1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.378 2006/07/14 14:42:57 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.
125 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
126 other effects) broke the use of negated acl sub-conditions.
128 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
131 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
132 "Deny" causes Exim to reject the incoming connection with a 554 error.
133 Unfortunately, if there is a major crisis, such as a disk failure,
134 tcp-wrappers gives "deny", whereas what one would like would be some
135 kind of temporary error. A kludge has been added to help with this.
136 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
137 554 error is used if errno is still zero or contains ENOENT (which occurs
138 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
141 PH/03 Add -lutil to the default FreeBSD LIBS setting.
143 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
144 errors. Otherwise a message that provokes a temporary error (when other
145 messages do not) can cause a whole host to time out.
147 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
148 addresses were routed directly to files or pipes from a redirect router.
149 File deliveries just didn't batch; pipe deliveries might have suffered
152 PH/06 A failure to get a lock for a hints database would erroneously always say
153 "Failed to get write lock", even when it was really a read lock.
155 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
156 of 0600. This has been changed to use the value of the lockfile_mode
157 option (which defaults to 0600).
159 PH/08 Applied small patch from the Sieve maintainer.
161 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
162 folder from quota calculations, a direct delivery into this folder messed
163 up the contents of the maildirsize file. This was because the regex was
164 used only to exclude .Trash (or whatever) when the size of the mailbox
165 was calculated. There was no check that a delivery was happening into an
166 excluded directory. This bug has been fixed by ignoring all quota
167 processing for deliveries into excluded directories.
169 PH/10 Added the maildirfolder_create_regex option to appendfile.
175 PH/01 The code for finding all the local interface addresses on a FreeBSD
176 system running IPv6 was broken. This may well have applied to all BSD
177 systems, as well as to others that have similar system calls. The broken
178 code found IPv4 interfaces correctly, but gave incorrect values for the
179 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
180 that it would not match correctly against @[] and not recognize the IPv6
183 PH/02 The ipliteral router was not recognizing addresses of the form user@
184 [ipv6:....] because it didn't know about the "ipv6:" prefix.
186 PH/03 Added disable_ipv6.
188 PH/04 Changed $reply_address to use the raw form of the headers instead of the
189 decoded form, because it is most often used to construct To: headers
190 lines in autoreplies, and the decoded form may well be syntactically
191 invalid. However, $reply_address has leading white space removed, and all
192 newlines turned into spaces so that the autoreply transport does not
195 PH/05 If group was specified without a user on a router, and no group or user
196 was specified on a transport, the group from the router was ignored.
198 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
199 for visible compile-time settings that can be used to change these
200 numbers, for those that want even more. Backwards compatibility with old
201 spool files has been maintained. However, going back to a previous Exim
202 release will lost any variables that are in spool files.
204 PH/07 Two small changes when running in the test harness: increase delay when
205 passing a TCP/IP connection to a new process, in case the original
206 process has to generate a bounce, and remove special handling of
207 127.0.0.2 (sic), which is no longer necessary.
209 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
210 be the same on different OS.
212 PH/09 Moved a debug statement in filter processing to avoid a race problem when
215 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
216 whether --show-vars was specified or not
218 JJ/02 exipick: Added support for new ACL variable spool format introduced
221 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
222 syntactically invalid From: or Reply-to: line, and a filter used this to
223 generate an autoreply, and therefore failed to obtain an address for the
224 autoreply, Exim could try to deliver to a non-existent relative file
225 name, causing unrelated and misleading errors. What now happens is that
226 it logs this as a hard delivery error, but does not attempt to create a
229 PH/11 The exinext utility has a -C option for testing purposes, but although
230 the given file was scanned by exinext itself; it wasn't being passed on
231 when Exim was called.
233 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
234 an end-of-file indication when reading a command response.
236 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
237 compiled. In many other places in Exim, IPv6 addresses are always
238 recognized, so I have changed this. It also means that IPv4 domain
239 literals of the form [IPV4:n.n.n.n] are now always recognized.
241 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
242 used if the router is not running as root, for example, when verifying at
243 ACL time, or when using -bh. The debugging output from this situation was
244 non-existent - all you got was a failure to exec. I have made two
247 (a) Failures to set uid/gid, the current directory, or a process leader
248 in a subprocess such as that created by queryprogram now generate
249 suitable debugging ouput when -d is set.
251 (b) The queryprogram router detects when it is not running as root,
252 outputs suitable debugging information if -d is set, and then runs
253 the subprocess without attempting to change uid/gid.
255 PH/15 Minor change to Makefile for building test_host (undocumented testing
258 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
259 additional section of a DNS packet that returns MX or SRV records.
260 Instead, it always explicitly searches for A/AAAA records. This avoids
261 major problems that occur when a DNS server includes only records of one
262 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
263 fixed another bug: if SRV records were looked up and the corresponding
264 address records were *not* found in the additional section, the port
265 values from the SRV records were lost.
267 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
268 using the correct key (the original address) when searching the retry
269 rules in order to find which one to use for generating the retry hint.
271 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
272 adding the default one. Similarly, if it contains a Reply-To: header, the
273 errors_reply_to option, if set, is not used.
275 PH/19 When calculating a retry time, Exim used to measure the "time since
276 failure" by looking at the "first failed" field in the retry record. Now
277 it does not use this if it is later than than the arrival time of the
278 message. Instead it uses the arrival time. This makes for better
279 behaviour in cases where some deliveries succeed, thus re-setting the
280 "first failed" field. An example is a quota failure for a huge message
281 when small messages continue to be delivered. Without this change, the
282 "time since failure" will always be short, possible causing more frequent
283 delivery attempts for the huge message than are intended.
284 [Note: This change was subsequently modified - see PH/04 for 4.62.]
286 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
287 $1, $2, $3) because the numerical variables can be reset during some
288 expansion items (e.g. "match"), thereby losing the authentication data.
290 PH/21 Make -bV show the size of off_t variables so that the test suite can
291 decide whether to run tests for quotas > 2G.
293 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
294 mailbox_size, and mailbox_filecount in the appendfile transport. If a
295 filecount value is greater than 2G or if a quota value is greater than 2G
296 on a system where the size of off_t is not greater than 4, a panic error
299 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
300 never match. The debug and -bh output now contains an explicit error
301 message indicating a malformed IPv4 address or mask.
303 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
304 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
307 PH/25 Do not write to syslog when running in the test harness. The only
308 occasion when this arises is a failure to open the main or panic logs
309 (for which there is an explicit test).
311 PH/26 Added the /no_tell option to "control=freeze".
313 PH/27 If a host name lookup failed very early in a connection, for example, if
314 the IP address matched host_lookup and the reverse lookup yielded a name
315 that did not have a forward lookup, an error message of the form "no IP
316 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
317 could be logged. Now it outputs the IP address instead of "NULL".
319 PH/28 An enabling patch from MH: add new function child_open_exim2() which
320 allows the sender and the authenticated sender to be set when
321 submitting a message from within Exim. Since child_open_exim() is
322 documented for local_scan(), the new function should be too.
324 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
325 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
326 results in an empty string is now treated as unset.
328 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
330 PH/31 Added sender_verify_fail logging option.
332 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
333 needed by RFC 822 but not by RFC 2822 was commented out. I have now
334 tidied the source and removed it altogether.
336 PH/33 When a queue run was abandoned because the load average was too high, a
337 log line was always written; now it is written only if the queue_run log
338 selector is set. In addition, the log line for abandonment now contains
339 information about the queue run such as the pid. This is always present
340 in "start" and "stop" lines but was omitted from the "abandon" line.
342 PH/34 Omit spaces between a header name and the colon in the error message that
343 is given when verify = headers_syntax fails (if there are lots of them,
344 the message gets confusing).
346 PH/35 Change the default for dns_check_names_pattern to allow slashes within
347 names, as there are now some PTR records that contain slashes. This check
348 is only to protect against broken name servers that fall over on strange
349 characters, so the fact that it applies to all lookups doesn't matter.
351 PH/36 Now that the new test suite is complete, we can remove some of the
352 special code in Exim that was needed for the old test suite. For example,
353 sorting DNS records because real resolvers return them in an arbitrary
354 order. The new test suite's fake resolver always returns records in the
357 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
358 bounces) except when queue_only is set, to avoid logging races between
359 the different processes.
361 PH/38 Panic-die if .include specifies a non-absolute path.
363 PH/39 A tweak to the "H" retry rule from its user.
365 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
366 a label. They prevented compilation on older perls.
368 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
369 a warning to be raised on newish perls.
371 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
372 on queue. Changes to match documented behaviour of showing count of
373 messages matching specified criteria.
375 PH/40 Changed the default ident timeout from 30s to 5s.
377 PH/41 Added support for the use of login_cap features, on those BSD systems
378 that have them, for controlling the resources used by pipe deliveries.
380 PH/42 The content-scanning code uses fopen() to create files in which to put
381 message data. Previously it was not paying any attention to the mode of
382 the files. Exim runs with umask(0) because the rest of the code creates
383 files with open(), and sets the required mode explicitly. Thus, these
384 files were ending up world-writeable. This was not a big issue, because,
385 being within the spool directory, they were not world-accessible. I have
386 created a function called modefopen, which takes an additional mode
387 argument. It sets umask(777), creates the file, chmods it to the required
388 mode, then resets the umask. All the relevant calls to fopen() in the
389 content scanning code have been changed to use this function.
391 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
392 to 24 hours. This avoids potential overflow problems when processing G
393 and H retry rules. I suspect nobody ever tinkers with this value.
395 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
397 PH/45 When the plaintext authenticator is running as a client, the server's
398 challenges are checked to ensure they are valid base64 strings. By
399 default, the authentication attempt is cancelled if an invalid string is
400 received. Setting client_ignore_invalid_base64 true ignores these errors.
401 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
402 they are received. Thus, the responses can be made to depend on the
403 challenges. If an invalid string is ignored, an empty string is placed in
406 PH/46 Messages that are created by the autoreply transport now contains a
407 References: header, in accordance with RFCs 2822 and 3834.
409 PH/47 Added authenticated_sender_force to the smtp transport.
411 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
413 PH/49 Installed latest patch from the Sieve maintainer.
415 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
416 was also set, the appendfile transport was unnecessarily scanning a
417 directory of message files (e.g. for maildir delivery) to find the count
418 of files (along with the size), even though it did not need this
419 information. It now does the scan only if it needs to find either the
420 size of the count of files.
422 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
424 PH/52 Two bugs concerned with error handling when the smtp transport is
427 (i) Exim was not creating retry information for temporary errors given
428 for individual recipients after the DATA command when the smtp transport
429 was used in LMTP mode. This meant that they could be retried too
430 frequently, and not timed out correctly.
432 (ii) Exim was setting the flag that allows error details to be returned
433 for LMTP errors on RCPT commands, but not for LMTP errors for individual
434 recipients that were returned after the DATA command.
436 PH/53 This is related to PH/52, but is more general: for any failing address,
437 when detailed error information was permitted to be returned to the
438 sender, but the error was temporary, then after the final timeout, only
439 "retry timeout exceeded" was returned. Now it returns the full error as
440 well as "retry timeout exceeded".
442 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
443 do this, and (what is worse) MTAs that accept it.
445 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
446 will now be deprecated.
448 PH/56 New os.c-cygwin from the Cygwin maintainer.
450 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
451 formats (previously only available in exim formats via -bpr, -bpru,
452 and -bpra. Now also available in native and exiqgrep formats)
454 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
455 with very large, slow to parse queues
457 JJ/08 exipick: added ! as generic prefix to negate any criteria format
459 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
461 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
462 responses to authentication challenges, though it was showing the
463 challenges; (ii) I've removed the CR characters from the debug output for
466 PH/58 Allow for the insertion of a newline as well as a space when a string
467 is turned into more than one encoded-word during RFC 2047 encoding. The
468 Sieve code now uses this.
470 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
471 data_4xx, lost_connection, tls_required.
473 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
474 message was being sent as an SMTP response.
476 PH/61 Add -l and -k options to exicyclog.
478 PH/62 When verifying, if an address was redirected to one new address, so that
479 verification continued, and the new address failed or deferred after
480 having set something in $address_data, the value of $address_data was not
481 passed back to the ACL. This was different to the case when no
482 redirection occurred. The value is now passed back in both cases.
484 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
485 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
486 use login_cap.h, so on its own it isn't the distinguishing feature. The
487 new name refers directly to the setclassresources() function.
489 PH/65 Added configuration files for NetBSD3.
491 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
493 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
494 is preferred over IPv4.
496 PH/68 The bounce_return_message and bounce_return_body options were not being
497 honoured for bounces generated during the reception of non-SMTP messages.
498 In particular, this applied to messages rejected by the ACL. This bug has
499 been fixed. However, if bounce_return_message is true and bounce_return_
500 body is false, the headers that are returned for a non-SMTP message
501 include only those that have been read before the error was detected.
502 (In the case of an ACL rejection, they have all been read.)
504 PH/69 The HTML version of the specification is now built in a directory called
505 spec_html instead of spec.html, because the latter looks like a path with
506 a MIME-type, and this confuses some software.
508 PH/70 Catch two compiler warnings in sieve.c.
510 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
511 function verify_get_ident() calls ip_connect() to connect a socket, but
512 if the "connect()" function timed out, ip_connect() used to close the
513 socket. However, verify_get_ident() also closes the socket later, and in
514 between Exim writes to the log, which may get opened at this point. When
515 the socket was closed in ip_connect(), the log could get the same file
516 descriptor number as the socket. This naturally causes chaos. The fix is
517 not to close the socket in ip_connect(); the socket should be closed by
518 the function that creates it. There was only one place in the code where
519 this was missing, in the iplookup router, which I don't think anybody now
520 uses, but I've fixed it anyway.
522 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
523 well as to direct DNS lookups. Otherwise the handling of names in host
524 lists is inconsistent and therefore confusing.
530 PH/01 Two changes to the default runtime configuration:
532 (1) Move the checks for relay_from_hosts and authenticated clients from
533 after to before the (commented out) DNS black list checks.
535 (2) Add control=submission to the relay_from_hosts and authenticated
536 clients checks, on the grounds that messages accepted by these
537 statements are most likely to be submissions.
539 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
541 (1) Generate an error if the third argument for the ${prvs expansion is
544 (2) Treat a missing third argument of ${prvscheck as if it were an empty
547 (3) Reset the variables that are obtained from the first argument of
548 ${prvscheck and used in the second argument before leaving the code,
549 because their memory is reclaimed, so using them afterwards may do
552 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
553 one (it's much easier than Tom thought :-).
555 (5) Because of (4), we can now allow for the use of $prvscheck_result
556 inside the third argument.
558 PH/03 For some reason, the default setting of PATH when running a command from
559 a pipe transport was just "/usr/bin". I have changed it to
562 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
563 anything to be listed in the output from -bV.
565 PH/05 When a filter generated an autoreply, the entire To: header line was
566 quoted in the delivery log line, like this:
568 => >A.N.Other <ano@some.domain> <original@ddress> ...
570 This has been changed so that it extracts the operative address. There
571 may be more than one such address. If so, they are comma-separated, like
574 => >ano@some.domain,ona@other.domain <original@ddress> ...
576 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
577 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
578 not being looked up in the rDNS to get a host name, Exim was showing the
579 IP address twice in Received: lines, even though the IP addresses were
580 identical. For example:
582 Received: from [1.2.3.4] (helo=[1.2.3.4])
584 However, if the real host name was known, it was omitting the HELO data
585 if it matched the actual IP address. This has been tidied up so that it
586 doesn't show the same IP address twice.
588 PH/07 When both +timestamp and +memory debugging was on, the value given by
589 $tod_xxx expansions could be wrong, because the tod_stamp() function was
590 called by the debug printing, thereby overwriting the timestamp buffer.
591 Debugging no longer uses the tod_stamp() function when +timestamp is set.
593 PH/08 When the original message was included in an autoreply transport, it
594 always said "this is a copy of the message, including all the headers",
595 even if body_only or headers_only was set. It now gives an appropriate
598 PH/09 Applied a patch from the Sieve maintainer which:
600 o fixes some comments
601 o adds the (disabled) notify extension core
602 o adds some debug output for the result of if/elsif tests
603 o points to the current vacation draft in the documentation
604 and documents the missing references header update
608 o fixes a bug in processing the envelope test (when testing
609 multiple envelope elements, the last element determinted the
612 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
613 Electronic Mail") by including:
615 Auto-submitted: auto-generated
617 in the messages that it generates (bounce messages and others, such as
618 warnings). In the case of bounce messages for non-SMTP mesages, there was
619 also a typo: it was using "Auto_submitted" (underscore instead of
620 hyphen). Since every message generated by Exim is necessarily in response
621 to another message, thes have all been changed to:
623 Auto-Submitted: auto-replied
625 in accordance with these statements in the RFC:
627 The auto-replied keyword:
629 - SHOULD be used on messages sent in direct response to another
630 message by an automatic process,
632 - MUST NOT be used on manually-generated messages,
634 - MAY be used on Delivery Status Notifications (DSNs) and Message
635 Disposition Notifications (MDNs),
637 - MUST NOT be used on messages generated by automatic or periodic
638 processes, except for messages which are automatic responses to
641 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
642 to the default Received: header definition.
644 PH/12 Added log selector acl_warn_skipped (default on).
646 PH/13 After a successful wildlsearch lookup, discard the values of numeric
647 variables because (a) they are in the wrong storage pool and (b) even if
648 they were copied, it wouldn't work properly because of the caching.
650 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
651 checking when decoding. Apparently there are clients that generate
652 overlong encoded strings. Why am I not surprised?
654 PH/15 If the first argument of "${if match_address" was not empty, but did not
655 contain an "@" character, Exim crashed. Now it writes a panic log message
656 and treats the condition as false.
658 PH/16 In autoreply, treat an empty string for "once" the same as unset.
660 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
661 extension "envelope-auth". The code is finished and in agreement with
662 other implementations, but there is no documentation so far and in fact,
663 nobody wrote the draft yet. This extension is currently #undef'ed, thus
664 not changing the active code.
666 Print executed "if" and "elsif" statements when debugging is used. This
667 helps a great deal to understand what a filter does.
669 Document more things not specified clearly in RFC3028. I had all this
670 sorted out, when out of a sudden new issues came to my mind. Oops."
672 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
675 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
676 canonical form (as documented). However, after a host name lookup from
677 the IP address, check_host() was doing a simple string comparison with
678 addresses acquired from the DNS when checking that the found name did
679 have the original IP as one of its addresses. Since any found IPv6
680 addresses are likely to be in abbreviated form, the comparison could
681 fail. Luckily, there already exists a function for doing the comparison
682 by converting both addresses to binary, so now that is used instead of
685 PH/20 There was another similar case to PH/19, when a complete host name was
686 given in a host list; looking up its IP address could give an abbreviated
687 form, whereas the current host's name might or might not be abbreviated.
688 The same fix has been applied.
694 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
695 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
698 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
699 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
701 PH/03 Typo: missing ".o" in src/pcre/Makefile.
703 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
704 header line, restrict the check to what is listed in RFCs 2369 and 2929.
705 Also, for "Auto-Submitted", treat anything other than "no" as
706 non-personal, in accordance with RFC 3834. (Previously it treated
707 anything starting "auto-" as non-personal.)
709 TF/01 The control=submission/name=... option had a problem with syntax
710 errors if the name included a slash character. The /name= option
711 now slurps the rest of the string, so it can include any characters
712 but it must come last in the list of options (after /sender_retain
715 PH/05 Some modifications to the interface to the fake nameserver for the new
723 TK/01 Added the "success_on_redirect" address verification option. See
724 NewStuff for rationale and an example.
726 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
728 PH/02 Patch to exigrep to allow it to work on syslog lines.
730 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
731 fread() to skip over the body file's header line, because in Cygwin the
732 header line is locked and is inaccessible.
734 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
735 co-exist for some time) to make it clear that it is the Exim ID that is
736 referenced, not the Message-ID: header line.
738 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
739 string_format() function, because snprintf() does not exist on all
742 PH/06 The use of forbid_filter_existstest now also locks out the use of the
743 ${stat: expansion item.
745 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
746 protocol synchronization error", to keep the pedants happy.
748 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
749 well as for IRIX systems, when gcc is being used. See the host.c source
752 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
754 PH/10 Named domain lists were not working if used in a queue_smtp_domains
757 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
758 transport and to the smtp transport in LMTP mode.
760 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
762 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
763 run a filter in a subprocess. This could lead to confusion in subsequent
764 lookups in the parent process. There should also be a search_tidyup() at
765 the end of the subprocess.
767 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
768 only if the host matched helo_try_verify_hosts, which caused the
769 verification to occur when the EHLO/HELO command was issued. The ACL just
770 tested the remembered result. Now, if a previous verification attempt has
771 not happened, "verify = helo" does it there and then.
773 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
775 TK/03 Fix log output including CR from clamd.
777 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
778 exist provoked a memory error which could cause a segfault.
780 PH/15 Installed PCRE 6.2
782 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
784 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
785 of the problem. Specifically, suggested +O2 rather than +O1 for the
788 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
790 PH/20 If a delivery was routed to a non-standard port by means of an SRV
791 record, the port was not correctly logged when the outgoing_port log
792 selector was set (it logged the transort's default port).
794 PH/21 Added support for host-specific ports to manualroute, queryprogram,
795 fallback_hosts, and "hosts" in the smtp transport.
797 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
798 host errors such as "Connection refused".
800 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
801 authentication with radiusclient 0.4.9:
803 - Error returned from rc_read_config was caught wrongly
804 - Username/password not passed on to radius server due to wrong length.
806 The presumption is that some radiusclient API changes for 4.51/PH/17
807 were not taken care of correctly. The code is still untested by me (my
808 Linux distribution still has 0.3.2 of radiusclient), but it was
809 contributed by a Radius user.
811 PH/24 When doing a callout, the value of $domain wasn't set correctly when
812 expanding the "port" option of the smtp transport.
814 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
815 while reading a MIME header. Thanks to Tom Hughes for a patch.
817 PH/24 Include config.h inside local_scan.h so that configuration settings are
820 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
821 that in an ACL for RCPT (for example), you can examine exactly what was
824 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
825 commands, but it was not correctly comparing the address with the actual
826 client host address. Thus, it would show the EHLO address in Received:
827 header lines when this was not necessary.
829 PH/27 Added the % operator to ${eval:}.
831 PH/28 Exim tries to create and chdir to its spool directory when it starts;
832 it should be ignoring failures (because with -C, for example, it has lost
833 privilege). It wasn't ignoring creation failures other than "already
836 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
839 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
840 because an input line was too long, either on its own, or by virtue of
841 too many continuations, the temporary file was not being removed, and the
842 return code was incorrect.
844 PH/31 Missing "BOOL" in function definition in filtertest.c.
846 PH/32 Applied Sieve patches from the maintainer.
848 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
850 PH/33 Added "verify = not_blind".
852 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
853 Local/Makefile (with some defaults set). These are used in built scripts
854 such as exicyclog, but they have never been used in the exim_install
855 script (though there are many overriding facilities there). I have
856 arranged that the exim_install script now takes note of these two
859 PH/35 Installed configuration files for Dragonfly.
861 PH/36 When a locally submitted message by a trusted user did not contain a
862 From: header, and the sender address was obtained from -f or from an SMTP
863 MAIL command, and the trusted user did not use -F to supply a sender
864 name, $originator_name was incorrectly used when constructing a From:
865 header. Furthermore, $originator_name was used for submission mode
866 messages from external hosts without From: headers in a similar way,
867 which is clearly wrong.
869 PH/37 Added control=suppress_local_fixups.
871 PH/38 When log_selector = +received_sender was set, and the addition of the
872 sender made the log line's construction buffer exactly full, or one byte
873 less than full, an overflow happened when the terminating "\n" was
876 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
877 when the result of a list match is failure because a DNS lookup failed.
879 PH/40 RM_COMMAND is now used in the building process.
881 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
882 the "build-* directories that it finds.
884 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
885 address in a domain literal was a prefix of an interface address.
887 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
888 when verifying a sender address, unless rewrite_headers is false.
890 PH/44 Wrote a long comment about why errors_to addresses are verified as
891 recipients, not senders.
893 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
894 the ratelimit ACL was added.
896 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
898 PH/46 Added extra information about PostgreSQL errors to the error string.
900 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
901 avoiding the need to install special zones in a real server. This is
902 backwards compatible; if it can't find the fake resolver, it drops back.
903 Thus, both old and new test suites can be run.
905 TF/02 Added util/ratelimit.pl
907 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
908 clock is set back in time.
910 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
911 Candler <B.Candler@pobox.com>.
913 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
914 OK for addresses that are the result of redirections.
916 PH/48 A number of further additions for the benefit of the new test suite,
917 including a fake gethostbyname() that interfaces to the fake DNS resolver
920 TF/06 The fix for widen_domains has also been applied to qualify_single and
921 search_parents which are the other dnslookup options that can cause
924 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
927 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
929 TF/07 Exim produced the error message "an SRV record indicated no SMTP
930 service" if it encountered an MX record with an empty target hostname.
931 The message is now "an MX or SRV record indicated no SMTP service".
933 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
934 if the DNS of the sending site is misconfigured. This is quite a
935 common situation. This change restores the behaviour of treating a
936 helo verification defer as a failure.
938 PH/51 If self=fail was set on a router, the bounce message did not include the
939 actual error message.
945 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
947 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
948 command itself ended in error, the underlying message about the transport
949 filter timeout was being overwritten with the pipe command error. Now the
950 underlying error message should be appended to the second error message.
952 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
954 PH/02 Reduce the amount of output that "make" produces by default. Full output
955 can still be requested.
957 PH/03 The warning log line about a condition test deferring for a "warn" verb
958 was being output only once per connection, rather than after each
959 occurrence (because it was using the same function as for successful
960 "warn" verbs). This seems wrong, so I have changed it.
962 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
963 it should not have, which might have caused a crash in the right
964 circumstances, but probably never did.
966 PH/04 Installed a modified version of Tony Finch's patch to make submission
967 mode fix the return path as well as the Sender: header line, and to
968 add a /name= option so that you can make the user's friendly name appear
971 TF/03 Added the control = fakedefer ACL modifier.
973 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
974 Mark Lowes for thorough testing.
976 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
978 TK/03 Merged latest SRS patch from Miles Wilton.
980 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
981 with the definition in sysexits.h (which is #included earlier).
982 Fortunately, Exim does not actually use EX_OK. The code used to try to
983 preserve the sysexits.h value, by assumimg that macro definitions were
984 scanned for macro replacements. I have been disabused of this notion,
985 so now the code just undefines EX_OK before #including unistd.h.
987 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
988 in the smtp transport. When a block could not be written in a single
989 write() function, the timeout was being re-applied to each part-write.
990 This seems wrong - if the receiver was accepting one byte at a time it
991 would take for ever. The timeout is now adjusted when this happens. It
992 doesn't have to be particularly precise.
994 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
995 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
997 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
998 without a domain if the check to <postmaster@domain> fails.
1000 SC/01 Eximstats: added -xls and the ability to specify output files
1001 (patch written by Frank Heydlauf).
1003 SC/02 Eximstats: use FileHandles for outputing results.
1005 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1007 SC/04 Eximstats: fixed display of large numbers with -nvr option
1009 SC/05 Eximstats: fixed merging of reports with empty tables.
1011 SC/06 Eximstats: added the -include_original_destination flag
1013 SC/07 Eximstats: removed tabs and trailing whitespace.
1015 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1017 TK/06 MBOX spool code: Add real "From " MBOX separator line
1018 so the .eml file is really in mbox format (even though
1019 most programs do not really care). Patch from Alex Miller.
1021 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1022 The latter is generated from $received_to and is only set if the
1023 message has one envelope recipient. SA can use these headers,
1024 obviously out-of-the-box. Patch from Alex Miller.
1026 PH/08 The ${def test on a variable was returning false if the variable's
1027 value was "0", contrary to what the specification has always said!
1028 The result should be true unless the variable is empty.
1030 PH/09 The syntax error of a character other than { following "${if
1031 def:variable_name" (after optional whitespace) was not being diagnosed.
1032 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1033 accidental colon was present, for example, could give incorrect results.
1035 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1036 result is used (not including appendfile, where other changes are about
1039 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1040 This involved changing a lot of size variables from int to off_t. It
1041 should work with maildirs and everything.
1043 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1044 spamd dying while we are connected to it.
1046 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1049 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1050 function for BDB 4.3.
1052 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1054 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1055 the sources for PCRE have been re-arranged and more files are now
1058 PH/15 The code I had for printing potentially long long variables in PH/11
1059 above was not the best (it lost precision). The length of off_t variables
1060 is now inspected at build time, and an appropriate printing format (%ld
1061 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1062 to be "long long int" or "long int". This is needed for the internal
1063 formatting function string_vformat().
1065 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1066 the configuration file to be ":syslog", then the script "guesses" where
1067 the logs files are, rather than using the compiled in default. In our
1068 case the guess is not the same as the compiled default, so the script
1069 suddenly stopped working when I started to use syslog. The patch checks
1070 to see if log_file_path is "". If so, it attempts to read it from exim
1071 with no configuration file to get the compiled in version, before it
1072 falls back to the previous guessing code."
1074 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1075 implementing BATV in an Exim configuration. See NewStuff for the gory
1078 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1079 Makefile that are specific to HP-UX.
1081 PH/18 If the "use_postmaster" option was set for a recipient callout together
1082 with the "random" option, the postmaster address was used as the MAIL
1083 FROM address for the random test, but not for the subsequent recipient
1084 test. It is now used for both.
1086 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1087 patch removes a few documentation additions to RFC 3028, because the
1088 latest draft now contains them. It adds the new en;ascii-case comparator
1089 and a new error check for 8bit text in MIME parts. Comparator and
1090 require names are now matched exactly. I enabled the subaddress
1091 extension, but it is not well tested yet (read: it works for me)."
1093 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1094 rework some of the code of TK/09 above to avoid the hardwired use of
1095 "%lld" and "long long". Replaced the call to snprintf() with a call to
1098 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1099 records point to non-existent hosts", "retry timeout exceeded", and
1100 "retry time not reached for any host after a long failure period".
1102 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1103 experimental DomainKeys support:
1105 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1106 (2) On an error such as an illegally used "control", the wrong name for
1107 the control was given.
1109 These problems did NOT occur unless DomainKeys support was compiled.
1111 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1113 PH/24 Added ${if match_ip condition.
1115 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1116 databases so that it will be absolutely obvious if a crash occurs in the
1117 DB library. This is a regular occurrence (often caused by mis-matched
1120 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1121 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1122 picked up on a user's system that detects such things. There doesn't seem
1123 to be a gcc warning option for this - only an attribute that has to be
1124 put on the function's prototype. It seems that in Fedora Core 4 they have
1125 set this on a number of new functions. No doubt there will be more in due
1128 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1129 specify a transport. However, if an address that was verified by such a
1130 router was the subject of a callout, Exim crashed because it tried to
1131 read the rcpt_include_affixes from the non-existent transport. Now it
1132 just assumes that the setting of that option is false. This bug was
1133 introduced by 4.51/PH/31.
1135 PH/28 Changed -d+all to exclude +memory, because that information is very
1136 rarely of interest, but it makes the output a lot bigger. People tend to
1137 do -d+all out of habit.
1139 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1140 code in os-type was giving problems when libc.so lives in lib64, like on
1143 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1144 aren't the modern standard, and it seems that some systems' include files
1145 don't always have them. Exim was already checking for some of the newer
1146 ones like T_AAAA, and defining it itself. I've added checks for all the
1147 record types that Exim uses.
1149 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1150 not automatically generating a new one, as it is supposed to. This
1151 prevented TLS from working. If the file did exist, but contained invalid
1152 data, a new version was generated, as expected. It was only the case of a
1153 non-existent file that was broken.
1155 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1156 with a change in libdomainkeys > 0.64.
1158 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1159 from DNS. If the selector record carries the flag, it now has
1160 precedence over the domain-wide flag.
1162 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1164 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1165 the use of a port name that isn't defined in /etc/services) occurred, the
1166 message was deferred as in a normal delivery, and thus remained on the
1167 spool, instead of being failed because of the mua_wrapper setting. This
1168 is now fixed, and I tidied up some of the mua_wrapper messages at the
1171 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1172 the messages in a hash of arrays rather than using individual hashes.
1173 This is a bit cleaner and results in dramatic memory savings, albeit
1174 at a slight CPU cost.
1176 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1177 as requested by Marc Sherman.
1179 SC/10 Eximstats: added histograms for user specified patterns as requested
1182 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1184 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1185 fopen() in the content-scanning modules that did not already have it.
1191 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1192 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1194 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1196 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1198 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1199 to test Sieve filters that use "vacation".
1201 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1202 that changes the way the GnuTLS parameters are stored in the cache file.
1203 The new format can be generated externally. For backward compatibility,
1204 if the data in the cache doesn't make sense, Exim assumes it has read an
1205 old-format file, and it generates new data and writes a new file. This
1206 means that you can't go back to an older release without removing the
1209 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1210 work if there are any delivery delays because "one_time" forces the
1211 parent to be marked "delivered", so its unseen clone is never tried
1212 again. For this reason, Exim now forbids the simultaneous setting of
1215 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1216 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1217 entry if you want to know the details. The fix, however, neglected to
1218 consider the case when local delivery batching is involved. The test for
1219 "previously delivered" was not happening when checking to see if an
1220 address could be batched with a previous (undelivered) one; under
1221 certain circumstances this could lead to multiple deliveries to the same
1224 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1225 in its include files, and this causes problems building Exim.
1227 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1228 header_syntax) but Exim was just ignoring anything given after a slash.
1229 In particular, this caused confusion with an attempt to use "verify =
1230 reverse_host_lookup/defer_ok". An error is now given when options are
1231 supplied for verify items that do not have them. (Maybe reverse_host_
1232 lookup should have a defer_ok option, but that's a different point.)
1234 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1235 defined by RFC 821) to 2048, because there were problems with some AUTH
1236 commands, and RFC 1869 says the size should be increased for extended
1237 SMTP commands that take arguments.
1239 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1242 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1243 "unknown" error; now it says that the functionality isn't in the binary.
1245 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1246 an address' error message when a string expansion fails (syntax or
1247 whatever). Otherwise the password may appear in the log. Following change
1248 PH/42 below, there is no longer a chance of it appearing in a bounce
1251 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1253 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1254 of its addresses was ever tried. (Bugzilla bug #2.)
1256 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1257 the result incorrectly in the debug output. (It correctly added a newline
1258 to what was transported.)
1260 TF/01 Added $received_time.
1262 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1263 commented out examples of how to interface to a virus scanner and to
1264 SpamAssassin. Also added commented examples of av_scanner and
1265 spamd_address settings.
1267 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1268 and controls are allowed in which ACLs. There were a couple of minor
1269 errors. Some of the entries in the conditions table (which is a table of
1270 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1271 negation of where the condition IS allowed.
1273 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1275 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1276 header file does not have a version number, so I've had to invent a new
1277 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1278 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1279 radiusclient), but it was contributed by a Radius user.
1281 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1282 files or directories, for OpenSSL.
1284 PH/19 When an Exim process that is running as root has to create an Exim log
1285 file, it does so in a subprocess that runs as exim:exim so as to get the
1286 ownership right at creation (otherwise, other Exim processes might see
1287 the file with the wrong ownership). There was no test for failure of this
1288 fork() call, which would lead to the process getting stuck as it waited
1289 for a non-existent subprocess. Forks do occasionally fail when resources
1290 run out. I reviewed all the other calls to fork(); they all seem to check
1293 PH/20 When checking for unexpected SMTP input at connect time (before writing
1294 the banner), Exim was not dealing correctly with a non-positive return
1295 from the read() function. If the client had disconnected by this time,
1296 the result was a log entry for a synchronization error with an empty
1297 string after "input=" when read() returned zero. If read() returned -1
1298 (an event I could not check), uninitialized data bytes were printed.
1299 There were reports of junk text (parts of files, etc) appearing after
1302 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1304 PH/22 Added support for macro redefinition, and (re)definition in between
1305 driver and ACL definitions.
1307 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1308 forgetting to use the resulting value; it was using the unexpanded value.
1310 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1311 hadn't been configured. The fix is from Juergen Kreileder, who
1312 understands it better than I do:
1314 "Here's what I see happening with three configured cyrus_sasl
1315 authenticators configured (plain, login, cram-md5):
1317 On startup auth_cyrus_sasl_init() gets called for each of these.
1318 This means three calls to sasl_listmech() without a specified mech_list.
1319 => SASL tests which mechs of all available mechs actually work
1320 => three warnings about OTP not working
1321 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1323 With the patch, sasl_listmech() also gets called three times. But now
1324 SASL's mech_list option is set to the server_mech specified in the the
1325 authenticator. Or in other words, the answer from sasl_listmech()
1326 gets limited to just the mech you're testing for (which is different
1328 => the return list contains just 'plain' or 'login', 'cram-md5' or
1329 nothing depending on the value of ob->server_mech.
1331 I've just tested the patch: Authentication still works fine,
1332 unavailable mechs specified in the exim configuration are still
1333 caught, and the auth.log warnings about OTP are gone."
1335 PH/25 When debugging is enabled, the contents of the command line are added
1336 to the debugging output, even when log_selector=+arguments is not
1339 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1340 answer is "GNU", and only if the return is "GNU/something" is the answer
1343 PH/27 $acl_verify_message is now set immediately after the failure of a
1344 verification in an ACL, and so is available in subsequent modifiers. In
1345 particular, the message can be preserved by coding like this:
1347 warn !verify = sender
1348 set acl_m0 = $acl_verify_message
1350 Previously, $acl_verify_message was set only while expanding "message"
1351 and "log_message" when a very denied access.
1353 PH/28 Modified OS/os.c-Linux with
1355 -#ifndef OS_LOAD_AVERAGE
1356 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1358 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1359 nomenclature these days.)
1361 PH/29 Installed patch from the Sieve maintainer that adds the options
1362 sieve_useraddress and sieve_subaddress to the redirect router.
1364 PH/30 In these circumstances:
1365 . Two addresses routed to the same list of hosts;
1366 . First host does not offer TLS;
1367 . First host accepts first address;
1368 . First host gives temporary error to second address;
1369 . Second host offers TLS and a TLS session is established;
1370 . Second host accepts second address.
1371 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1372 and peerdn, if requested) that were in fact used only for the second
1375 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1376 attention to any local part prefix or suffix that was matched by the
1377 router that accepted the address. It now behaves in the same way as it
1378 does for delivery: the affixes are removed from the local part unless
1379 rcpt_include_affixes is set on the transport.
1381 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1382 timeout during the DATA phase of an incoming message.
1384 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1385 applied a patch sanctioned by the Sieve maintainer.
1387 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1388 the uid or gid is negative. A case of a negative gid caused this to be
1389 noticed. The fix allows for either to be negative.
1391 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1392 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1395 PH/36 The change PH/12 above was broken. Fixed it.
1397 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1398 the grounds that routing the same address twice would always produce the
1399 same answer. This might have been true once, but it is certainly no
1400 longer true now. Routing a child address may depend on the previous
1401 routing that produced that child. Some complicated redirection strategies
1402 went wrong when messages had multiple recipients, and made Exim's
1403 behaviour dependent on the order in which the addresses were given.
1405 I have moved the duplicate checking until after the routing is complete.
1406 Exim scans the addresses that are assigned to local and remote
1407 transports, and removes any duplicates. This means that more work will be
1408 done, as duplicates will always all be routed, but duplicates are
1409 presumably rare, so I don't expect this is of any significance.
1411 For deliveries to pipes, files, and autoreplies, the duplicate checking
1412 still happens during the routing process, since they are not going to be
1415 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1416 It corrects a timeout issue with spamd. This is Ian's comment: "The
1417 background is that sometimes spamd either never reads data from a
1418 connection it has accepted, or it never writes response data. The exiscan
1419 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1420 blindly assumes that writes won't block so it may never time out."
1422 PH/39 Allow G after quota size as well as K and M.
1424 PH/40 The value set for $authenticated_id in an authenticator may not contain
1425 binary zeroes or newlines because the value is written to log lines and
1426 to spool files. There was no check on this. Now the value is run through
1427 the string_printing() function so that such characters are converted to
1428 printable escape sequences.
1430 PH/41 $message_linecount is a new variable that contains the total number of
1431 lines in the message. Compare $body_linecount, which is the count for the
1434 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1435 bounce and delay warning messages, except in certain special cases, which
1438 (a) An SMTP error message from a remote host;
1439 (b) A message specified in a :fail: redirection;
1440 (c) A message specified in a "fail" command in a system filter;
1441 (d) A message specified in a FAIL return from the queryprogram router;
1442 (e) A message specified by the cannot_route_message router option.
1444 In these cases only, Exim does include the error details in bounce and
1445 warning messages. There are also a few cases where bland messages such
1446 as "unrouteable address" or "local delivery error" are given.
1448 PH/43 $value is now also set for the "else" part of a ${run expansion.
1450 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1451 being worked on, but at least Exim now implements the latest version to
1454 PH/45 In a pipe transport, although a timeout while waiting for the pipe
1455 process to complete was treated as a delivery failure, a timeout while
1456 writing the message to the pipe was logged, but erroneously treated as a
1457 successful delivery. Such timeouts include transport filter timeouts. For
1458 consistency with the overall process timeout, these timeouts are now
1459 treated as errors, giving rise to delivery failures by default. However,
1460 there is now a new Boolean option for the pipe transport called
1461 timeout_defer, which, if set TRUE, converts the failures into defers for
1462 both kinds of timeout. A transport filter timeout is now identified in
1465 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1466 systems where "make" and "gmake" are different, calling "gmake" at top
1467 level broke things. I've arranged for the value of $(MAKE) to be passed
1468 from the Makefile to this script so that it can call the same version of
1472 A note about Exim versions 4.44 and 4.50
1473 ----------------------------------------
1475 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1476 changes of various kinds. As a consequence, a big documentation update was
1477 needed. This delayed the release for rather longer than seemed good, especially
1478 in the light of a couple of (minor) security issues. Therefore, the changes
1479 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1480 release. So 4.44 and 4.50 are in effect two different branches that both start
1483 I have left the 4.50 change log unchanged; it contains all the changes since
1484 4.43. The change log for 4.44 is below; many of its items are identical to
1485 those for 4.50. This seems to be the most sensible way to preserve the
1486 historical information.
1492 1. Minor wording change to the doc/README.SIEVE file.
1494 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1495 computation of the current number of files was incorrect.
1497 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1498 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1501 4. Give more explanation in the error message when the command for a transport
1502 filter fails to execute.
1504 5. There are several places where Exim runs a non-Exim command in a
1505 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1506 was being done only for the command run by the queryprogram router. It is
1507 now done for all such subprocesses. The other cases are: ${run, transport
1508 filters, and the commands run by the lmtp and pipe transports.
1510 6. Added CONFIGURE_GROUP build-time option.
1512 7. Some older OS have a limit of 256 on the maximum number of file
1513 descriptors. Exim was using setrlimit() to set 1000 as a large value
1514 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1515 systems. I've change it so that if it can't get 1000, it tries for 256.
1517 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1518 was an oversight, and furthermore, ever since the addition of extra
1519 controls (e.g. 4.43/32), the checks on when to allow different forms of
1520 "control" were broken. There should now be diagnostics for all cases when a
1521 control that does not make sense is encountered.
1523 9. Added the /retain_sender option to "control=submission".
1525 10. $recipients is now available in the predata ACL (oversight).
1527 11. Tidy the search cache before the fork to do a delivery from a message
1528 received from the command line. Otherwise the child will trigger a lookup
1529 failure and thereby defer the delivery if it tries to use (for example) a
1530 cached ldap connection that the parent has called unbind on.
1532 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1533 of $address_data from the recipient verification was clobbered by the
1534 sender verification.
1536 13. The value of address_data from a sender verification is now available in
1537 $sender_address_data in subsequent conditions in the ACL statement.
1539 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1541 15. Added a new option "connect=<time>" to callout options, to set a different
1544 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1545 was its contents. (It was OK if the option was not defined at all.)
1547 17. A "Completed" log line is now written for messages that are removed from
1548 the spool by the -Mrm option.
1550 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1551 information about exactly what failed.
1553 19. Added -dd to debug only the daemon process.
1555 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1556 handles timeouts, both on the server side and network timeouts. Renamed the
1557 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1559 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1560 It is now set to "smtps".
1562 22. $host_address is now set to the target address during the checking of
1563 ignore_target_hosts.
1565 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1566 being passed; this would have caused $sender_host_name to have been used if
1567 matching the list had actually called for a host name (not very likely,
1568 since this list is usually IP addresses). A host name is now passed as
1571 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1572 code that specifies a non-restarting handler (typically sigaction() in
1573 modern systems) in an attempt to fix a rare and obscure crash bug.
1575 25. Narrowed the window for a race in the daemon that could cause it to ignore
1576 SIGCHLD signals. This is not a major problem, because they are used only to
1577 wake it up if nothing else does.
1579 26. A malformed maildirsize file could cause Exim to calculate negative values
1580 for the mailbox size or file count. Odd effects could occur as a result.
1581 The maildirsize information is now recalculated if the size or filecount
1584 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1585 support for a long time. Removed HAVE_SYS_VFS_H.
1587 28. Installed the latest version of exipick from John Jetmore.
1589 29. In an address list, if the pattern was not a regular expression, an empty
1590 subject address (from a bounce message) matched only if the pattern was an
1591 empty string. Non-empty patterns were not even tested. This was the wrong
1592 because it is perfectly reasonable to use an empty address as part of a
1593 database query. An empty address is now tested by patterns that are
1594 lookups. However, all the other forms of pattern expect the subject to
1595 contain a local part and a domain, and therefore, for them, an empty
1596 address still always fails if the pattern is not itself empty.
1598 30. Exim went into a mad DNS loop when attempting to do a callout where the
1599 host was specified on an smtp transport, and looking it up yielded more
1600 than one IP address.
1602 31. Re-factored the code for checking spool and log partition space into a
1603 function that finds that data and another that does the check. The former
1604 is then used to implement four new variables: $spool_space, $log_space,
1605 $spool_inodes, and $log_inodes.
1607 32. The RFC2047 encoding function was originally intended for short strings
1608 such as real names; it was not keeping to the 75-character limit for
1609 encoded words that the RFC imposes. It now respects the limit, and
1610 generates multiple encoded words if necessary. To be on the safe side, I
1611 have increased the buffer size for the ${rfc2047: expansion operator from
1614 33. It is now permitted to omit both strings after an "if" condition; if the
1615 condition is true, the result is "true". As before, when the second string
1616 is omitted, a false condition yields an empty string. This makes it less
1617 cumbersome to write custom ACL and router conditions.
1619 34. Failure to deliver a bounce message always caused it to be frozen, even if
1620 there was an errors_to setting on the router. The errors_to setting is now
1623 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1624 canonical form (fully expanded) before being placed in
1625 $sender_host_address.
1627 36. The table in the code that translates DNS record types into text (T_A to
1628 "A" for instance) was missing entries for NS and CNAME. It is just possible
1629 that this could have caused confusion if both these types were looked up
1630 for the same domain, because the text type is used as part of Exim's
1631 per-process caching. But the chance of anyone hitting this buglet seems
1634 37. The dnsdb lookup has been extended in a number of ways.
1636 (1) There is a new type, "zns", which walks up the domain tree until it
1637 finds some nameserver records. It should be used with care.
1639 (2) There is a new type, "mxh", which is like "mx" except that it returns
1640 just the host names, not the priorities.
1642 (3) It is now possible to give a list of domains (or IP addresses) to be
1643 looked up. The behaviour when one of the lookups defers can be
1644 controlled by a keyword.
1646 (4) It is now possible to specify the separator character for use when
1647 multiple records are returned.
1649 38. The dnslists ACL condition has been extended: it is now possible to supply
1650 a list of IP addresses and/or domains to be looked up in a particular DNS
1653 39. Added log_selector=+queue_time_overall.
1655 40. When running the queue in the test harness, wait just a tad after forking a
1656 delivery process, to get repeatability of debugging output.
1658 41. Include certificate and key file names in error message when GnuTLS fails
1659 to set them up, because the GnuTLS error message doesn't include the name
1660 of the failing file when there is a problem reading it.
1662 42. Allow both -bf and -bF in the same test run.
1664 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1666 44. The "Exiscan patch" is now merged into the mainline Exim source.
1668 45. Sometimes the final signoff response after QUIT could fail to get
1669 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1670 < 0 before doing a fflush(). This bug looks as though it goes back to the
1671 introduction of TLS in release 3.20, but "sometimes" must have been rare
1672 because the tests only now provoked it.
1674 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1675 (this can affect the format of dates).
1677 47. exim_tidydb, when checking for the continued existence of a message for
1678 which it has found a message-specific retry record, was not finding
1679 messages that were in split spool directories. Consequently, it was
1680 deleting retry records that should have stayed in existence.
1682 48. Steve fixed some bugs in eximstats.
1684 49. The SPA authentication driver was not abandoning authentication and moving
1685 on to the next authenticator when an expansion was forced to fail,
1686 contradicting the general specification for all authenticators. Instead it
1687 was generating a temporary error. It now behaves as specified.
1689 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1690 (the order specifies the preference for clients). The order is now AES256,
1691 AES128, 3DES, ARCFOUR128.
1693 51. Small patch to Sieve code - explicitly set From: when generating an
1696 52. Exim crashed if a remote delivery caused a very long error message to be
1697 recorded - for instance if somebody sent an entire SpamAssassin report back
1698 as a large number of 550 error lines. This bug was coincidentally fixed by
1699 increasing the size of one of Exim's internal buffers (big_buffer) that
1700 happened as part of the Exiscan merge. However, to be on the safe side, I
1701 have made the code more robust (and fixed the comments that describe what
1704 53. Now that there can be additional text after "Completed" in log lines (if
1705 the queue_time_overall log selector is set), a one-byte patch to exigrep
1706 was needed to allow it to recognize "Completed" as not the last thing in
1709 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1710 patch that reportedly fixes this has been added. I am not expert enough to
1711 create a test for it. This is what the patch creator wrote:
1713 "I found a little strange behaviour of ldap code when working with
1714 Windows 2003 AD Domain, where users was placed in more than one
1715 Organization Units. When I tried to give exim partial DN, the exit code
1716 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1717 But simultaneously result of request was absolutely normal ldap result,
1718 so I produce this patch..."
1720 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1721 REFERENCE, so I have modified the code to exclude the patch when that macro
1724 55. Some experimental protocols are using DNS PTR records for new purposes. The
1725 keys for these records are domain names, not reversed IP addresses. The
1726 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1727 leaves it alone. Component reversal etc. now happens only for IP addresses.
1730 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1732 57. Double the size of the debug message buffer (to 2048) so that more of very
1733 long debug lines gets shown.
1735 58. The exicyclog utility now does better if the number of log files to keep
1736 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1738 59. Two changes related to the smtp_active_hostname option:
1740 (1) $smtp_active_hostname is now available as a variable.
1741 (2) The default for smtp_banner uses $smtp_active_hostname instead
1742 of $primary_hostname.
1744 60. The host_aton() function is supposed to be passed a string that is known
1745 to be a valid IP address. However, in the case of IPv6 addresses, it was
1746 not checking this. This is a hostage to fortune. Exim now panics and dies
1747 if the condition is not met. A case was found where this could be provoked
1748 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1749 components; fortuitously, this particular loophole had already been fixed
1750 by change 4.50/55 above.
1752 If there are any other similar loopholes, the new check in host_aton()
1753 itself should stop them being exploited. The report I received stated that
1754 data on the command line could provoke the exploit when Exim was running as
1755 exim, but did not say which command line option was involved. All I could
1756 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1757 running as the user.
1760 61. There was a buffer overflow vulnerability in the SPA authentication code
1761 (which came originally from the Samba project). I have added a test to the
1762 spa_base64_to_bits() function which I hope fixes it.
1765 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1766 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1768 63. The daemon start-up calls getloadavg() while still root for those OS that
1769 need the first call to be done as root, but it missed one case: when
1770 deliver_queue_load_max is set with deliver_drop_privilege. This is
1771 necessary for the benefit of the queue runner, because there is no re-exec
1772 when deliver_drop_privilege is set.
1774 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1775 This has been fixed.
1777 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1778 was in use, was not putting the data itself into the right store pool;
1779 consequently, it could be overwritten for a subsequent message in the same
1780 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1783 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1785 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1786 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1787 the function were treating the return as a boolean value, which happened to
1788 work because 0=false and not-0=true, but is not correct code.
1790 68. The host_aton() function was not handling scoped IPv6 addresses (those
1791 with, for example, "%eth0" on the end) correctly.
1793 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1794 negated items (that is, ~something) in unsigned ints. Some compilers
1795 apparently mutter when there is no cast.
1797 70. If an address verification called from an ACL failed, and did not produce a
1798 user-specific message (i.e. there was only a "system" message), nothing was
1799 put in $acl_verify_message. In this situation, it now puts the system
1802 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1803 session; change 4.31/43 added the unwanted input to the log line - except
1804 that it did not do this in the start of session case. It now does.
1806 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1807 This is wrong and can cause the other end to generate a synchronization
1808 error if it is another Exim or anything else that does the synchronization
1809 check. A QUIT command is no longer sent after a timeout.
1811 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1812 during host lookups.
1814 74. The defer_ok option of callout verification was not working if it was used
1815 when verifying addresses in header lines, that is, for this case:
1817 verify = header_sender/callout=defer_ok
1819 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1820 those file descriptors could be used for SMTP connections. If anything
1821 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1822 could be sent to the SMTP client, causing chaos. The daemon now opens
1823 stdin, stdout, and stderr to /dev/null when it puts itself into the
1826 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1827 log by default. The user can override this with suitable Perl magic.
1829 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1830 the log message when discard triggers, was not working for the DATA ACL or
1831 for the non-SMTP ACL.
1833 78. Error message wording change in sieve.c.
1835 79. If smtp_accept_max_per_host was set, the number of connections could be
1836 restricted to fewer than expected, because the daemon was trying to set up
1837 a new connection before checking whether the processes handling previous
1838 connections had finished. The check for completed processes is now done
1839 earlier. On busy systems, this bug wouldn't be noticed because something
1840 else would have woken the daemon, and it would have reaped the completed
1843 80. If a message was submitted locally by a user whose login name contained one
1844 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1845 It caused a spool format error. I have fixed the spool reading code. A
1846 related problem was that the "from" clause in the Received: line became
1847 illegal because of the space(s). It is now covered by ${quote_local_part.
1849 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
1852 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
1854 83. Patch from Sieve maintainer to fix unterminated string problem in
1855 "vacation" handling.
1857 84. Some minor changes to the Linux configuration files to help with other
1858 OS variants using glibc.
1860 85. One more patch for Sieve to update vacation handling to latest spec.
1863 ----------------------------------------------------
1864 See the note above about the 4.44 and 4.50 releases.
1865 ----------------------------------------------------
1871 1. Change 4.43/35 introduced a bug that caused file counts to be
1872 incorrectly computed when quota_filecount was set in an appendfile
1875 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1876 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1879 3. Give more explanation in the error message when the command for a transport
1880 filter fails to execute.
1882 4. There are several places where Exim runs a non-Exim command in a
1883 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1884 was being done only for the command run by the queryprogram router. It is
1885 now done for all such subprocesses. The other cases are: ${run, transport
1886 filters, and the commands run by the lmtp and pipe transports.
1888 5. Some older OS have a limit of 256 on the maximum number of file
1889 descriptors. Exim was using setrlimit() to set 1000 as a large value
1890 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1891 systems. I've change it so that if it can't get 1000, it tries for 256.
1893 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1894 was an oversight, and furthermore, ever since the addition of extra
1895 controls (e.g. 4.43/32), the checks on when to allow different forms of
1896 "control" were broken. There should now be diagnostics for all cases when a
1897 control that does not make sense is encountered.
1899 7. $recipients is now available in the predata ACL (oversight).
1901 8. Tidy the search cache before the fork to do a delivery from a message
1902 received from the command line. Otherwise the child will trigger a lookup
1903 failure and thereby defer the delivery if it tries to use (for example) a
1904 cached ldap connection that the parent has called unbind on.
1906 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1907 of $address_data from the recipient verification was clobbered by the
1908 sender verification.
1910 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1911 was its contents. (It was OK if the option was not defined at all.)
1913 11. A "Completed" log line is now written for messages that are removed from
1914 the spool by the -Mrm option.
1916 12. $host_address is now set to the target address during the checking of
1917 ignore_target_hosts.
1919 13. When checking ignore_target_hosts for an ipliteral router, no host name was
1920 being passed; this would have caused $sender_host_name to have been used if
1921 matching the list had actually called for a host name (not very likely,
1922 since this list is usually IP addresses). A host name is now passed as
1925 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1926 code that specifies a non-restarting handler (typically sigaction() in
1927 modern systems) in an attempt to fix a rare and obscure crash bug.
1929 15. Narrowed the window for a race in the daemon that could cause it to ignore
1930 SIGCHLD signals. This is not a major problem, because they are used only to
1931 wake it up if nothing else does.
1933 16. A malformed maildirsize file could cause Exim to calculate negative values
1934 for the mailbox size or file count. Odd effects could occur as a result.
1935 The maildirsize information is now recalculated if the size or filecount
1938 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1939 support for a long time. Removed HAVE_SYS_VFS_H.
1941 18. Updated exipick to current release from John Jetmore.
1943 19. Allow an empty sender to be matched against a lookup in an address list.
1944 Previously the only cases considered were a regular expression, or an
1947 20. Exim went into a mad DNS lookup loop when doing a callout where the
1948 host was specified on the transport, if the DNS lookup yielded more than
1951 21. The RFC2047 encoding function was originally intended for short strings
1952 such as real names; it was not keeping to the 75-character limit for
1953 encoded words that the RFC imposes. It now respects the limit, and
1954 generates multiple encoded words if necessary. To be on the safe side, I
1955 have increased the buffer size for the ${rfc2047: expansion operator from
1958 22. Failure to deliver a bounce message always caused it to be frozen, even if
1959 there was an errors_to setting on the router. The errors_to setting is now
1962 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1963 canonical form (fully expanded) before being placed in
1964 $sender_host_address.
1966 24. Updated eximstats to version 1.33
1968 25. Include certificate and key file names in error message when GnuTLS fails
1969 to set them up, because the GnuTLS error message doesn't include the name
1970 of the failing file when there is a problem reading it.
1972 26. Expand error message when OpenSSL has problems setting up cert/key files.
1975 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
1976 (this can affect the format of dates).
1978 28. exim_tidydb, when checking for the continued existence of a message for
1979 which it has found a message-specific retry record, was not finding
1980 messages that were in split spool directories. Consequently, it was
1981 deleting retry records that should have stayed in existence.
1983 29. eximstats updated to version 1.35
1984 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1985 1.35 - bugfix such that pie charts by volume are generated correctly
1987 30. The SPA authentication driver was not abandoning authentication and moving
1988 on to the next authenticator when an expansion was forced to fail,
1989 contradicting the general specification for all authenticators. Instead it
1990 was generating a temporary error. It now behaves as specified.
1992 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
1993 (the order specifies the preference for clients). The order is now AES256,
1994 AES128, 3DES, ARCFOUR128.
1996 31. Small patch to Sieve code - explicitly set From: when generating an
1999 32. Exim crashed if a remote delivery caused a very long error message to be
2000 recorded - for instance if somebody sent an entire SpamAssassin report back
2001 as a large number of 550 error lines. This bug was coincidentally fixed by
2002 increasing the size of one of Exim's internal buffers (big_buffer) that
2003 happened as part of the Exiscan merge. However, to be on the safe side, I
2004 have made the code more robust (and fixed the comments that describe what
2007 33. Some experimental protocols are using DNS PTR records for new purposes. The
2008 keys for these records are domain names, not reversed IP addresses. The
2009 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2010 leaves it alone. Component reversal etc. now happens only for IP addresses.
2013 34. The host_aton() function is supposed to be passed a string that is known
2014 to be a valid IP address. However, in the case of IPv6 addresses, it was
2015 not checking this. This is a hostage to fortune. Exim now panics and dies
2016 if the condition is not met. A case was found where this could be provoked
2017 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2018 components; fortuitously, this particular loophole had already been fixed
2019 by change 4.50/55 or 4.44/33 above.
2021 If there are any other similar loopholes, the new check in host_aton()
2022 itself should stop them being exploited. The report I received stated that
2023 data on the command line could provoke the exploit when Exim was running as
2024 exim, but did not say which command line option was involved. All I could
2025 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2026 running as the user.
2029 35. There was a buffer overflow vulnerability in the SPA authentication code
2030 (which came originally from the Samba project). I have added a test to the
2031 spa_base64_to_bits() function which I hope fixes it.
2034 36. The daemon start-up calls getloadavg() while still root for those OS that
2035 need the first call to be done as root, but it missed one case: when
2036 deliver_queue_load_max is set with deliver_drop_privilege. This is
2037 necessary for the benefit of the queue runner, because there is no re-exec
2038 when deliver_drop_privilege is set.
2040 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2041 was in use, was not putting the data itself into the right store pool;
2042 consequently, it could be overwritten for a subsequent message in the same
2043 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2046 38. Sometimes the final signoff response after QUIT could fail to get
2047 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2048 < 0 before doing a fflush(). This bug looks as though it goes back to the
2049 introduction of TLS in release 3.20, but "sometimes" must have been rare
2050 because the tests only now provoked it.
2056 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2057 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2058 Now it returns an integer. A number of calls were still expecting a T/F
2059 return. Fortuitously, in all cases, the tests worked in OK situations,
2060 which is the norm. However, things would have gone wrong on any write
2061 failures on the smtp file descriptor. This function is used when sending
2062 messages over SMTP and also when doing verify callouts.
2064 2. When Exim is called to do synchronous delivery of a locally submitted
2065 message (the -odf or -odi options), it no longer closes stderr before doing
2068 3. Implemented the mua_wrapper option.
2070 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2072 5. Implemented the functions header_remove(), header_testname(),
2073 header_add_at_position(), and receive_remove_recipient(), and exported them
2076 6. If an ACL "warn" statement specified the addition of headers, Exim already
2077 inserted X-ACL-Warn: at the start if there was no header name. However, it
2078 was not making this test for the second and subsequent header lines if
2079 there were newlines in the string. This meant that an invalid header could
2080 be inserted if Exim was badly configured.
2082 7. Allow an ACL "warn" statement to add header lines at the start or after all
2083 the Received: headers, as well as at the end.
2085 8. Added the rcpt_4xx retry error code.
2087 9. Added postmaster_mailfrom=xxx to callout verification option.
2089 10. Added mailfrom=xxxx to the callout verification option, for verify=
2092 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2093 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2095 12. Inserted some casts to stop certain compilers warning when using pointer
2096 differences as field lengths or precisions in printf-type calls (mostly
2097 affecting debugging statements).
2099 13. Added optional readline() support for -be (dynamically loaded).
2101 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2102 same clock tick as a message's arrival, so that its received time was the
2103 same as the "first fail" time on the retry record, and that message
2104 remained on the queue past the ultimate address timeout, every queue runner
2105 would try a delivery (because it was past the ultimate address timeout) but
2106 after another failure, the ultimate address timeout, which should have then
2107 bounced the address, did not kick in. This was a "< instead of <=" error;
2108 in most cases the first failure would have been in the next clock tick
2109 after the received time, and all would be well.
2111 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2112 being recognized when the domain list was tested by the match_domain
2113 condition in an expansion string.
2115 16. Added the ${str2b64: operator.
2117 17. Exim was always calling setrlimit() to set a large limit for the number of
2118 processes, without checking whether the existing limit was already
2119 adequate. (It did check for the limit on file descriptors.) Furthermore,
2120 errors from getrlimit() and setrlimit() were being ignored. Now they are
2121 logged to the main and panic logs, but Exim does carry on, to try to do its
2122 job under whatever limits there are.
2124 18. Imported PCRE 5.0.
2126 19. Trivial typo in log message " temporarily refused connection" (the leading
2129 20. If the log selector return_path_on_delivery was set and an address was
2130 redirected to /dev/null, the delivery process crashed because it assumed
2131 that a return path would always be set for a "successful" delivery. In this
2132 case, the whole delivery is bypassed as an optimization, and therefore no
2135 21. Internal re-arrangement: the function for sending a challenge and reading
2136 a response while authentication was assuming a zero-terminated challenge
2137 string. It's now changed to take a pointer and a length, to allow for
2138 binary data in such strings.
2140 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2142 23. Exim was not respecting finduser_retries when seeking the login of the
2143 uid under which it was called; it was always trying 10 times. (The default
2144 setting of finduser_retries is zero.) Also, it was sleeping after the final
2145 failure, which is pointless.
2147 24. Implemented tls_on_connect_ports.
2149 25. Implemented acl_smtp_predata.
2151 26. If the domain in control=submission is set empty, Exim assumes that the
2152 authenticated id is a complete email address when it generates From: or
2153 Sender: header lines.
2155 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2156 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2157 chown and chgrp in /bin and hostname in /usr/bin.
2159 28. Exim was keeping the "process log" file open after each use, just as it
2160 does for the main log. This opens the possibility of it remaining open for
2161 long periods when the USR1 signal hits a daemon. Occasional processlog
2162 errors were reported, that could have been caused by this. Anyway, it seems
2163 much more sensible not to leave this file open at all, so that is what now
2166 29. The long-running daemon process does not normally write to the log once it
2167 has entered its main loop, and it closes the log before doing so. This is
2168 so that log files can straightforwardly be renamed and moved. However,
2169 there are a couple of unusual error situations where the daemon does write
2170 log entries, and I had neglected to close the log afterwards.
2172 30. The text of an SMTP error response that was received during a remote
2173 delivery was being truncated at 512 bytes. This is too short for some of
2174 the long messages that one sometimes sees. I've increased the limit to
2177 31. It is now possible to make retry rules that apply only when a message has a
2178 specific sender, in particular, an empty sender.
2180 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2181 it possible to be selective about when SMTP synchronization is enforced.
2183 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2185 32. Implemented hosts_connection_nolog.
2187 33. Added an ACL for QUIT.
2189 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2192 35. Added mailbox_size and mailbox_filecount to appendfile.
2194 36. Added control = no_multiline_responses to ACLs.
2196 37. There was a bug in the logic of the code that waits for the clock to tick
2197 in the case where the clock went backwards by a substantial amount such
2198 that the microsecond fraction of "now" was more than the microsecond
2199 fraction of "then" (but the whole seconds number was less).
2201 38. Added support for the libradius Radius client library this is found on
2202 FreeBSD (previously only the radiusclient library was supported).
2208 1. When certain lookups returned multiple values in the form name=value, the
2209 quoting of the values was not always being done properly. Specifically:
2210 (a) If the value started with a double quote, but contained no whitespace,
2212 (b) If the value contained whitespace other than a space character (i.e.
2213 tabs or newlines or carriage returns) it was not quoted.
2214 This fix has been applied to the mysql and pgsql lookups by writing a
2215 separate quoting function and calling it from the lookup code. The fix
2216 should probably also be applied to nisplus, ibase and oracle lookups, but
2217 since I cannot test any of those, I have not disturbed their existing code.
2219 2. A hit in the callout cache for a specific address caused a log line with no
2220 reason for rejecting RCPT. Now it says "Previous (cached) callout
2221 verification failure".
2223 3. There was an off-by-one bug in the queryprogram router. An over-long
2224 return line was truncated at 256 instead of 255 characters, thereby
2225 overflowing its buffer with the terminating zero. As well as fixing this, I
2226 have increased the buffer size to 1024 (and made a note to document this).
2228 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2229 when Exim is waiting for an SMTP response from a remote server, Exim
2230 restarts its select() call on the socket, thereby resetting its timeout.
2231 This is not a problem when such interrupts are rare. Somebody set up a cron
2232 job to run exiwhat every 2 minutes, which is less than the normal select()
2233 timeout (5 or 10 minutes). This meant that the select() timeout never
2234 kicked in because it was always reset. I have fixed this by comparing the
2235 time when an interrupt arrives with the time at the start of the first call
2236 to select(). If more time than the timeout has elapsed, the interrupt is
2237 treated as a timeout.
2239 5. Some internal re-factoring in preparation for the addition of Sieve
2240 extensions (by MH). In particular, the "personal" test is moved to a
2241 separate function, and given an option for scanning Cc: and Bcc: (which is
2242 not set for Exim filters).
2244 6. When Exim created an email address using the login of the caller as the
2245 local part (e.g. when creating a From: or Sender: header line), it was not
2246 quoting the local part when it contained special characters such as @.
2248 7. Installed new OpenBSD configuration files.
2250 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2251 try to make them clearer.
2253 9. Callout options, other than the timeout value, were being ignored when
2254 verifying sender addresses in header lines. For example, when using
2256 verify = header_sender/callout=no_cache
2258 the cache was (incorrectly) being used.
2260 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2261 only the Cygwin environment.
2263 11. When return_path_on_delivery was set as a log selector, if different remote
2264 addresses in the same message used different return paths and parallel
2265 remote delivery occurred, the wrong values would sometimes be logged.
2266 (Whenever a remote delivery process finished, the return path value from
2267 the most recently started remote delivery process was logged.)
2269 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2270 lines when AUTH and/or TLS are in use. This is the "received protocol"
2271 field. Exim used to use "asmtp" for authenticated SMTP, without any
2272 indication (in the protocol name) for TLS use. Now it follows the RFC and
2273 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2274 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2275 names appear in log lines as well as in Received: header lines.
2277 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2278 extensions, and comparison tests, and to fix some bugs.
2280 14. Changes to the "personal" filter test:
2282 (1) The test was buggy in that it was just doing the equivalent of
2283 "contains" tests on header lines. For example, if a user's address was
2284 anne@some.where, the "personal" test would incorrectly be true for
2286 To: susanne@some.where
2288 This test is now done by extracting each address from the header in turn,
2289 and checking the entire address. Other tests that are part of "personal"
2290 are now done using regular expressions (for example, to check local parts
2291 of addresses in From: header lines).
2293 (2) The list of non-personal local parts in From: addresses has been
2294 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2295 taken from the Sieve specification recommendations.
2297 (3) If the message contains any header line starting with "List-" it is
2298 treated as non-personal.
2300 (4) The test for "circular" in the Subject: header line has been removed
2301 because it now seems ill-conceived.
2303 15. Minor typos in src/EDITME comments corrected.
2305 16. Installed latest exipick from John Jetmore.
2307 17. If headers_add on a router specified a text string that was too long for
2308 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2309 of string_sprintf() is now avoided.
2311 18. $message_body_size was not set (it was always zero) when running the DATA
2312 ACL and the local_scan() function.
2314 19. For the "mail" command in an Exim filter, no default was being set for
2315 the once_repeat time, causing a random time value to be used if "once" was
2316 specified. (If the value happened to be <= 0, no repeat happened.) The
2317 default is now 0s, meaning "never repeat". The "vacation" command was OK
2318 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2319 (I found it when inspecting the code).
2321 20. There is now an overall timeout for performing a callout verification. It
2322 defaults to 4 times the callout timeout, which applies to individual SMTP
2323 commands during the callout. The overall timeout applies when there is more
2324 than one host that can be tried. The timeout is checked before trying the
2325 next host. This prevents very long delays if there are a large number of
2326 hosts and all are timing out (e.g. when the network connections are timing
2327 out). The value of the overall timeout can be changed by specifying an
2328 additional sub-option for "callout", called "maxwait". For example:
2330 verify = sender/callout=5s,maxwait=20s
2332 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2333 to the end before writing, but this should make it even safer).
2335 22. Exim was forgetting that it had advertised PIPELINING for the second and
2336 subsequent messages on an SMTP connection. It was also not resetting its
2337 memory on STARTTLS and an internal HELO.
2339 23. When Exim logs an SMTP synchronization error within a session, it now
2340 records whether PIPELINING has been advertised or not.
2342 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2343 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2344 rather than long int.
2346 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2348 26. Added the never_mail option to autoreply.
2354 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2355 crash if the getsockname() call failed; this can happen if a connection is
2356 closed very soon after it is established. The problem was simply in the
2357 order in which certain operations were done, causing Exim to try to write
2358 to the SMTP stream before it had set up the file descriptor. The bug has
2359 been fixed by making things happen in the correct order.
2365 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2366 before the connection was closed, thus losing the rejection response.
2368 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2369 some early Solaris releases, but causes trouble in current releases where
2370 socklen_t is defined.
2372 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2375 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2378 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2379 that releasing the top of it at the end releases what was used for sub-
2380 expansions (unless the block got too big). However, discard this block if
2381 the first thing is a variable or header, so that we can use its block when
2382 it is dynamic (useful for very large $message_headers, for example).
2384 6. Lookups now cache *every* query, not just the most recent. A new, separate
2385 store pool is used for this. It can be recovered when all lookup caches are
2386 flushed. Lookups now release memory at the end of their result strings.
2387 This has involved some general refactoring of the lookup sources.
2389 7. Some code has been added to the store_xxx() functions to reduce the amount
2390 of flapping under certain conditions.
2392 8. log_incoming_interface used to affect only the <= reception log lines. Now
2393 it causes the local interface and port to be added to several more SMTP log
2394 lines, for example "SMTP connection from", and rejection lines.
2396 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2398 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2400 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2401 could be overwritten at the end of the current message (or the start of a
2402 new message if it was set in a HELO ACL). The value is now preserved for
2403 the duration of the SMTP connection.
2405 12. If a transport had a headers_rewrite setting, and a matching header line
2406 contained an unqualified address, that address was qualified, even if it
2407 did not match any rewriting rules. The underlying bug was that the values
2408 of the flags that permit the existence of unqualified sender and recipient
2409 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2410 non-local messages, and by -bnq for local messages) were not being
2411 preserved with the message after it was received.
2413 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2414 "next input=" as part of the text comprising the host identity instead of
2415 the correct text. The code was using the same buffer for two different
2416 strings. However, depending on which order the printing function evaluated
2417 its arguments, the bug did not always show up. Under Linux, for example, my
2418 test suite worked just fine.
2420 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2421 doesn't work with some older versions of Perl. It has been changed to "my",
2422 which in any case is probably the better facility to use.
2424 15. A really picky compiler found some instances of statements for creating
2425 error messages that either had too many or two few arguments for the format
2428 16. The size of the buffer for calls to the DNS resolver has been increased
2429 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2430 for addresses that have a lot of PTR records. This alleviates a problem; it
2431 does not fully solve it.
2433 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2434 buffer now truncates the list and logs the incident, which is the same
2435 action as happens when Exim is looking up a host name and its aliases.
2436 Previously in this situation something unpredictable would happen;
2437 sometimes it was "internal error: store_reset failed".
2439 18. If a server dropped the connection unexpectedly when an Exim client was
2440 using GnuTLS and trying to read a response, the client delivery process
2441 crashed while trying to generate an error log message.
2443 19. If a "warn" verb in an ACL added multiple headers to a message in a single
2444 string, for example:
2446 warn message = H1: something\nH2: something
2448 the text was added as a single header line from Exim's point of view
2449 though it ended up OK in the delivered message. However, searching for the
2450 second and subsequent header lines using $h_h2: did not work. This has been
2451 fixed. Similarly, if a system filter added multiple headers in this way,
2452 the routers could not see them.
2454 20. Expanded the error message when iplsearch is called with an invalid key to
2455 suggest using net-iplsearch in a host list.
2457 21. When running tests using -bh, any delays imposed by "delay" modifiers in
2458 ACLs are no longer actually imposed (and a message to that effect is
2461 22. If a "gecos" field in a passwd entry contained escaped characters, in
2462 particular, if it contained a \" sequence, Exim got it wrong when building
2463 a From: or a Sender: header from that name. A second bug also caused
2464 incorrect handling when an unquoted " was present following a character
2465 that needed quoting.
2467 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2468 was not being matched caselessly.
2470 24. Arranged for all hyphens in the exim.8 source to be escaped with
2473 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2474 properly. Recipient callout cache records were still being keyed to include
2475 the sender, even when use_sender was set false. This led to far more
2476 callouts that were necessary. The sender is no longer included in the key
2477 when use_sender is false.
2479 26. Added "control = submission" modifier to ACLs.
2481 27. Added the ${base62d: operator to decode base 62 numbers.
2483 28. dnsdb lookups can now access SRV records.
2485 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2486 the configuration file.
2488 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2489 (-v) mode. This makes the output for a verbose queue run more intelligible.
2491 31. Added a use_postmaster feature to recipient callouts.
2493 32. Added the $body_zerocount variable, containing the number of binary zero
2494 bytes in the message body.
2496 33. The time of last modification of the "new" subdirectory is now used as the
2497 "mailbox time last read" when there is a quota error for a maildir
2500 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2502 35. Added +ignore_unknown as a special item in host lists.
2504 36. Code for decoding IPv6 addresses in host lists is now included, even if
2505 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2506 address was recognized as an IP address, but was then not correctly decoded
2507 into binary, causing unexpected and incorrect effects when compared with
2514 1. Very minor rewording of debugging text in manualroute to say "list of
2515 hosts" instead of "hostlist".
2517 2. If verify=header_syntax was set, and a header line with an unqualified
2518 address (no domain) and a large number of spaces between the end of the
2519 name and the colon was received, the reception process suffered a buffer
2520 overflow, and (when I tested it) crashed. This was caused by some obsolete
2521 code that should have been removed. The fix is to remove it!
2523 3. When running in the test harness, delay a bit after writing a bounce
2524 message to get a bit more predictability in the log output.
2526 4. Added a call to search_tidyup() just before forking a reception process. In
2527 theory, someone could use a lookup in the expansion of smtp_accept_max_
2528 per_host which, without the tidyup, could leave open a database connection.
2530 5. Added the variables $recipient_data and $sender_data which get set from a
2531 lookup success in an ACL "recipients" or "senders" condition, or a router
2532 "senders" option, similar to $domain_data and $local_part_data.
2534 6. Moved the writing of debug_print from before to after the "senders" test
2537 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2538 problems for message scanning, either using a data ACL, or using
2539 local_scan() because the Received: header was not generated till after they
2540 were called (in order to set the time as the time of reception completion).
2541 I have revised the way this works. The header is now generated after the
2542 body is received, but before the ACL or local_scan() are called. After they
2543 are run, the timestamp in the header is updated.
2549 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2550 before starting a queue runner without re-exec. This happened only when
2551 deliver_drop_privilege was set or when the Exim user was set to root. The
2552 effect of the bug was that timeouts during subsequent deliveries caused
2553 crashes instead of being properly handled. The handler is now left at its
2554 default (and expected) setting.
2556 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2557 message, again when deliver_drop_privilege is set or Exim is run as root.
2558 The bug described in (1) was not present in this case, but the tidying up
2559 of the other signals was missing. I have made the two cases consistent.
2561 3. The ignore_target_hosts setting on a manualroute router was being ignored
2562 for hosts that were looked up using the /MX notation.
2564 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2567 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2568 operated on the sender address. After changing the $sender_address to <>
2569 for the sender address verify, Exim was re-instated it as the original
2570 (before rewriting) address, but remembering that it had rewritten it, so it
2571 wasn't rewriting it again. This bug also had the effect of breaking the
2572 sender address verification caching when the sender address was rewritten.
2574 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2575 This has been changed so that if the ip literal address matches
2576 ignore_target_hosts, the router declines.
2578 7. Added expansion conditions match_domain, match_address, and match_local_
2579 part (NOT match_host).
2581 8. The placeholder for the Received: header didn't have a length field set.
2583 9. Added code to Exim itself and to exim_lock to test for a specific race
2584 condition that could lead to file corruption when using MBX delivery. The
2585 issue is with the lockfile that is created in /tmp. If this file is removed
2586 after a process has opened it but before that process has acquired a lock,
2587 there is the potential for a second process to recreate the file and also
2588 acquire a lock. This could lead to two Exim processes writing to the file
2589 at the same time. The added code performs the same test as UW imapd; it
2590 checks after acquiring the lock that its file descriptor still refers to
2591 the same named file.
2593 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2594 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2597 11. Added the smtp_active_hostname option. If used, this will typically be made
2598 to depend on the incoming interface address. Because $interface_address is
2599 not set up until the daemon has forked a reception process, error responses
2600 that can happen earlier (such as "too many connections") no longer contain
2603 12. If an expansion in a condition on a "warn" statement fails because a lookup
2604 defers, the "warn" statement is abandoned, and the next ACL statement is
2605 processed. Previously this caused the whole ACL to be aborted.
2607 13. Added the iplsearch lookup type.
2609 14. Added ident_timeout as a log selector.
2611 15. Added tls_certificate_verified as a log selector.
2613 16. Added a global option tls_require_ciphers (compare the smtp transport
2614 option of the same name). This controls incoming TLS connections.
2616 17. I finally figured out how to make tls_require_ciphers do a similar thing
2617 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2618 before starting the TLS session.
2620 18. Tabs are now shown as \t in -bP output.
2622 19. If the log selector return_path_on_delivery was set, Exim crashed when
2623 bouncing a message because it had too many Received: header lines.
2625 20. If two routers both had headers_remove settings, and the first one included
2626 a superfluous trailing colon, the final name in the first list and the
2627 first name in the second list were incorrectly joined into one item (with a
2628 colon in the middle).
2634 1. Added -C and -D options to the exinext utility, mainly to make it easier
2635 to include in the automated testing, but these could be helpful when
2636 multiple configurations are in use.
2638 2. The exinext utility was not formatting the output nicely when there was
2639 an alternate port involved in the retry record key, nor when there was a
2640 message id as well (for retries that were specific to a specific message
2641 and a specific host). It was also confused by IPv6 addresses, because of
2642 the additional colons they contain. I have fixed the IPv4 problem, and
2643 patched it up to do a reasonable job for IPv6.
2645 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2646 delivery, the log line now contains "pipelined" if PIPELINING was used.
2648 4. An SMTP transport process used to panic and die if the bind() call to set
2649 an explicit outgoing interface failed. This has been changed; it is now
2650 treated in the same way as a connect() failure.
2652 5. A reference to $sender_host_name in the part of a conditional expansion
2653 that was being skipped was still causing a DNS lookup. This no longer
2656 6. The def: expansion condition was not recognizing references to header lines
2657 that used bh_ and bheader_.
2659 7. Added the _cache feature to named lists.
2661 8. The code for checking quota_filecount in the appendfile transport was
2662 allowing one more file than it should have been.
2664 9. For compatibility with Sendmail, the command line option
2672 and sets the incoming protocol and host name (for trusted callers). The
2673 host name and its colon can be omitted when only the protocol is to be set.
2674 Note the Exim already has two private options, -pd and -ps, that refer to
2675 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2676 "s", but I don't think that's a major issue.
2678 10. A number of refactoring changes to the code, none of which should affect
2681 (a) The number of logging options was getting close to filling up the
2682 32-bit word that was used as a bit map. I have split them into two classes:
2683 those that are passed in the argument to log_write(), and those that are
2684 only ever tested independently outside of that function. These are now in
2685 separate 32-bit words, so there is plenty of room for expansion again.
2686 There is no change in the user interface or the logging behaviour.
2688 (b) When building, for example, log lines, the code previously used a
2689 macro that called string_cat() twice, in order to add two strings. This is
2690 not really sufficiently general. Furthermore, there was one instance where
2691 it was actually wrong because one of the argument was used twice, and in
2692 one call a function was used. (As it happened, calling the function twice
2693 did not affect the overall behaviour.) The macro has been replaced by a
2694 function that can join an arbitrary number of extra strings onto a growing
2697 (c) The code for expansion conditions now uses a table and a binary chop
2698 instead of a serial search (which was left over from when there were very
2699 few conditions). Also, it now recognizes conditions like "pam" even when
2700 the relevant support is not compiled in: a suitably worded error message is
2701 given if an attempt is made to use such a condition.
2703 11. Added ${time_interval:xxxxx}.
2705 12. A bug was causing one of the ddress fields not to be passed back correctly
2706 from remote delivery subprocesses. The field in question was not being
2707 subsequently used, so this caused to problems in practice.
2709 13. Added new log selectors queue_time and deliver_time.
2711 14. Might have fixed a bug in maildirsizefile handling that threw up
2712 "unexpected character" debug warnings, and recalculated the data
2713 unnecessarily. In any case, I expanded the warning message to give more
2716 15. Added the message "Restricted characters in address" to the statements in
2717 the default ACL that block characters like @ and % in local parts.
2719 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2720 Three changes have been made:
2722 (a) There was a serious bug; a negative response to MAIL caused the whole
2723 recipient domain to be cached as invalid, thereby blocking all messages
2724 to all local parts at the same domain, from all senders. This bug has
2725 been fixed. The domain is no longer cached after a negative response to
2726 MAIL if the sender used is not empty.
2728 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2731 (c) A new callout option, "use_sender" has been added for people who want
2732 the modified behaviour.
2738 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2741 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2742 indeed breaks things for older releases.
2744 3. Added additional logging to the case where there is a problem reading data
2745 from a filter that is running in a subprocess using a pipe, in order to
2746 try to track down a specific problem.
2748 4. Testing facility fudge: when running in the test harness and attempting
2749 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2750 getting "No route to host". Convert this to a timeout.
2752 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2755 6. Some OS don't have socklen_t but use size_t instead. This affects the
2756 fifth argument of getsockopt() amongst other things. This is now
2757 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2758 can be set for individual OS. I have set it for SunOS5, OSF1, and
2759 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2760 some earlier ones do not.
2762 7. Change 4.30/15 was not doing the test caselessly.
2764 8. The standard form for an IPv6 address literal was being rejected by address
2765 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2766 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2767 this, as well as the form without the "IPv6" on the front (but only when
2768 address literals are enabled, of course).
2770 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2772 10. Exim crashed if a message with an empty sender address specified by -f
2773 encountered a router with an errors_to setting. This could be provoked only
2774 by a command such as
2778 where an empty string was supplied; "<>" did not hit this bug.
2780 11. Installed PCRE release 4.5.
2782 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2783 remained set. It is now erased.
2785 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2786 times from message ids (which are base 36 rather than the normal 62).
2788 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2789 were being counted as actual protocol errors, and logged if the log
2790 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2791 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2792 and DATA following a set of rejected RCPTs do not count as protocol errors.
2793 In other words, Exim assumes they were pipelined, though this may not
2794 actually be the case. Of course, in all cases the client gets an
2795 appropriate error code.
2797 15. If a lookup fails in an ACL condition, a message about the failure may
2798 be available; it is used if testing the ACL cannot continue, because most
2799 such messages specify what the cause of the deferral is. However, some
2800 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2801 that caused an old message to be retained and used if a later statement
2802 caused a defer, replacing the real cause of the deferral.
2804 16. If an IP address had so many PTR records that the DNS lookup buffer
2805 was not large enough to hold them, Exim could crash while trying to process
2806 the truncated data. It now detects and logs this case.
2808 17. Further to 4.21/58, another change has been made: if (and only if) the
2809 first line of a message (the first header line) ends with CRLF, a bare LF
2810 in a subsequent header line has a space inserted after it, so as not to
2811 terminate the header.
2813 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2814 unnecessarily, used atoi() instead of strtol(), and didn't check the
2815 termination when getting file sizes from file names by regex.
2817 19. Completely re-implemented the support for maildirsize files, in the light
2818 of a number of problems with the previous contributed implementation
2819 (4.30/29). In particular:
2821 . If the quota is zero, the maildirsize file is maintained, but no quota is
2824 . If the maildir directory does not exist, it is created before any attempt
2825 to write a maildirsize file.
2827 . The quota value in the file is just a cache; if the quota is changed in
2828 the transport, the new value overrides.
2830 . A regular expression is available for excluding directories from the
2833 20. The autoreply transport checks the characters in options that define the
2834 message's headers; it allows continued headers, but it was checking with
2835 isspace() after an embedded newline instead of explicitly looking for a
2838 21. If all the "regular" hosts to which an address was routed had passed their
2839 expiry times, and had not reached their retry times, the address was
2840 bounced, even if fallback hosts were defined. Now Exim should go on to try
2843 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2844 equivalent code in the SMTP transport. Some hosts send humungous responses
2845 to HELO/EHLO, more than 1024 it seems.
2847 23. Refactoring: code in filter.c used (void *) for "any old type" but this
2848 gives compiler warnings in some environments. I've now done it "properly",
2851 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2852 (because of problems with the built-in one) was declared to return uschar *
2853 instead of char *, causing compiler failure.
2855 25. Fixed a file descriptor leak when processing alias/forward files.
2857 26. Fixed a minor format string issue in dbfn.c.
2859 27. Typo in exim.c: ("dmbnz" for "dbmnz").
2861 28. If a filter file refered to $h_xxx or $message_headers, and the headers
2862 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2865 29. When a sender address is verified, it is cached, to save repeating the test
2866 when there is more than one recipient in a message. However, when the
2867 verification involves a callout, it is possible for different callout
2868 options to be set for different recipients. It is too complicated to keep
2869 track of this in the cache, so now Exim always runs a verification when a
2870 callout is required, relying on the callout cache for the optimization.
2871 The overhead is duplication of the address routing, but this should not be
2874 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
2875 to be verified with callout=postmaster, and the main callout worked but the
2876 postmaster check failed, the verification correctly failed. However, if a
2877 subsequent RCPT command asked for sender verification *without* the
2878 postmaster check, incorrect caching caused this verification also to fail,
2881 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2882 it was not caching the DNS options (qualify_single, search_parents) that
2883 were used when the lookup failed. A subsequent lookup with different
2884 options therefore always gave the same answer, though there were cases
2885 where it should not have. (Example: a "domains = !$mx_any" option on a
2886 dnslookup router: the "domains" option is always processed without any
2887 widening, but the router might have qualify_single set.) Now Exim uses the
2888 cached value only when the same options are set.
2890 32. Added John Jetmore's "exipick" utility to the distribution.
2892 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
2893 than a timeout (e.g. a certificate is required, and is not provided), an
2894 Exim server now closes the connection immediately. Previously it waited for
2895 the client to close - but if the client is SSL, it seems that they each
2896 wait for each other, leading to a delay before one of them times out.
2898 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2899 maintained 0.8.x compatibility because I don't think many are using it, and
2900 it is clearly obsolete.
2902 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2905 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2906 client certificate was expired. A simple patch fixes this, though I don't
2907 understand the full logic of why the verify callback is called multiple
2910 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2911 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2912 which causes problems with some clients (such as the Certicom SSL Plus
2913 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2914 disables the coutermeasure allowing Eudora to connect."
2916 38. Exim was not checking that a write() to a log file succeeded. This could
2917 lead to Bad Things if a log got too big, in particular if it hit a file
2918 size limit. Exim now panics and dies if it cannot write to a log file, just
2919 as it does if it cannot open a log file.
2921 39. Modified OS/Makefile-Linux so that it now contains
2923 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2925 The two -D definitions ensure that Exim is compiled with large file
2926 support, which makes it possible to handle log files that are bigger than
2929 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2930 instance) a domain was checked against a named list that involved a lookup,
2931 causing $domain_data to be set, then another domain was checked against the
2932 same list, then the first domain was re-checked, the value of $domain_data
2933 after the final check could be wrong. In particular, if the second check
2934 failed, it could be set empty. This bug probably also applied to
2937 41. The strip_trailing_dot option was not being applied to the address given
2938 with the -f command-line option.
2940 42. The code for reading a message's header from the spool was incrementing
2941 $received_count, but never initializing it. This meant that the value was
2942 incorrect (doubled) while delivering a message in the same process in which
2943 it was received. In the most common configuration of Exim, this never
2944 happens - a fresh exec is done - but it can happen when
2945 deliver_drop_privilege is set.
2947 43. When Exim logs an SMTP synchronization error - client data sent too soon -
2948 it now includes up to 150 characters of the unexpected data in the log
2951 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2952 and building data strings. The size of both of these buffers was 10 000
2953 bytes - far larger than anybody would *ever* want, thought I. Needless to
2954 say, somebody hit the limit. I have increased the maximum line length to
2955 20 000 and the maximum data length of concatenated lines to 100 000. I have
2956 also fixed two bugs, because there was no checking on these buffers. Tsk,
2957 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2958 buffer is too small.
2960 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2961 lsearch lookups. Now it does.
2963 46. When parsing a route_list item in a manualroute router, a fixed-length
2964 buffer was used for the list of hosts. I made this 1024 bytes long,
2965 thinking that nobody would ever have a list of hosts that long. Wrong.
2966 Somebody had a whole pile of complicated expansion conditions, and the
2967 string was silently truncated, leading to an expansion error. It turns out
2968 that it is easier to change to an unlimited length (owing to other changes
2969 that have happened since this code was originally written) than to build
2970 structure for giving a limitation error. The length of the item that
2971 expands into the list of hosts is now unlimited.
2973 47. The lsearch lookup could not handle data where the length of text line was
2974 more than 4095 characters. Such lines were truncated, leading to shortened
2975 data being returned. It should now handle lines of any length.
2977 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2978 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2981 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
2982 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2983 no longer happens in comment lines. A list of replacements is now placed
2984 at the head of all of the source files, except those whose only change is
2985 to replace PERL_COMMAND in the very first #! line.
2987 50. Replaced the slow insertion sort in queue.c, for sorting the list of
2988 messages on the queue, with a bottom-up merge sort, using code contributed
2989 by Michael Haardt. This should make operations like -bp somewhat faster on
2990 large queues. It won't affect queue runners, except when queue_run_in_order
2993 51. Installed eximstats 1.31 in the distribution.
2995 52. Added support for SRV lookups to the dnslookup router.
2997 53. If an ACL referred to $message_body or $message_body_end, the value was not
2998 reset for any messages that followed in the same SMTP session.
3000 54. The store-handling optimization for building very long strings was not
3001 differentiating between the different store pools. I don't think this
3002 actually made any difference in practice, but I've tidied it.
3004 55. While running the routers to verify a sender address, $sender_address
3005 was still set to the sender address. This is wrong, because when routing to
3006 send a bounce to the sender, it would be empty. Therefore, I have changed
3007 it so that, while verifying a sender address, $sender_address is set to <>.
3008 (There is no change to what happens when verifying a recipient address.)
3010 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3011 target A or AAAA records (if not already returned) without resetting the
3012 qualify_single or search_parents options of the DNS resolver. These are
3013 inappropriate in this case because the targets of MX and SRV records must
3014 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3015 target that, when qualified, matched something in the local domain. These
3016 two options are now turned off when doing these lookups.
3018 57. It seems that at least some releases of Reiserfs (which does not have the
3019 concept of a fixed number of inodes) returns zero and not -1 for the
3020 number of available inodes. This interacted badly with check_spool_inodes,
3021 which assumed that -1 was the "no such thing" setting. What I have done is
3022 to check that the total number of inodes is greater than zero before doing
3023 the test of how many are available.
3025 58. When a "warn" ACL statement has a log_message modifier, the message is
3026 remembered, and not repeated. This is to avoid a lot of repetition when a
3027 message has many recipients that cause the same warning to be written.
3028 Howewer, Exim was preserving the list of already written lines for an
3029 entire SMTP session, which doesn't seem right. The memory is now reset if a
3030 new message is started.
3032 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3033 debugging output unless log_rewrite was also set.
3035 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3036 of (int)(handle) when we know that handle contains (void *)(-1).
3038 61. The Exim daemon panic-logs an error return when it closes the incoming
3039 connection. However "connection reset by peer" seems to be common, and
3040 isn't really an error worthy of noting specially, so that particular error
3043 62. When Exim is trying to find all the local interfaces, it used to panic and
3044 die if the ioctl to get the interface flags failed. However, it seems that
3045 on at least one OS (Solaris 9) it is possible to have an interface that is
3046 included in the list of interfaces, but for which you get a failure error
3047 for this call. This happens when the interface is not "plumbed" into a
3048 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3049 failure of the "get flags" call assumes that the interface is down.
3051 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3052 makes life easier for people who are doing arithmetic on fields extracted
3053 from dates, where you often get leading zeros that should not be
3054 interpreted as octal.
3056 64. Added qualify_domain to the redirect router, to override the global
3059 65. If a pathologically long header line contained very many addresses (the
3060 report of this problem mentioned 10 000) and each of them was rewritten,
3061 Exim could use up a very large amount of memory. (It kept on making new
3062 copies of the header line as it rewrote, and never released the old ones.)
3063 At the expense of a bit more processing, the header rewriting function has
3064 been changed so that it no longer eats memory in this way.
3066 66. The generation of the Received: header has been moved from the time that a
3067 message starts to be received, to the time that it finishes. The timestamp
3068 in the Received: header should now be very close to that of the <= log
3069 line. There are two side-effects of this change:
3071 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3072 logged header lines no longer include the local Received: line, because
3073 it has not yet been created. The same applies to a copy of the message
3074 that is returned to a non-SMTP sender when a message is rejected.
3076 (b) When a filter file is tested using -bf, no additional Received: header
3077 is added to the test message. After some thought, I decided that this
3080 This change does not affect the value of $received_for. It is still set
3081 after address rewriting, but before local_scan() is called.
3083 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3085 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3086 gave an unhelpful panic error message, and a defer error. I have managed to
3087 change this behaviour so that it now rejects any supplied certificate,
3088 which seems right, as the list of acceptable certificates is empty.
3090 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3091 gave an unhelpful defer error. I have not managed to make this reject any
3092 supplied certificates, but the error message it gives is "no certificate
3093 supplied", which is not helpful.
3095 70. exigrep's output now also includes lines that are not associated with any
3096 message, but which match the given pattern. Implemented by a patch from
3097 Martin Sluka, which also tidied up the Perl a bit.
3099 71. Recipient callout verification, like sender verification, was using <> in
3100 the MAIL FROM command. This isn't really the right thing, since the actual
3101 sender may affect whether the remote host accepts the recipient or not. I
3102 have changed it to use the actual sender in the callout; this means that
3103 the cache record is now keyed on a recipient/sender pair, not just the
3104 recipient address. There doesn't seem to be a real danger of callout loops,
3105 since a callout by the remote host to check the sender would use <>.
3106 [SEE ABOVE: changed after hitting problems.]
3108 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3109 temporary errors. However, in the case of such a code being given after
3110 the end of a data transmission (i.e. after ".") Exim was failing to write
3111 a retry record for the message. (Yes, there was some broken host that was
3112 actually sending 8xx at this point.)
3114 73. An unknown lookup type in a host list could cause Exim to panic-die when
3115 the list was checked. (An example that provoked this was putting <; in the
3116 middle of a list instead of at the start.) If this happened during a DATA
3117 ACL check, a -D file could be left lying around. This kind of configuration
3118 error no longer causes Exim to die; instead it causes a defer errror. The
3119 incident is still logged to the main and panic logs.
3121 74. Buglet left over from Exim 3 conversion. The message "too many messages
3122 in one connection" was written to the rejectlog but not the mainlog, except
3123 when address rewriting (yes!) was being logged.
3125 75. Added write_rejectlog option.
3127 76. When a system filter was run not as root (that is, when system_filter_user
3128 was set), the values of the $n variables were not being returned to the
3129 main process; thus, they were not subsequently available in the $sn
3132 77. Added +return_path_on_delivery log selector.
3134 78. A connection timeout was being treated differently from recipients deferred
3135 when testing hosts_max_try with a message that was older than the host's
3136 retry timeout. (The host should not be counted, thus allowing all hosts to
3137 be tried at least once before bouncing.) This may have been the cause of an
3138 occasionally reported bug whereby a message would remain on the queue
3139 longer than the retry timeout, but would be bounced if a delivery was
3140 forced. I say "may" because I never totally pinned down the problem;
3141 setting up timeout/retry tests is difficult. See also the next item.
3143 79. The ultimate address timeout was not being applied to errors that involved
3144 a combination of host plus message (for example, a timeout on a MAIL
3145 command). When an address resolved to a number of possible hosts, and they
3146 were not all tried for each delivery (e.g. because of hosts_max_try), a
3147 message could remain on the queue longer than the retry timeout.
3149 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3152 81. Fixed an obscure SMTP outgoing bug which required at least the following
3153 conditions: (a) there was another message waiting for the same server;
3154 (b) the server returned 5xx to all RCPT commands in the first message so
3155 that the message was not completed; (c) the server dropped the connection
3156 or gave a negative response to the RSET that Exim sends to abort the
3157 transaction. The observed case was a dropped connection after DATA that had
3158 been sent in pipelining mode. That is, the server had advertised PIPELINING
3159 but was not implementing it correctly. The effect of the bug was incorrect
3160 behaviour, such as trying another host, and this could lead to a crash.
3166 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3167 and daemon.c were passed as pointers to ints; they should have been
3168 pointers to socklen_t variables (which are typically unsigned ints).
3170 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3173 3. Fixed a really odd bug that affected only the testing scheme; patching a
3174 certain fixed string in the binary changed the value of another string that
3175 happened to be identical to the end of the original first string.
3177 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3178 name", it returns that address as the IP address. On some operating
3179 systems (e.g. Solaris), it also passes back the IP address string as the
3180 "host name". However, on others (e.g. Linux), it passes back an empty
3181 string. Exim wasn't checking for this, and was changing the host name to an
3182 empty string, assuming it had been canonicized.
3184 5. Although rare, it is permitted to have more than one PTR record for a given
3185 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3186 all the names associated with an address, because they do in Solaris.
3187 However, it seems that they do not in Linux for data that comes from the
3188 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3189 I found this out when I moved to a new Linux workstation and tried to run
3190 the Exim test suite.
3192 To get round this problem I have changed the code so that it now does its
3193 own call to the DNS to look up PTR records when searching for a host name.
3194 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3195 addresses that are only in /etc/hosts are still found.
3197 This behaviour is, however, controlled by an option called host_lookup_
3198 order, which defaults to "bydns:byaddr". If people want to use the other
3199 order, or indeed, just use one or the other means of lookup, they can
3200 specify it in this variable.
3202 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3203 some operating systems, this comes back from gethostbyaddr() as an empty
3204 string, and this is what Exim used to test for. However, it seems that in
3205 other systems, "." is yielded. Exim now tests for this case too.
3207 7. The values of check_spool_space and check_log_space are now held internally
3208 as a number of kilobytes instead of an absolute number of bytes. If a
3209 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3210 kilobyte. This means that much larger values can be stored.
3212 8. Exim monitor: an attempt to get the action menu when not actually pointing
3213 at a message produces an empty menu entitled "No message selected". This
3214 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3215 no entries in it ("Shell widget menu has zero width and/or height"). So I
3216 have added a single, blank menu entry in this case.
3218 9. Added ${quote_local_part.
3220 10. MIME decoding is now applied to the contents of Subject: header lines when
3223 11. Now that a reference to $sender_host_address automatically causes a reverse
3224 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3225 host lookup before query-style lookups in lists that might use this
3226 variable. This has therefore been abolished, and the "net-" prefix is no
3227 longer necessary for query-style lookups.
3229 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3230 have been a typo for LFLAGS, so it has been changed.
3232 13. The install script calls Exim with "-C /dev/null" in order to find the
3233 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3234 to be output. Howeve, since Exim outputs its version number before the
3235 error, it didn't break the script. It just looked ugly. I fixed this by
3236 always allowing "-C /dev/null" if the caller is root.
3238 14. Ignore overlarge ACL variable number when reading spool file - insurance
3239 against a later release with more variables having written the file.
3241 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3242 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3243 this, as well as the form without the "IPv6" on the front.
3245 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3246 OS/Makefile-Darwin file.
3248 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3249 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3251 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3252 variable. However, some OS use ints for inodes. I've added cast to long int
3253 to get rid of the compiler warning.
3255 19. I had forgotten to lock out "/../" in configuration file names when
3256 ALT_CONFIG_PREFIX was set.
3258 20. Routers used for verification do not need to specify transports. However,
3259 if such a router generated a host list, and callout was configured, Exim
3260 crashed, because it could not find a port number from the (non-existent)
3261 transport. It now assumes port 25 in this circumstance.
3263 21. Added the -t option to exigrep.
3265 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3266 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3267 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3268 them is set, LOOKUP_LSEARCH is forced.
3270 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3272 24. Added sender and host information to the "rejected by local_scan()" log
3273 line; previously there was no indication of these.
3275 25. Added .include_if_exists.
3277 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3278 Linux. It turns out that not all file systems support this. Apparently some
3279 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3280 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3281 not supported on the file descriptor, is now ignored when Exim is trying to
3282 sync a directory. This applies only to Linux.
3284 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3286 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3287 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3290 29. Added support for maildirsize files from supplied patch (modified a bit).
3292 30. The use of :fail: followed by an empty string could lead Exim to respond to
3293 sender verification failures with (e.g.):
3295 550 Verification failed for <xxx>
3296 550 Sender verify failed
3298 where the first response line was missing the '-' that indicates it is not
3299 the final line of the response.
3301 31. The loop for finding the name of the user that called Exim had a hardwired
3302 limit of 10; it now uses the value of finduser_retries, which is used for
3303 all other user lookups.
3305 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3308 33. Exim was neglecting to zero errno before one call of strtol() when
3309 expanding a string and expecting an integer value. On some systems this
3310 resulted in spurious "integer overflow" errors. Also, it was casting the
3311 result into an int without checking.
3313 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3314 did not work. The code looks as if it has *never* worked, though it appears
3315 to have been documented since at least releast 1.62. I have made it work.
3317 35. The "timeout_DNS" error in retry rules, also documented since at least
3318 1.62, also never worked. As it isn't clear exactly what this means, and
3319 clearly it isn't a major issue, I have abolished the feature by treating it
3320 as "timeout", and writing a warning to the main and panic logs.
3322 36. The display of retry rules for -brt wasn't always showing the error code
3325 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3326 timeout_connect_A, timeout_connect_MX.
3328 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3329 to the empty sender.
3331 39. The daemon was not analysing the content of -oX till after it had closed
3332 stderr and disconnected from the controlling terminal. This meant that any
3333 syntax errors were only noted on the panic log, and the return code from
3334 the command was 0. By re-arranging the code a little, I've made the
3335 decoding happen first, so such errors now appear on stderr, and the return
3336 code is 1. However, the actual setting up of the sockets still happens in
3337 the disconnected process, so errors there are still only recorded on the
3340 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3341 connections (as happens on some IP stacks) was logged at start up time as
3342 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3343 it from "IPv6 and IPv4", which means that two separate sockets are being
3346 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3347 says whether AF_INET or AF_INET6 was passed as an argument.
3349 42. Exiwhat output was messed up when time zones were included in log
3352 43. Exiwhat now gives more information about the daemon's listening ports,
3353 and whether -tls-on-connect was used.
3355 44. The "port" option of the smtp transport is now expanded.
3357 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3358 silently ignored. Now an error message is written to the main and panic
3361 46. There's a new ACL modifier called "logwrite" which writes to a log file
3362 as soon as it is encountered.
3364 47. Added $local_user_uid and $local_user_gid at routing time.
3366 48. Exim crashed when trying to verify a sender address that was being
3369 49. Exim was recognizing only a space character after ".include". It now also
3370 recognizes a tab character.
3372 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3373 extracting the relevant information from the specification. The man page no
3374 longer contains scrambled data for the -d option, and I've added a section
3375 at the front about calling Exim under different names.
3377 51. Added "extra_headers" argument to the "mail" command in filter files.
3379 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3382 53. Installed eximstats 1.29.
3384 54. Added transport_filter_timeout as a generic transport option.
3386 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3387 Cc: header lines. This was required by RFC 822, but it not required by RFC
3390 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3391 incoming messages that did not have them. Now it does so only if the
3392 message originates locally, that is, if there is no associated remote host
3393 address. When Resent- header lines are present, this applies to the Resent-
3394 lines rather than the non-Resent- lines.
3396 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3397 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3399 58. Messages for configuration errors now include the name of the main
3400 configuration file - useful now that there may be more than one file in a
3401 list (.included file names were always shown).
3403 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3404 for those rare installations that do not start the daemon as root or run it
3405 setuid root. I've cut out the call to initgroups() if the daemon is not
3408 60. The Exim user and group can now be bound into the binary as text strings
3409 that are looked up at the start of Exim's processing.
3411 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3413 62. Added $mailstore_basename variable.
3415 63. Installed patch to sieve.c from Michael Haardt.
3417 64. When Exim failed to open the panic log after failing to open the main log,
3418 the original message it was trying to log was written to stderr and debug
3419 output, but if they were not available (the usual case in production), it
3420 was lost. Now it is written to syslog before the two lines that record the
3421 failures to open the logs.
3423 65. Users' Exim filters run in subprocesses under the user's uid. It is
3424 possible for a "deliver" command or an alias in a "personal" command to
3425 provoke an address rewrite. If logging of address rewriting is configured,
3426 this fails because the process is not running as root or exim. There may be
3427 a better way of dealing with this, but for the moment (because 4.30 needs
3428 to be released), I have disabled address rewrite logging when running a
3429 filter in a non-root, non-exim process.
3435 1. The buildconfig auxiliary program wasn't quoting the value set for
3436 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3437 not defined. This bug was masked in 4.22 by the effect that was fixed in
3440 2. Some messages that were rejected after a message id was allocated were
3441 shown as "incomplete" by exigrep. It no longer does this for messages that
3442 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3444 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3445 have allow_domain_literals set, the ID did not get logged in the <= line.
3446 Domain literals are now always recognized in Message-ID: header lines.
3448 4. The first argument for a ${extract expansion item is the key name or field
3449 number. Leading and trailing spaces in this item were not being ignored,
3450 causing some misleading effects.
3452 5. When deliver_drop_privilege was set, single queue runner processes started
3453 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3454 same command in the process it spins off) were not dropping privilege.
3456 6. When the daemon running as "exim" started a queue runner, it always
3457 re-executed Exim in the spun-off process. This is a waste of effort when
3458 deliver_drop_privilege is set. The new process now just calls the
3459 queue-runner function directly.
3465 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3468 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3469 ignored. Though the use of -oP was forcing the writing of a pid file, it
3470 was always written to the default place.
3472 3. If the message "no IP address found for host xxxx" is generated during
3473 incoming verification, it is now followed by identification of the incoming
3474 connection (so you can more easily find what provoked it).
3476 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3478 5. Added some features to "harden" Exim a bit more against certain attacks:
3480 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3481 be put in Local/Makefile. This is like the never_users runtime option,
3482 but it cannot be overridden. The default setting is "root".
3484 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3485 prefix string with which any file named in a -C command line option
3488 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3489 is retained for -C and -D only if the caller of Exim is root. Without
3490 it, the exim user may also use -C and -D and retain privilege.
3492 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3493 command line option is disabled.
3495 6. Macro names set by the -D option must start with an upper case letter, just
3496 like macro names defined in the configuration file.
3498 7. Added "dereference=" facility to LDAP.
3500 8. Two instances of the typo "uknown" in the source files are fixed.
3502 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3503 the Configure-Makefile script screwed up while processing it.
3505 10. Incorporated PCRE 4.4.
3507 11. The SMTP synchronization check was not operating right at the start of an
3508 SMTP session. For example, it could not catch a HELO sent before the client
3509 waited for the greeting. There is now a check for outstanding input at the
3510 point when the greeting is written. Because of the duplex, asynchronous
3511 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3512 way, but not yet received, when the check is performed.
3514 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3515 on TCP/IP sockets, because this apparently causes some broken clients to
3518 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3519 unchanged) from the Cygwin maintainer.
3521 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3522 was supported instead of testing for mbx. Effectively a typo.
3524 15. The spa authenticator server code was not checking that the input it
3525 received was valid base64.
3527 16. The debug output line for the "set" modifier in ACLs was not showing the
3528 name of the variable that was being set.
3530 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3532 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3533 lookup on its own. Something else was needed to trigger the lookup. For
3534 example, a match in host_lookup or the need for a host name in a host list.
3535 Now, if $sender_host_name is referenced and the host name has not yet been
3536 looked up, a lookup is performed. If the lookup fails, the variable remains
3537 empty, and $host_lookup_failed is set to "1".
3539 19. Added "eqi" as a case-independent comparison operator.
3541 20. The saslauthd authentication condition could segfault if neither service
3542 nor realm was specified.
3544 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3545 error message that was logged was misleading, and incoming SMTP
3546 connections were dropped. The message is now more accurate, and temporary
3547 errors are given to SMTP connections.
3549 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3550 (except RSET) with a 421 error, until QUIT is received. However, it was
3551 failing to send a response to QUIT.
3553 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3554 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3555 ACL is now run after the helo_try_verify_hosts code.
3557 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3558 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3559 case-independent, so other case variants are also recognized). Apparently
3560 some systems use these upper case variants.
3562 25. If more than two messages were waiting for the same host, and a transport
3563 filter was specified for the transport, Exim sent two messages over the
3564 same TCP/IP connection, and then failed with "socket operation on non-
3565 socket" when it tried to send the third.
3567 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3569 27. The extern definition of crypt16() in expand.c was not being excluded when
3570 the OS had its own crypt16() function.
3572 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3573 as a preferred synonym for return_size_limit, both as an option and as an
3576 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3578 30. Changed the default configuration ACL to relax the local part checking rule
3579 for addresses that are not in any local domains. For these addresses,
3580 slashes and pipe symbols are allowed within local parts, but the sequence
3581 /../ is explicitly forbidden.
3583 31. SPA server authentication was not clearing the challenge buffer before
3586 32. log_message in a "warn" ACL statement was writing to the reject log as
3587 well as to the main log, which contradicts the documentation and doesn't
3588 seem right (because no rejection is happening). So I have stopped it.
3590 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3591 However, I am unable to do any testing of this.
3593 34. Fixed an infelicity in the appendfile transport. When checking directories
3594 for a mailbox, to see if any needed to be created, it was accidentally
3595 using path names with one or more superfluous leading slashes; tracing
3596 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3598 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3599 contents are added to the log line that is written for every discarded
3600 recipient. (Previously a log_message setting was ignored.)
3602 36. The ${quote: operator now quotes the string if it is empty.
3604 37. The install script runs exim in order to find its version number. If for
3605 some reason other than non-existence or emptiness, which it checks, it
3606 could not run './exim', it was installing it with an empty version number,
3607 i.e. as "exim-". This error state is now caught, and the installation is
3610 38. An argument was missing from the function that creates an error message
3611 when Exim fails to connect to the socket for saslauthd authentication.
3612 This could cause Exim to crash, or give a corrupted message.
3614 39. Added isip, isip4, and isip6 to ${if conditions.
3616 40. The ACL variables $acl_xx are now saved with the message, and can be
3617 accessed later in routers, transports, and filters.
3619 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3620 strings in the file are not string-expanded.
3622 42. If a MAIL command specified a SIZE value that was too large to fit into an
3623 int variable, the check against message_size_limit failed. Such values are
3624 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3625 day this will have to be increased, but I don't think I want to be around
3626 when emails are that large.
3633 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3634 iconv() is not standard in FreeBSD.
3636 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3637 IPv6 enabled. The observed symptom was a segmentation fault on return from
3638 the function os_common_find_running_interfaces() in src/os.c.
3640 3. In the check_special_case() function in daemon.c I had used "errno" as an
3641 argument name, which causes warnings on some systems. This was basically a
3642 typo, since it was named "eno" in the comments!
3644 4. The code that waits for the clock to tick (at a resolution of some fraction
3645 of a second) so as to ensure message-id uniqueness was always waiting for
3646 at least one whole tick, when it could have waited for less. [This is
3647 almost certainly not relevant at current processor speeds, where it is
3648 unlikely to ever wait at all. But we try to future-proof.]
3650 5. The function that sleeps for a time interval that includes fractions of a
3651 second contained a race. It did not block SIGALRM between setting the
3652 timer, and suspending (a couple of lines later). If the interval was short
3653 and the sigsuspend() was delayed until after it had expired, the suspension
3654 never ended. On busy systems this could lead to processes getting stuck for
3657 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3658 process, before it forks any delivery processes. The open lookup caching
3659 mechanism meant that the open file or database connection was passed into
3660 the delivery process. The problem was that delivery processes always tidy
3661 up cached lookup data. This could cause a problem for the next delivery
3662 process started by the queue runner, because the external queue runner
3663 process does not know about the closure. So the next delivery process
3664 still has data in the lookup cache. In the case of a file lookup, there was
3665 no problem because closing a file descriptor in a subprocess doesn't affect
3666 the parent. However, if the lookup was caching a connection to a database,
3667 the connection was closed, and the second delivery process was likely to
3668 see errors such as "PGSQL: query failed: server closed the connection
3669 unexpectedly". The problem has been fixed by closing all cached lookups
3670 in a queue runner before running a delivery process.
3672 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3673 seem to have the "const" qualifier which it has on other OS. I've
3676 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3677 *of the same type* (client or server) with the same public name that an
3678 error should be diagnosed.
3680 9. When Exim looked up a host name for an IP address, but failed to find the
3681 original IP address when looking up the host name (a safety check), it
3682 output the message "<ip address> does not match any IP for NULL", which was
3683 confusing, to say the least. The bug was that the host name should have
3684 appeared instead of "NULL".
3686 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3687 user that is built into the binary, and the -C or -D options is used, root
3688 privilege is dropped before the configuration file is read. In addition,
3689 logging is switched to stderr instead of the normal log files. If the
3690 configuration then re-defines the Exim user, the unprivileged environment
3691 is probably not what is expected, so Exim logs a panic warning message (but
3694 However, if deliver_drop_privilege is set, the unprivileged state may well
3695 be exactly what is intended, so the warning has been cut out in that case,
3696 and Exim is allowed to try to write to its normal log files.
3702 1. smtp_return_error_details was not giving details for temporary sender
3703 or receiver verification errors.
3705 2. Diagnose a configuration error if two authenticators have the same public
3708 3. Exim used not to create the message log file for a message until the first
3709 delivery attempt. This could be confusing when incoming messages were held
3710 for policy or load reasons. The message log file is now created at the time
3711 the message is received, and an initial "Received" line is written to it.
3713 4. The automatically generated man page for command line options had a minor
3714 bug that caused no ill effects; however, a more serious problem was that
3715 the procedure for building the man page automatically didn't always
3716 operate. Consequently, release 4.20 contains an out-of-date version. This
3717 shouldn't happen again.
3719 5. When building Exim with embedded Perl support, the script that builds the
3720 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3721 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3723 6. The freeze_tell option was not being used for messages that were frozen on
3724 arrival, either by an ACL or by local_scan().
3726 7. Added the smtp_incomplete_transaction log selector.
3728 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3729 was accepting AUTH without a new EHLO.
3731 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3732 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3735 10. Logging of TCP/IP connections (when configured) now happens in the main
3736 daemon process instead of the child process, so that the TCP/IP connection
3737 count is more accurate (but it can never be perfect).
3739 11. The use of "drop" in a nested ACL was not being handled correctly in the
3740 outer ACL. Now, if condition failure induced by the nested "drop" causes
3741 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3742 or "require"), the connection is dropped.
3744 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3745 that yield "discard" can now be used with an "accept" or a "discard" verb,
3746 but an error is generated for any others (because I can't see a useful way
3747 to define what should happen).
3749 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3750 are now processed in the same way as lines in the Exim configuration file.
3751 In particular, continuation lines are supported.
3753 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3755 15. Added -ti meaning -t -i.
3757 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3758 domains, and warn by logging if others are found.
3760 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3761 returned from GIFCONF when Exim is trying to find the list of interfaces on
3762 a host. The code in os.c has been modified to copy each ifreq to an aligned
3763 structure in all cases.
3765 Also, in some cases, the returned ifreq's were being copied to a 'struct
3766 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3767 means the last couple of bytes of an IPv6 address could be chopped if the
3768 ifreq contained only a normal sockaddr (14 bytes storage).
3770 18. Named domain lists were not supported in the hosts_treat_as_local option.
3771 An entry such as +xxxx was not recognized, and was treated as a literal
3774 19. Ensure that header lines added by a DATA ACL are included in the reject log
3775 if the ACL subsequently rejects the message.
3777 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3778 MD5 (which is deprecated).
3780 21. When testing a filter file using -bf, Exim was writing a message when it
3781 took the sender from a "From " line in the message, but it was not doing so
3782 when it took $return_path from a Return-Path: header line. It now does.
3784 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3785 with a valid header line field name (a series of printing characters
3786 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3788 23. Changed "disc" in the source to "disk" to conform to the documentation and
3789 the book and for uniformity.
3791 24. Ignore Sendmail's -Ooption=value command line item.
3793 25. When execve() failed while trying to run a command in a pipe transport,
3794 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3795 could be confused with a return value of 69 from the command itself. This
3796 has been changed to 127, the value the shell returns if it is asked to run
3797 a non-existent command. The wording for the related log line suggests a
3798 non-existent command as the problem.
3800 26. If received_header_text expands to an empty string, do not add a Received:
3801 header line to the message. (Well, it adds a token one on the spool, but
3802 marks it "old" so that it doesn't get used or transmitted.)
3804 27. Installed eximstats 1.28 (addition of -nt option).
3806 28. There was no check for failure on the call to getsockname() in the daemon
3807 code. This can fail if there is a shortage of resources on the system, with
3808 ENOMEM, for example. A temporary error is now given on failure.
3810 29. Contrary to the C standard, it seems that in some environments, the
3811 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3812 program. Exim now does this explicitly; it affects the formatting of
3813 timestamps using strftime().
3815 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3816 complaints. I've now initialized all the variables, to avoid this.
3818 32. Header lines added by a system filter were not being "seen" during
3819 transport-time rewrites.
3821 33. The info_callback() function passed to OpenSSL is set up with type void
3822 (*)(SSL *, int, int), as described somewhere. However, when calling the
3823 function (actually a macro) that sets it up, the type void(*)() is
3824 expected. I've put in a cast to prevent warnings from picky compilers.
3826 34. If a DNS black list lookup found a CNAME record, but there were no A
3827 records associated with the domain it pointed at, Exim crashed.
3829 35. If a DNS black list lookup returned more than one A record, Exim ignored
3830 all but the first. It now scans all returned addresses if a particular IP
3831 value is being sought. In this situation, the contents of the
3832 $dnslist_value variable are a list of all the addresses, separated by a
3835 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3836 to accept a host name and all its aliases if the forward lookup for any of
3837 them yielded the IP address of the incoming connection. Now it accepts only
3838 those names whose forward lookup yields the correct IP address. Any other
3839 names are discarded. This closes a loophole whereby a rogue DNS
3840 administrator could create reverse DNS records to break through a
3841 wildcarded host restriction in an ACL.
3843 37. If a user filter or a system filter that ran in a subprocess used any of
3844 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3845 the wrong values were passed to the pipe command ($thisaddress had the
3846 value of $0, $0 had the value of $1, etc). This bug was introduced by
3847 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3849 38. Improved the line breaking for long SMTP error messages from ACLs.
3850 Previously, if there was no break point between 40 and 75 characters, Exim
3851 left the rest of the message alone. Two changes have been made: (a) I've
3852 reduced the minimum length to 35 characters; (b) if it can't find a break
3853 point between 35 and 75 characters, it looks ahead and uses the first one
3854 that it finds. This may give the occasional overlong line, but at least the
3855 remaining text gets split now.
3857 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3858 file descriptors might be low, and that setting 1000 would always raise it.
3859 It turns out that in some environments, the limit is already over 1000 and
3860 that lowering it causes trouble. So now Exim takes care not to decrease it.
3862 40. When delivering a message, the value of $return_path is set to $sender_
3863 address at the start of routing (routers may change the value). By an
3864 oversight, this default was not being set up when an address was tested by
3865 -bt or -bv, which affected the outcome if any router or filter referred to
3868 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
3869 log only when "message" or "log_message" are set. However, if one of the
3870 conditions was an address verification, or a call to a nested ACL, the
3871 messages generated by the underlying test were being passed through. This
3872 no longer happens. The underlying message is available in $acl_verify_
3873 message for both "message" and "log_message" expansions, so it can be
3874 passed through if needed.
3876 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3877 new expansion $bh_ to give the encoded byte string without charset
3878 translation. Translation happens only if iconv() is available; HAVE_ICONV
3879 indicates this at build time. HEADERS_CHARSET gives the charset to
3880 translate to; headers_charset can change it in the configuration, and
3881 "headers charset" can change it in an individual filter file.
3883 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
3884 that creates RFC 2047 encoded "words" labels them as that charset instead
3885 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3886 expansion operator; (ii) when Exim creates a From: line for a local
3887 message; (iii) when a header line is rewritten to include a "phrase" part.
3889 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3890 buggy, causing it to skip the first lines of messages whose message ID
3891 ended in 'D'. This would not have bitten before Exim release 4.14, because
3892 message IDs were unlikely to end in 'D' before then. The effect was to have
3893 incorrect size information for certain domains.
3895 45. #include "config.h" was missing at the start of the crypt16.c module. This
3896 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3899 46. If there was a timeout during a "random" callout check, Exim treated it as
3900 a failure of the random address, and carried on sending RSET and the real
3901 address. If the delay was just some slowness somewhere, the response to the
3902 original RCPT would be taken as a response to RSET and so on, causing
3903 mayhem of various kinds.
3905 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3906 when I implemented it. It didn't allow for the fact that some option values
3907 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3908 the right test for positive values.
3910 48. Domain names in DNS records are case-independent. Exim always looks them up
3911 in lower case. Some resolvers return domain names in exactly the case they
3912 appear in the zone file, that is, they may contain uppercase letters. Not
3913 all resolvers do this - some return always lower case. Exim was treating a
3914 change of case by a resolver as a change of domain, similar to a widening
3915 of a domain abbreviation. This triggered its re-routing code and so it was
3916 trying to route what was effectively the same domain again. This normally
3917 caused routing to fail (because the router wouldn't handle the domain
3918 twice). Now Exim checks for this case specially, and just changes the
3919 casing of the domain that it ultimately uses when it transmits the message
3922 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3925 50. If a filter generated a file delivery with a non-absolute name (possible if
3926 no home directory exists for the router), the forbid_file option was not
3929 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
3930 the existing equality matching.
3932 52. Exim was using ints instead of ino_t variables in some places where it was
3933 dealing with inode numbers.
3935 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3936 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3937 called TMPDIR, and if it finds it is different, it changes its value.
3939 54. The smtp_printf() function is now made available to local_scan() so
3940 additional output lines can be written before returning. There is also an
3941 smtp_fflush() function to enable the detection of a dropped connection.
3942 The variables smtp_input and smtp_batched_input are exported to
3945 55. Changed the default runtime configuration: the message "Unknown user"
3946 has been removed from the ACL, and instead placed on the localuser router,
3947 using the cannot_route_message feature. This means that any verification
3948 failures that generate their own messages won't get overridden. Similarly,
3949 the "Unrouteable address" message that was in the ACL for unverifiable
3950 relay addresses has also been removed.
3952 56. Added hosts_avoid_esmtp to the smtp transport.
3954 57. The exicyclog script was not checking for the esoteric option
3955 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
3956 will work only if exicyclog is run under the appropriate euid.
3958 58. Following a discussion on the list, the rules by which Exim recognises line
3959 endings on incoming messages have been changed. The -dropcr and drop_cr
3960 options are now no-ops, retained only for backwards compatibility. The
3961 following line terminators are recognized: LF CRLF CR. However, special
3962 processing applies to CR:
3964 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
3965 nor a local message in the state where . is a terminator.
3967 (ii) If a bare CR is encountered in a header line, an extra space is added
3968 after the line terminator so as not to end the header. The reasoning
3969 behind this is that bare CRs in header lines are most likely either
3970 to be mistakes, or people trying to play silly games.
3972 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
3973 was being incorrectly given as 18 bytes larger than it should have been.
3974 This is a VOB (very old bug).
3976 60. This may never have affected anything current, but just in case it has:
3977 When the local host is found other than at the start of a list of hosts,
3978 the local host, those with the same MX, and any that follow, are discarded.
3979 When the list in question was part of a longer list of hosts, the following
3980 hosts (not currently being processed) were also being discarded. This no
3981 longer happens. I'm not sure if this situation could ever has previously
3984 61. Added the "/MX" feature to lists of hosts in the manualroute and query
3987 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
3988 header. This is something that is recommended in a new Internet Draft, and
3989 is something that is documented as being done by Sendmail. There are two
3990 possible values. For messages generated by the autoreply transport, Exim
3993 Auto-Submitted: auto-replied
3995 whereas for all other generated messages (e.g. bounces) it adds
3997 Auto-Submitted: auto-generated
3999 63. The "personal" condition in filters now includes a test for the
4000 Auto-Submitted: header. If it contains the string "auto-" the message it
4001 not considered personal.
4003 64. Added rcpt_include_affixes as a generic transport option.
4005 65. Added queue_only_override (default true).
4007 66. Added the syslog_duplication option.
4009 67. If what should have been the first header line of a message consisted of
4010 a space followed by a colon, Exim was mis-interpreting it as a header line.
4011 It isn't of course - it is syntactically invalid and should therefore be
4012 treated as the start of the message body. The misbehaviour could have
4013 caused a number of strange effects, including loss of data in subsequent
4014 header lines, and spool format errors.
4016 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4017 client host had authenticated. This control can now be exercised by an ACL
4018 for more flexibility.
4020 69. By default, callouts do not happen when testing with -bh. There is now a
4021 variant, -bhc, which does actually run the callout code, including
4022 consulting and updating the callout cache.
4024 70. Added support for saslauthd authentication, courtesy of Alexander
4027 71. If statvfs() failed on the spool or log directories while checking their
4028 size for availability, Exim confusingly gave the error "space shortage".
4029 Furthermore, in debugging mode it crashed with a floating point exception.
4030 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4031 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4032 serious problem, Exim now writes to the main and panic logs when this
4033 happens, with details of the failure. It then refuses to accept the
4034 incoming message, giving the message "spool directory problem" or "log
4035 directory problem" with a 421 code for SMTP messages.
4037 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4038 0, 1, and 2 exist, because some OS complain for execs without them (see
4039 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4040 descriptors. However, the code omitted to check that the open succeeded,
4041 causing mysterious errors if for some reason the permissions on /dev/null
4042 got screwed. Now Exim writes a message to the main and panic logs, and
4043 bombs out if it can't open /dev/null.
4045 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4046 interact so that it is all more flexible. It is supposed to remain
4047 backwards compatible. Also added extra_local_interfaces.
4049 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4050 to bomb out with an assertion failure - to the client this appears as a
4051 connection drop. This problem occurs in the part of the code that was taken
4052 from the Samba project. Fortunately, the assertion is in a very simple
4053 function, so I have fixed this by reproducing the function inline in the
4054 one place where it is called, and arranging for authentication to fail
4055 instead of killing the process with assert().
4057 75. The SPA client code was not working when the server requested OEM rather
4058 than Unicode encoding.
4060 76. Added code to make require_files with a specific uid setting more usable in
4061 the case where statting the file as root fails - usually a non-root-mounted
4062 NFS file system. When this happens and the failure is EACCES, Exim now
4063 forks a subprocess and does the per-uid checking as the relevant uid.
4065 77. Added process_log_path.
4067 78. If log_file_path was not explicitly set, a setting of check_log_space or
4068 check_log_inodes was ignored.
4070 79. If a space check for the spool or log partitions fails, the incident is now
4071 logged. Of course, in the latter case the data may get lost...
4073 80. Added the %p formatting code to string_format() so that it can be used to
4074 print addresses in debug_print(). Adjusted all the address printing in the
4075 debugging in store.c to use %p rather than %d.
4077 81. There was a concern that a line of code in smtp_in.c could overflow a
4078 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4079 initially expressed, the concern was not well-founded, because trailing
4080 spaces are removed early. However, if the trailing spaces were followed by
4081 a NULL, they did not get removed, so the overflow was possible. Two fixes
4084 (a) I re-wrote the offending code in a cleaner fashion.
4085 (b) If an incoming SMTP command contains a NULL character, it is rejected
4088 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4089 runs initgroups(), so that if the Exim user is in any additional groups,
4090 they will be used during message reception.
4096 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob