line. There are two side-effects of this change:
3084 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3085 logged header lines no longer include the local Received: line, because
3086 it has not yet been created. The same applies to a copy of the message
3087 that is returned to a non-SMTP sender when a message is rejected.
3089 (b) When a filter file is tested using -bf, no additional Received: header
3090 is added to the test message. After some thought, I decided that this
3093 This change does not affect the value of $received_for. It is still set
3094 after address rewriting, but before local_scan() is called.
3096 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3098 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3099 gave an unhelpful panic error message, and a defer error. I have managed to
3100 change this behaviour so that it now rejects any supplied certificate,
3101 which seems right, as the list of acceptable certificates is empty.
3103 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3104 gave an unhelpful defer error. I have not managed to make this reject any
3105 supplied certificates, but the error message it gives is "no certificate
3106 supplied", which is not helpful.
3108 70. exigrep's output now also includes lines that are not associated with any
3109 message, but which match the given pattern. Implemented by a patch from
3110 Martin Sluka, which also tidied up the Perl a bit.
3112 71. Recipient callout verification, like sender verification, was using <> in
3113 the MAIL FROM command. This isn't really the right thing, since the actual
3114 sender may affect whether the remote host accepts the recipient or not. I
3115 have changed it to use the actual sender in the callout; this means that
3116 the cache record is now keyed on a recipient/sender pair, not just the
3117 recipient address. There doesn't seem to be a real danger of callout loops,
3118 since a callout by the remote host to check the sender would use <>.
3119 [SEE ABOVE: changed after hitting problems.]
3121 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3122 temporary errors. However, in the case of such a code being given after
3123 the end of a data transmission (i.e. after ".") Exim was failing to write
3124 a retry record for the message. (Yes, there was some broken host that was
3125 actually sending 8xx at this point.)
3127 73. An unknown lookup type in a host list could cause Exim to panic-die when
3128 the list was checked. (An example that provoked this was putting <; in the
3129 middle of a list instead of at the start.) If this happened during a DATA
3130 ACL check, a -D file could be left lying around. This kind of configuration
3131 error no longer causes Exim to die; instead it causes a defer errror. The
3132 incident is still logged to the main and panic logs.
3134 74. Buglet left over from Exim 3 conversion. The message "too many messages
3135 in one connection" was written to the rejectlog but not the mainlog, except
3136 when address rewriting (yes!) was being logged.
3138 75. Added write_rejectlog option.
3140 76. When a system filter was run not as root (that is, when system_filter_user
3141 was set), the values of the $n variables were not being returned to the
3142 main process; thus, they were not subsequently available in the $sn
3145 77. Added +return_path_on_delivery log selector.
3147 78. A connection timeout was being treated differently from recipients deferred
3148 when testing hosts_max_try with a message that was older than the host's
3149 retry timeout. (The host should not be counted, thus allowing all hosts to
3150 be tried at least once before bouncing.) This may have been the cause of an
3151 occasionally reported bug whereby a message would remain on the queue
3152 longer than the retry timeout, but would be bounced if a delivery was
3153 forced. I say "may" because I never totally pinned down the problem;
3154 setting up timeout/retry tests is difficult. See also the next item.
3156 79. The ultimate address timeout was not being applied to errors that involved
3157 a combination of host plus message (for example, a timeout on a MAIL
3158 command). When an address resolved to a number of possible hosts, and they
3159 were not all tried for each delivery (e.g. because of hosts_max_try), a
3160 message could remain on the queue longer than the retry timeout.
3162 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3165 81. Fixed an obscure SMTP outgoing bug which required at least the following
3166 conditions: (a) there was another message waiting for the same server;
3167 (b) the server returned 5xx to all RCPT commands in the first message so
3168 that the message was not completed; (c) the server dropped the connection
3169 or gave a negative response to the RSET that Exim sends to abort the
3170 transaction. The observed case was a dropped connection after DATA that had
3171 been sent in pipelining mode. That is, the server had advertised PIPELINING
3172 but was not implementing it correctly. The effect of the bug was incorrect
3173 behaviour, such as trying another host, and this could lead to a crash.
3179 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3180 and daemon.c were passed as pointers to ints; they should have been
3181 pointers to socklen_t variables (which are typically unsigned ints).
3183 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3186 3. Fixed a really odd bug that affected only the testing scheme; patching a
3187 certain fixed string in the binary changed the value of another string that
3188 happened to be identical to the end of the original first string.
3190 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3191 name", it returns that address as the IP address. On some operating
3192 systems (e.g. Solaris), it also passes back the IP address string as the
3193 "host name". However, on others (e.g. Linux), it passes back an empty
3194 string. Exim wasn't checking for this, and was changing the host name to an
3195 empty string, assuming it had been canonicized.
3197 5. Although rare, it is permitted to have more than one PTR record for a given
3198 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3199 all the names associated with an address, because they do in Solaris.
3200 However, it seems that they do not in Linux for data that comes from the
3201 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3202 I found this out when I moved to a new Linux workstation and tried to run
3203 the Exim test suite.
3205 To get round this problem I have changed the code so that it now does its
3206 own call to the DNS to look up PTR records when searching for a host name.
3207 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3208 addresses that are only in /etc/hosts are still found.
3210 This behaviour is, however, controlled by an option called host_lookup_
3211 order, which defaults to "bydns:byaddr". If people want to use the other
3212 order, or indeed, just use one or the other means of lookup, they can
3213 specify it in this variable.
3215 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3216 some operating systems, this comes back from gethostbyaddr() as an empty
3217 string, and this is what Exim used to test for. However, it seems that in
3218 other systems, "." is yielded. Exim now tests for this case too.
3220 7. The values of check_spool_space and check_log_space are now held internally
3221 as a number of kilobytes instead of an absolute number of bytes. If a
3222 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3223 kilobyte. This means that much larger values can be stored.
3225 8. Exim monitor: an attempt to get the action menu when not actually pointing
3226 at a message produces an empty menu entitled "No message selected". This
3227 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3228 no entries in it ("Shell widget menu has zero width and/or height"). So I
3229 have added a single, blank menu entry in this case.
3231 9. Added ${quote_local_part.
3233 10. MIME decoding is now applied to the contents of Subject: header lines when
3236 11. Now that a reference to $sender_host_address automatically causes a reverse
3237 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3238 host lookup before query-style lookups in lists that might use this
3239 variable. This has therefore been abolished, and the "net-" prefix is no
3240 longer necessary for query-style lookups.
3242 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3243 have been a typo for LFLAGS, so it has been changed.
3245 13. The install script calls Exim with "-C /dev/null" in order to find the
3246 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3247 to be output. Howeve, since Exim outputs its version number before the
3248 error, it didn't break the script. It just looked ugly. I fixed this by
3249 always allowing "-C /dev/null" if the caller is root.
3251 14. Ignore overlarge ACL variable number when reading spool file - insurance
3252 against a later release with more variables having written the file.
3254 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3255 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3256 this, as well as the form without the "IPv6" on the front.
3258 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3259 OS/Makefile-Darwin file.
3261 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3262 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3264 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3265 variable. However, some OS use ints for inodes. I've added cast to long int
3266 to get rid of the compiler warning.
3268 19. I had forgotten to lock out "/../" in configuration file names when
3269 ALT_CONFIG_PREFIX was set.
3271 20. Routers used for verification do not need to specify transports. However,
3272 if such a router generated a host list, and callout was configured, Exim
3273 crashed, because it could not find a port number from the (non-existent)
3274 transport. It now assumes port 25 in this circumstance.
3276 21. Added the -t option to exigrep.
3278 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3279 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3280 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3281 them is set, LOOKUP_LSEARCH is forced.
3283 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3285 24. Added sender and host information to the "rejected by local_scan()" log
3286 line; previously there was no indication of these.
3288 25. Added .include_if_exists.
3290 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3291 Linux. It turns out that not all file systems support this. Apparently some
3292 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3293 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3294 not supported on the file descriptor, is now ignored when Exim is trying to
3295 sync a directory. This applies only to Linux.
3297 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3299 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3300 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3303 29. Added support for maildirsize files from supplied patch (modified a bit).
3305 30. The use of :fail: followed by an empty string could lead Exim to respond to
3306 sender verification failures with (e.g.):
3308 550 Verification failed for <xxx>
3309 550 Sender verify failed
3311 where the first response line was missing the '-' that indicates it is not
3312 the final line of the response.
3314 31. The loop for finding the name of the user that called Exim had a hardwired
3315 limit of 10; it now uses the value of finduser_retries, which is used for
3316 all other user lookups.
3318 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3321 33. Exim was neglecting to zero errno before one call of strtol() when
3322 expanding a string and expecting an integer value. On some systems this
3323 resulted in spurious "integer overflow" errors. Also, it was casting the
3324 result into an int without checking.
3326 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3327 did not work. The code looks as if it has *never* worked, though it appears
3328 to have been documented since at least releast 1.62. I have made it work.
3330 35. The "timeout_DNS" error in retry rules, also documented since at least
3331 1.62, also never worked. As it isn't clear exactly what this means, and
3332 clearly it isn't a major issue, I have abolished the feature by treating it
3333 as "timeout", and writing a warning to the main and panic logs.
3335 36. The display of retry rules for -brt wasn't always showing the error code
3338 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3339 timeout_connect_A, timeout_connect_MX.
3341 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3342 to the empty sender.
3344 39. The daemon was not analysing the content of -oX till after it had closed
3345 stderr and disconnected from the controlling terminal. This meant that any
3346 syntax errors were only noted on the panic log, and the return code from
3347 the command was 0. By re-arranging the code a little, I've made the
3348 decoding happen first, so such errors now appear on stderr, and the return
3349 code is 1. However, the actual setting up of the sockets still happens in
3350 the disconnected process, so errors there are still only recorded on the
3353 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3354 connections (as happens on some IP stacks) was logged at start up time as
3355 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3356 it from "IPv6 and IPv4", which means that two separate sockets are being
3359 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3360 says whether AF_INET or AF_INET6 was passed as an argument.
3362 42. Exiwhat output was messed up when time zones were included in log
3365 43. Exiwhat now gives more information about the daemon's listening ports,
3366 and whether -tls-on-connect was used.
3368 44. The "port" option of the smtp transport is now expanded.
3370 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3371 silently ignored. Now an error message is written to the main and panic
3374 46. There's a new ACL modifier called "logwrite" which writes to a log file
3375 as soon as it is encountered.
3377 47. Added $local_user_uid and $local_user_gid at routing time.
3379 48. Exim crashed when trying to verify a sender address that was being
3382 49. Exim was recognizing only a space character after ".include". It now also
3383 recognizes a tab character.
3385 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3386 extracting the relevant information from the specification. The man page no
3387 longer contains scrambled data for the -d option, and I've added a section
3388 at the front about calling Exim under different names.
3390 51. Added "extra_headers" argument to the "mail" command in filter files.
3392 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3395 53. Installed eximstats 1.29.
3397 54. Added transport_filter_timeout as a generic transport option.
3399 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3400 Cc: header lines. This was required by RFC 822, but it not required by RFC
3403 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3404 incoming messages that did not have them. Now it does so only if the
3405 message originates locally, that is, if there is no associated remote host
3406 address. When Resent- header lines are present, this applies to the Resent-
3407 lines rather than the non-Resent- lines.
3409 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3410 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3412 58. Messages for configuration errors now include the name of the main
3413 configuration file - useful now that there may be more than one file in a
3414 list (.included file names were always shown).
3416 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3417 for those rare installations that do not start the daemon as root or run it
3418 setuid root. I've cut out the call to initgroups() if the daemon is not
3421 60. The Exim user and group can now be bound into the binary as text strings
3422 that are looked up at the start of Exim's processing.
3424 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3426 62. Added $mailstore_basename variable.
3428 63. Installed patch to sieve.c from Michael Haardt.
3430 64. When Exim failed to open the panic log after failing to open the main log,
3431 the original message it was trying to log was written to stderr and debug
3432 output, but if they were not available (the usual case in production), it
3433 was lost. Now it is written to syslog before the two lines that record the
3434 failures to open the logs.
3436 65. Users' Exim filters run in subprocesses under the user's uid. It is
3437 possible for a "deliver" command or an alias in a "personal" command to
3438 provoke an address rewrite. If logging of address rewriting is configured,
3439 this fails because the process is not running as root or exim. There may be
3440 a better way of dealing with this, but for the moment (because 4.30 needs
3441 to be released), I have disabled address rewrite logging when running a
3442 filter in a non-root, non-exim process.
3448 1. The buildconfig auxiliary program wasn't quoting the value set for
3449 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3450 not defined. This bug was masked in 4.22 by the effect that was fixed in
3453 2. Some messages that were rejected after a message id was allocated were
3454 shown as "incomplete" by exigrep. It no longer does this for messages that
3455 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3457 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3458 have allow_domain_literals set, the ID did not get logged in the <= line.
3459 Domain literals are now always recognized in Message-ID: header lines.
3461 4. The first argument for a ${extract expansion item is the key name or field
3462 number. Leading and trailing spaces in this item were not being ignored,
3463 causing some misleading effects.
3465 5. When deliver_drop_privilege was set, single queue runner processes started
3466 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3467 same command in the process it spins off) were not dropping privilege.
3469 6. When the daemon running as "exim" started a queue runner, it always
3470 re-executed Exim in the spun-off process. This is a waste of effort when
3471 deliver_drop_privilege is set. The new process now just calls the
3472 queue-runner function directly.
3478 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3481 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3482 ignored. Though the use of -oP was forcing the writing of a pid file, it
3483 was always written to the default place.
3485 3. If the message "no IP address found for host xxxx" is generated during
3486 incoming verification, it is now followed by identification of the incoming
3487 connection (so you can more easily find what provoked it).
3489 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3491 5. Added some features to "harden" Exim a bit more against certain attacks:
3493 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3494 be put in Local/Makefile. This is like the never_users runtime option,
3495 but it cannot be overridden. The default setting is "root".
3497 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3498 prefix string with which any file named in a -C command line option
3501 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3502 is retained for -C and -D only if the caller of Exim is root. Without
3503 it, the exim user may also use -C and -D and retain privilege.
3505 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3506 command line option is disabled.
3508 6. Macro names set by the -D option must start with an upper case letter, just
3509 like macro names defined in the configuration file.
3511 7. Added "dereference=" facility to LDAP.
3513 8. Two instances of the typo "uknown" in the source files are fixed.
3515 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3516 the Configure-Makefile script screwed up while processing it.
3518 10. Incorporated PCRE 4.4.
3520 11. The SMTP synchronization check was not operating right at the start of an
3521 SMTP session. For example, it could not catch a HELO sent before the client
3522 waited for the greeting. There is now a check for outstanding input at the
3523 point when the greeting is written. Because of the duplex, asynchronous
3524 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3525 way, but not yet received, when the check is performed.
3527 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3528 on TCP/IP sockets, because this apparently causes some broken clients to
3531 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3532 unchanged) from the Cygwin maintainer.
3534 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3535 was supported instead of testing for mbx. Effectively a typo.
3537 15. The spa authenticator server code was not checking that the input it
3538 received was valid base64.
3540 16. The debug output line for the "set" modifier in ACLs was not showing the
3541 name of the variable that was being set.
3543 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3545 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3546 lookup on its own. Something else was needed to trigger the lookup. For
3547 example, a match in host_lookup or the need for a host name in a host list.
3548 Now, if $sender_host_name is referenced and the host name has not yet been
3549 looked up, a lookup is performed. If the lookup fails, the variable remains
3550 empty, and $host_lookup_failed is set to "1".
3552 19. Added "eqi" as a case-independent comparison operator.
3554 20. The saslauthd authentication condition could segfault if neither service
3555 nor realm was specified.
3557 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3558 error message that was logged was misleading, and incoming SMTP
3559 connections were dropped. The message is now more accurate, and temporary
3560 errors are given to SMTP connections.
3562 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3563 (except RSET) with a 421 error, until QUIT is received. However, it was
3564 failing to send a response to QUIT.
3566 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3567 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3568 ACL is now run after the helo_try_verify_hosts code.
3570 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3571 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3572 case-independent, so other case variants are also recognized). Apparently
3573 some systems use these upper case variants.
3575 25. If more than two messages were waiting for the same host, and a transport
3576 filter was specified for the transport, Exim sent two messages over the
3577 same TCP/IP connection, and then failed with "socket operation on non-
3578 socket" when it tried to send the third.
3580 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3582 27. The extern definition of crypt16() in expand.c was not being excluded when
3583 the OS had its own crypt16() function.
3585 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3586 as a preferred synonym for return_size_limit, both as an option and as an
3589 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3591 30. Changed the default configuration ACL to relax the local part checking rule
3592 for addresses that are not in any local domains. For these addresses,
3593 slashes and pipe symbols are allowed within local parts, but the sequence
3594 /../ is explicitly forbidden.
3596 31. SPA server authentication was not clearing the challenge buffer before
3599 32. log_message in a "warn" ACL statement was writing to the reject log as
3600 well as to the main log, which contradicts the documentation and doesn't
3601 seem right (because no rejection is happening). So I have stopped it.
3603 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3604 However, I am unable to do any testing of this.
3606 34. Fixed an infelicity in the appendfile transport. When checking directories
3607 for a mailbox, to see if any needed to be created, it was accidentally
3608 using path names with one or more superfluous leading slashes; tracing
3609 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3611 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3612 contents are added to the log line that is written for every discarded
3613 recipient. (Previously a log_message setting was ignored.)
3615 36. The ${quote: operator now quotes the string if it is empty.
3617 37. The install script runs exim in order to find its version number. If for
3618 some reason other than non-existence or emptiness, which it checks, it
3619 could not run './exim', it was installing it with an empty version number,
3620 i.e. as "exim-". This error state is now caught, and the installation is
3623 38. An argument was missing from the function that creates an error message
3624 when Exim fails to connect to the socket for saslauthd authentication.
3625 This could cause Exim to crash, or give a corrupted message.
3627 39. Added isip, isip4, and isip6 to ${if conditions.
3629 40. The ACL variables $acl_xx are now saved with the message, and can be
3630 accessed later in routers, transports, and filters.
3632 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3633 strings in the file are not string-expanded.
3635 42. If a MAIL command specified a SIZE value that was too large to fit into an
3636 int variable, the check against message_size_limit failed. Such values are
3637 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3638 day this will have to be increased, but I don't think I want to be around
3639 when emails are that large.
3646 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3647 iconv() is not standard in FreeBSD.
3649 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3650 IPv6 enabled. The observed symptom was a segmentation fault on return from
3651 the function os_common_find_running_interfaces() in src/os.c.
3653 3. In the check_special_case() function in daemon.c I had used "errno" as an
3654 argument name, which causes warnings on some systems. This was basically a
3655 typo, since it was named "eno" in the comments!
3657 4. The code that waits for the clock to tick (at a resolution of some fraction
3658 of a second) so as to ensure message-id uniqueness was always waiting for
3659 at least one whole tick, when it could have waited for less. [This is
3660 almost certainly not relevant at current processor speeds, where it is
3661 unlikely to ever wait at all. But we try to future-proof.]
3663 5. The function that sleeps for a time interval that includes fractions of a
3664 second contained a race. It did not block SIGALRM between setting the
3665 timer, and suspending (a couple of lines later). If the interval was short
3666 and the sigsuspend() was delayed until after it had expired, the suspension
3667 never ended. On busy systems this could lead to processes getting stuck for
3670 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3671 process, before it forks any delivery processes. The open lookup caching
3672 mechanism meant that the open file or database connection was passed into
3673 the delivery process. The problem was that delivery processes always tidy
3674 up cached lookup data. This could cause a problem for the next delivery
3675 process started by the queue runner, because the external queue runner
3676 process does not know about the closure. So the next delivery process
3677 still has data in the lookup cache. In the case of a file lookup, there was
3678 no problem because closing a file descriptor in a subprocess doesn't affect
3679 the parent. However, if the lookup was caching a connection to a database,
3680 the connection was closed, and the second delivery process was likely to
3681 see errors such as "PGSQL: query failed: server closed the connection
3682 unexpectedly". The problem has been fixed by closing all cached lookups
3683 in a queue runner before running a delivery process.
3685 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3686 seem to have the "const" qualifier which it has on other OS. I've
3689 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3690 *of the same type* (client or server) with the same public name that an
3691 error should be diagnosed.
3693 9. When Exim looked up a host name for an IP address, but failed to find the
3694 original IP address when looking up the host name (a safety check), it
3695 output the message "<ip address> does not match any IP for NULL", which was
3696 confusing, to say the least. The bug was that the host name should have
3697 appeared instead of "NULL".
3699 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3700 user that is built into the binary, and the -C or -D options is used, root
3701 privilege is dropped before the configuration file is read. In addition,
3702 logging is switched to stderr instead of the normal log files. If the
3703 configuration then re-defines the Exim user, the unprivileged environment
3704 is probably not what is expected, so Exim logs a panic warning message (but
3707 However, if deliver_drop_privilege is set, the unprivileged state may well
3708 be exactly what is intended, so the warning has been cut out in that case,
3709 and Exim is allowed to try to write to its normal log files.
3715 1. smtp_return_error_details was not giving details for temporary sender
3716 or receiver verification errors.
3718 2. Diagnose a configuration error if two authenticators have the same public
3721 3. Exim used not to create the message log file for a message until the first
3722 delivery attempt. This could be confusing when incoming messages were held
3723 for policy or load reasons. The message log file is now created at the time
3724 the message is received, and an initial "Received" line is written to it.
3726 4. The automatically generated man page for command line options had a minor
3727 bug that caused no ill effects; however, a more serious problem was that
3728 the procedure for building the man page automatically didn't always
3729 operate. Consequently, release 4.20 contains an out-of-date version. This
3730 shouldn't happen again.
3732 5. When building Exim with embedded Perl support, the script that builds the
3733 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3734 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3736 6. The freeze_tell option was not being used for messages that were frozen on
3737 arrival, either by an ACL or by local_scan().
3739 7. Added the smtp_incomplete_transaction log selector.
3741 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3742 was accepting AUTH without a new EHLO.
3744 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3745 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3748 10. Logging of TCP/IP connections (when configured) now happens in the main
3749 daemon process instead of the child process, so that the TCP/IP connection
3750 count is more accurate (but it can never be perfect).
3752 11. The use of "drop" in a nested ACL was not being handled correctly in the
3753 outer ACL. Now, if condition failure induced by the nested "drop" causes
3754 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3755 or "require"), the connection is dropped.
3757 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3758 that yield "discard" can now be used with an "accept" or a "discard" verb,
3759 but an error is generated for any others (because I can't see a useful way
3760 to define what should happen).
3762 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3763 are now processed in the same way as lines in the Exim configuration file.
3764 In particular, continuation lines are supported.
3766 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3768 15. Added -ti meaning -t -i.
3770 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3771 domains, and warn by logging if others are found.
3773 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3774 returned from GIFCONF when Exim is trying to find the list of interfaces on
3775 a host. The code in os.c has been modified to copy each ifreq to an aligned
3776 structure in all cases.
3778 Also, in some cases, the returned ifreq's were being copied to a 'struct
3779 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3780 means the last couple of bytes of an IPv6 address could be chopped if the
3781 ifreq contained only a normal sockaddr (14 bytes storage).
3783 18. Named domain lists were not supported in the hosts_treat_as_local option.
3784 An entry such as +xxxx was not recognized, and was treated as a literal
3787 19. Ensure that header lines added by a DATA ACL are included in the reject log
3788 if the ACL subsequently rejects the message.
3790 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3791 MD5 (which is deprecated).
3793 21. When testing a filter file using -bf, Exim was writing a message when it
3794 took the sender from a "From " line in the message, but it was not doing so
3795 when it took $return_path from a Return-Path: header line. It now does.
3797 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3798 with a valid header line field name (a series of printing characters
3799 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3801 23. Changed "disc" in the source to "disk" to conform to the documentation and
3802 the book and for uniformity.
3804 24. Ignore Sendmail's -Ooption=value command line item.
3806 25. When execve() failed while trying to run a command in a pipe transport,
3807 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3808 could be confused with a return value of 69 from the command itself. This
3809 has been changed to 127, the value the shell returns if it is asked to run
3810 a non-existent command. The wording for the related log line suggests a
3811 non-existent command as the problem.
3813 26. If received_header_text expands to an empty string, do not add a Received:
3814 header line to the message. (Well, it adds a token one on the spool, but
3815 marks it "old" so that it doesn't get used or transmitted.)
3817 27. Installed eximstats 1.28 (addition of -nt option).
3819 28. There was no check for failure on the call to getsockname() in the daemon
3820 code. This can fail if there is a shortage of resources on the system, with
3821 ENOMEM, for example. A temporary error is now given on failure.
3823 29. Contrary to the C standard, it seems that in some environments, the
3824 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3825 program. Exim now does this explicitly; it affects the formatting of
3826 timestamps using strftime().
3828 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3829 complaints. I've now initialized all the variables, to avoid this.
3831 32. Header lines added by a system filter were not being "seen" during
3832 transport-time rewrites.
3834 33. The info_callback() function passed to OpenSSL is set up with type void
3835 (*)(SSL *, int, int), as described somewhere. However, when calling the
3836 function (actually a macro) that sets it up, the type void(*)() is
3837 expected. I've put in a cast to prevent warnings from picky compilers.
3839 34. If a DNS black list lookup found a CNAME record, but there were no A
3840 records associated with the domain it pointed at, Exim crashed.
3842 35. If a DNS black list lookup returned more than one A record, Exim ignored
3843 all but the first. It now scans all returned addresses if a particular IP
3844 value is being sought. In this situation, the contents of the
3845 $dnslist_value variable are a list of all the addresses, separated by a
3848 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3849 to accept a host name and all its aliases if the forward lookup for any of
3850 them yielded the IP address of the incoming connection. Now it accepts only
3851 those names whose forward lookup yields the correct IP address. Any other
3852 names are discarded. This closes a loophole whereby a rogue DNS
3853 administrator could create reverse DNS records to break through a
3854 wildcarded host restriction in an ACL.
3856 37. If a user filter or a system filter that ran in a subprocess used any of
3857 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3858 the wrong values were passed to the pipe command ($thisaddress had the
3859 value of $0, $0 had the value of $1, etc). This bug was introduced by
3860 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3862 38. Improved the line breaking for long SMTP error messages from ACLs.
3863 Previously, if there was no break point between 40 and 75 characters, Exim
3864 left the rest of the message alone. Two changes have been made: (a) I've
3865 reduced the minimum length to 35 characters; (b) if it can't find a break
3866 point between 35 and 75 characters, it looks ahead and uses the first one
3867 that it finds. This may give the occasional overlong line, but at least the
3868 remaining text gets split now.
3870 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3871 file descriptors might be low, and that setting 1000 would always raise it.
3872 It turns out that in some environments, the limit is already over 1000 and
3873 that lowering it causes trouble. So now Exim takes care not to decrease it.
3875 40. When delivering a message, the value of $return_path is set to $sender_
3876 address at the start of routing (routers may change the value). By an
3877 oversight, this default was not being set up when an address was tested by
3878 -bt or -bv, which affected the outcome if any router or filter referred to
3881 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
3882 log only when "message" or "log_message" are set. However, if one of the
3883 conditions was an address verification, or a call to a nested ACL, the
3884 messages generated by the underlying test were being passed through. This
3885 no longer happens. The underlying message is available in $acl_verify_
3886 message for both "message" and "log_message" expansions, so it can be
3887 passed through if needed.
3889 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3890 new expansion $bh_ to give the encoded byte string without charset
3891 translation. Translation happens only if iconv() is available; HAVE_ICONV
3892 indicates this at build time. HEADERS_CHARSET gives the charset to
3893 translate to; headers_charset can change it in the configuration, and
3894 "headers charset" can change it in an individual filter file.
3896 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
3897 that creates RFC 2047 encoded "words" labels them as that charset instead
3898 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3899 expansion operator; (ii) when Exim creates a From: line for a local
3900 message; (iii) when a header line is rewritten to include a "phrase" part.
3902 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3903 buggy, causing it to skip the first lines of messages whose message ID
3904 ended in 'D'. This would not have bitten before Exim release 4.14, because
3905 message IDs were unlikely to end in 'D' before then. The effect was to have
3906 incorrect size information for certain domains.
3908 45. #include "config.h" was missing at the start of the crypt16.c module. This
3909 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3912 46. If there was a timeout during a "random" callout check, Exim treated it as
3913 a failure of the random address, and carried on sending RSET and the real
3914 address. If the delay was just some slowness somewhere, the response to the
3915 original RCPT would be taken as a response to RSET and so on, causing
3916 mayhem of various kinds.
3918 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3919 when I implemented it. It didn't allow for the fact that some option values
3920 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3921 the right test for positive values.
3923 48. Domain names in DNS records are case-independent. Exim always looks them up
3924 in lower case. Some resolvers return domain names in exactly the case they
3925 appear in the zone file, that is, they may contain uppercase letters. Not
3926 all resolvers do this - some return always lower case. Exim was treating a
3927 change of case by a resolver as a change of domain, similar to a widening
3928 of a domain abbreviation. This triggered its re-routing code and so it was
3929 trying to route what was effectively the same domain again. This normally
3930 caused routing to fail (because the router wouldn't handle the domain
3931 twice). Now Exim checks for this case specially, and just changes the
3932 casing of the domain that it ultimately uses when it transmits the message
3935 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3938 50. If a filter generated a file delivery with a non-absolute name (possible if
3939 no home directory exists for the router), the forbid_file option was not
3942 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
3943 the existing equality matching.
3945 52. Exim was using ints instead of ino_t variables in some places where it was
3946 dealing with inode numbers.
3948 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3949 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3950 called TMPDIR, and if it finds it is different, it changes its value.
3952 54. The smtp_printf() function is now made available to local_scan() so
3953 additional output lines can be written before returning. There is also an
3954 smtp_fflush() function to enable the detection of a dropped connection.
3955 The variables smtp_input and smtp_batched_input are exported to
3958 55. Changed the default runtime configuration: the message "Unknown user"
3959 has been removed from the ACL, and instead placed on the localuser router,
3960 using the cannot_route_message feature. This means that any verification
3961 failures that generate their own messages won't get overridden. Similarly,
3962 the "Unrouteable address" message that was in the ACL for unverifiable
3963 relay addresses has also been removed.
3965 56. Added hosts_avoid_esmtp to the smtp transport.
3967 57. The exicyclog script was not checking for the esoteric option
3968 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
3969 will work only if exicyclog is run under the appropriate euid.
3971 58. Following a discussion on the list, the rules by which Exim recognises line
3972 endings on incoming messages have been changed. The -dropcr and drop_cr
3973 options are now no-ops, retained only for backwards compatibility. The
3974 following line terminators are recognized: LF CRLF CR. However, special
3975 processing applies to CR:
3977 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
3978 nor a local message in the state where . is a terminator.
3980 (ii) If a bare CR is encountered in a header line, an extra space is added
3981 after the line terminator so as not to end the header. The reasoning
3982 behind this is that bare CRs in header lines are most likely either
3983 to be mistakes, or people trying to play silly games.
3985 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
3986 was being incorrectly given as 18 bytes larger than it should have been.
3987 This is a VOB (very old bug).
3989 60. This may never have affected anything current, but just in case it has:
3990 When the local host is found other than at the start of a list of hosts,
3991 the local host, those with the same MX, and any that follow, are discarded.
3992 When the list in question was part of a longer list of hosts, the following
3993 hosts (not currently being processed) were also being discarded. This no
3994 longer happens. I'm not sure if this situation could ever has previously
3997 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4000 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4001 header. This is something that is recommended in a new Internet Draft, and
4002 is something that is documented as being done by Sendmail. There are two
4003 possible values. For messages generated by the autoreply transport, Exim
4006 Auto-Submitted: auto-replied
4008 whereas for all other generated messages (e.g. bounces) it adds
4010 Auto-Submitted: auto-generated
4012 63. The "personal" condition in filters now includes a test for the
4013 Auto-Submitted: header. If it contains the string "auto-" the message it
4014 not considered personal.
4016 64. Added rcpt_include_affixes as a generic transport option.
4018 65. Added queue_only_override (default true).
4020 66. Added the syslog_duplication option.
4022 67. If what should have been the first header line of a message consisted of
4023 a space followed by a colon, Exim was mis-interpreting it as a header line.
4024 It isn't of course - it is syntactically invalid and should therefore be
4025 treated as the start of the message body. The misbehaviour could have
4026 caused a number of strange effects, including loss of data in subsequent
4027 header lines, and spool format errors.
4029 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4030 client host had authenticated. This control can now be exercised by an ACL
4031 for more flexibility.
4033 69. By default, callouts do not happen when testing with -bh. There is now a
4034 variant, -bhc, which does actually run the callout code, including
4035 consulting and updating the callout cache.
4037 70. Added support for saslauthd authentication, courtesy of Alexander
4040 71. If statvfs() failed on the spool or log directories while checking their
4041 size for availability, Exim confusingly gave the error "space shortage".
4042 Furthermore, in debugging mode it crashed with a floating point exception.
4043 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4044 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4045 serious problem, Exim now writes to the main and panic logs when this
4046 happens, with details of the failure. It then refuses to accept the
4047 incoming message, giving the message "spool directory problem" or "log
4048 directory problem" with a 421 code for SMTP messages.
4050 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4051 0, 1, and 2 exist, because some OS complain for execs without them (see
4052 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4053 descriptors. However, the code omitted to check that the open succeeded,
4054 causing mysterious errors if for some reason the permissions on /dev/null
4055 got screwed. Now Exim writes a message to the main and panic logs, and
4056 bombs out if it can't open /dev/null.
4058 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4059 interact so that it is all more flexible. It is supposed to remain
4060 backwards compatible. Also added extra_local_interfaces.
4062 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4063 to bomb out with an assertion failure - to the client this appears as a
4064 connection drop. This problem occurs in the part of the code that was taken
4065 from the Samba project. Fortunately, the assertion is in a very simple
4066 function, so I have fixed this by reproducing the function inline in the
4067 one place where it is called, and arranging for authentication to fail
4068 instead of killing the process with assert().
4070 75. The SPA client code was not working when the server requested OEM rather
4071 than Unicode encoding.
4073 76. Added code to make require_files with a specific uid setting more usable in
4074 the case where statting the file as root fails - usually a non-root-mounted
4075 NFS file system. When this happens and the failure is EACCES, Exim now
4076 forks a subprocess and does the per-uid checking as the relevant uid.
4078 77. Added process_log_path.
4080 78. If log_file_path was not explicitly set, a setting of check_log_space or
4081 check_log_inodes was ignored.
4083 79. If a space check for the spool or log partitions fails, the incident is now
4084 logged. Of course, in the latter case the data may get lost...
4086 80. Added the %p formatting code to string_format() so that it can be used to
4087 print addresses in debug_print(). Adjusted all the address printing in the
4088 debugging in store.c to use %p rather than %d.
4090 81. There was a concern that a line of code in smtp_in.c could overflow a
4091 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4092 initially expressed, the concern was not well-founded, because trailing
4093 spaces are removed early. However, if the trailing spaces were followed by
4094 a NULL, they did not get removed, so the overflow was possible. Two fixes
4097 (a) I re-wrote the offending code in a cleaner fashion.
4098 (b) If an incoming SMTP command contains a NULL character, it is rejected
4101 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4102 runs initgroups(), so that if the Exim user is in any additional groups,
4103 they will be used during message reception.
4109 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / jgh / exim.git / blob