1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.384 2006/07/27 13:50:43 ph10 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 SC/01 Use a glob alias rather than an array ref in eximstats generated
10 parser. This improves both readability and performance.
12 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
13 Don't display local sender or destination tables in eximstats unless
14 there is data to show.
15 Added average volumes into the eximstats top table text output.
17 SC/03 Collect data on the number of addresses (recipients) as well
18 as the number of messages in eximstats.
20 TF/01 Correct an error in the documentation for the redirect router. Exim
21 does (usually) call initgroups() when daemonizing.
23 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
24 with consistent privilege compared to when running as a daemon.
26 TF/03 Note in the spec that $authenticated_id is not set for local
27 submissions from trusted users.
29 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
30 Thanks to Dean Brooks <dean@iglou.com> for the patch.
32 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
33 by adding some example configuration directives to the default
34 configuration file. A little bit of work is required to uncomment the
35 directives and define how usernames and passwords are checked, but
36 there is now a framework to start from.
38 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
39 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
40 without this. I don't know how relevant this is to other LDAP libraries.
42 PH/02 Add the verb name to the "unknown ACL verb" error.
44 PH/03 Magnus Holmgren's patch for filter_prepend_home.
46 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
48 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
49 directory not expanded when it should be if an expanded home directory
50 was set for the address (which is overridden by the transport).
52 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
55 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
56 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
57 because it is too late at that time, and has no effect.
59 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
60 security issue with \' (bugzilla #107). I could not use the
61 PQescapeStringConn() function, because it needs a PGconn value as one of
64 PH/08 When testing addresses using -bt, indicate those final addresses that
65 are duplicates that would not cause an additional delivery. At least one
66 person was confused, thinking that -bt output corresponded to deliveries.
67 (Suppressing duplicates isn't a good idea as you lose the information
68 about possibly different redirections that led to the duplicates.)
70 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
71 systems where poll() doesn't work, in particular OS X.
73 PH/10 Added more information to debugging output for retry time not reached.
75 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
76 operations in malware.c.
78 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
81 PH/13 If write_rejectlog was set false when logging was sent to syslog with
82 syslog_duplication set false, log lines that would normally be written
83 both the the main log and to the reject log were not written to syslog at
86 PH/14 In the default configuration, change the use of "message" in ACL warn
87 statements to "add_header".
89 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
90 not followed by a command (e.g. "seen endif").
92 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
93 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
96 PH/17 Added an extra condition to the default value of delay_warning_condition
100 { match{$h_precedence:}{(?i)bulk|list|junk} }
101 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} }
104 The Auto-Submitted: header is standardised, whereas I don't think
105 Precedence: ever was.
107 PH/18 Refactored debugging code in route_finduser() to show more information,
108 in particular, the error code if getpwnam() issues one.
110 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
111 This is apparently needed in addition to the PH/07 change above to avoid
112 any possible encoding problems.
114 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
115 but not after initializing Perl.
117 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
118 output them only if debugging. By default they are written stderr,
119 apparently, which is not desirable.
121 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
124 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
127 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
129 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
130 authenticated or an ident call has been made. Suppress the default
131 values for $authenticated_id and $authenticated_sender (but permit -oMai
132 and -oMas) when testing with -bh.
134 PH/24 Re-jigged the order of the tests in the default configuration so that the
135 tests for valid domains and recipients precede the DNS black list and CSA
136 tests, on the grounds that those ones are more expensive.
138 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
139 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
140 This bug exists in every version of Exim that I still have, right back to
143 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
144 However, an attempt to turn on case-sensitivity in a regex key by
145 including (?-i) didn't work because the subject string was already
146 lowercased, and the effects were non-intuitive. It turns out that a
147 one-line patch can be used to allow (?-i) to work as expected.
153 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
154 other effects) broke the use of negated acl sub-conditions.
156 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
159 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
160 "Deny" causes Exim to reject the incoming connection with a 554 error.
161 Unfortunately, if there is a major crisis, such as a disk failure,
162 tcp-wrappers gives "deny", whereas what one would like would be some
163 kind of temporary error. A kludge has been added to help with this.
164 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
165 554 error is used if errno is still zero or contains ENOENT (which occurs
166 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
169 PH/03 Add -lutil to the default FreeBSD LIBS setting.
171 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
172 errors. Otherwise a message that provokes a temporary error (when other
173 messages do not) can cause a whole host to time out.
175 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
176 addresses were routed directly to files or pipes from a redirect router.
177 File deliveries just didn't batch; pipe deliveries might have suffered
180 PH/06 A failure to get a lock for a hints database would erroneously always say
181 "Failed to get write lock", even when it was really a read lock.
183 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
184 of 0600. This has been changed to use the value of the lockfile_mode
185 option (which defaults to 0600).
187 PH/08 Applied small patch from the Sieve maintainer.
189 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
190 folder from quota calculations, a direct delivery into this folder messed
191 up the contents of the maildirsize file. This was because the regex was
192 used only to exclude .Trash (or whatever) when the size of the mailbox
193 was calculated. There was no check that a delivery was happening into an
194 excluded directory. This bug has been fixed by ignoring all quota
195 processing for deliveries into excluded directories.
197 PH/10 Added the maildirfolder_create_regex option to appendfile.
203 PH/01 The code for finding all the local interface addresses on a FreeBSD
204 system running IPv6 was broken. This may well have applied to all BSD
205 systems, as well as to others that have similar system calls. The broken
206 code found IPv4 interfaces correctly, but gave incorrect values for the
207 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
208 that it would not match correctly against @[] and not recognize the IPv6
211 PH/02 The ipliteral router was not recognizing addresses of the form user@
212 [ipv6:....] because it didn't know about the "ipv6:" prefix.
214 PH/03 Added disable_ipv6.
216 PH/04 Changed $reply_address to use the raw form of the headers instead of the
217 decoded form, because it is most often used to construct To: headers
218 lines in autoreplies, and the decoded form may well be syntactically
219 invalid. However, $reply_address has leading white space removed, and all
220 newlines turned into spaces so that the autoreply transport does not
223 PH/05 If group was specified without a user on a router, and no group or user
224 was specified on a transport, the group from the router was ignored.
226 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
227 for visible compile-time settings that can be used to change these
228 numbers, for those that want even more. Backwards compatibility with old
229 spool files has been maintained. However, going back to a previous Exim
230 release will lost any variables that are in spool files.
232 PH/07 Two small changes when running in the test harness: increase delay when
233 passing a TCP/IP connection to a new process, in case the original
234 process has to generate a bounce, and remove special handling of
235 127.0.0.2 (sic), which is no longer necessary.
237 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
238 be the same on different OS.
240 PH/09 Moved a debug statement in filter processing to avoid a race problem when
243 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
244 whether --show-vars was specified or not
246 JJ/02 exipick: Added support for new ACL variable spool format introduced
249 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
250 syntactically invalid From: or Reply-to: line, and a filter used this to
251 generate an autoreply, and therefore failed to obtain an address for the
252 autoreply, Exim could try to deliver to a non-existent relative file
253 name, causing unrelated and misleading errors. What now happens is that
254 it logs this as a hard delivery error, but does not attempt to create a
257 PH/11 The exinext utility has a -C option for testing purposes, but although
258 the given file was scanned by exinext itself; it wasn't being passed on
259 when Exim was called.
261 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
262 an end-of-file indication when reading a command response.
264 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
265 compiled. In many other places in Exim, IPv6 addresses are always
266 recognized, so I have changed this. It also means that IPv4 domain
267 literals of the form [IPV4:n.n.n.n] are now always recognized.
269 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
270 used if the router is not running as root, for example, when verifying at
271 ACL time, or when using -bh. The debugging output from this situation was
272 non-existent - all you got was a failure to exec. I have made two
275 (a) Failures to set uid/gid, the current directory, or a process leader
276 in a subprocess such as that created by queryprogram now generate
277 suitable debugging ouput when -d is set.
279 (b) The queryprogram router detects when it is not running as root,
280 outputs suitable debugging information if -d is set, and then runs
281 the subprocess without attempting to change uid/gid.
283 PH/15 Minor change to Makefile for building test_host (undocumented testing
286 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
287 additional section of a DNS packet that returns MX or SRV records.
288 Instead, it always explicitly searches for A/AAAA records. This avoids
289 major problems that occur when a DNS server includes only records of one
290 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
291 fixed another bug: if SRV records were looked up and the corresponding
292 address records were *not* found in the additional section, the port
293 values from the SRV records were lost.
295 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
296 using the correct key (the original address) when searching the retry
297 rules in order to find which one to use for generating the retry hint.
299 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
300 adding the default one. Similarly, if it contains a Reply-To: header, the
301 errors_reply_to option, if set, is not used.
303 PH/19 When calculating a retry time, Exim used to measure the "time since
304 failure" by looking at the "first failed" field in the retry record. Now
305 it does not use this if it is later than than the arrival time of the
306 message. Instead it uses the arrival time. This makes for better
307 behaviour in cases where some deliveries succeed, thus re-setting the
308 "first failed" field. An example is a quota failure for a huge message
309 when small messages continue to be delivered. Without this change, the
310 "time since failure" will always be short, possible causing more frequent
311 delivery attempts for the huge message than are intended.
312 [Note: This change was subsequently modified - see PH/04 for 4.62.]
314 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
315 $1, $2, $3) because the numerical variables can be reset during some
316 expansion items (e.g. "match"), thereby losing the authentication data.
318 PH/21 Make -bV show the size of off_t variables so that the test suite can
319 decide whether to run tests for quotas > 2G.
321 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
322 mailbox_size, and mailbox_filecount in the appendfile transport. If a
323 filecount value is greater than 2G or if a quota value is greater than 2G
324 on a system where the size of off_t is not greater than 4, a panic error
327 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
328 never match. The debug and -bh output now contains an explicit error
329 message indicating a malformed IPv4 address or mask.
331 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
332 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
335 PH/25 Do not write to syslog when running in the test harness. The only
336 occasion when this arises is a failure to open the main or panic logs
337 (for which there is an explicit test).
339 PH/26 Added the /no_tell option to "control=freeze".
341 PH/27 If a host name lookup failed very early in a connection, for example, if
342 the IP address matched host_lookup and the reverse lookup yielded a name
343 that did not have a forward lookup, an error message of the form "no IP
344 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
345 could be logged. Now it outputs the IP address instead of "NULL".
347 PH/28 An enabling patch from MH: add new function child_open_exim2() which
348 allows the sender and the authenticated sender to be set when
349 submitting a message from within Exim. Since child_open_exim() is
350 documented for local_scan(), the new function should be too.
352 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
353 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
354 results in an empty string is now treated as unset.
356 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
358 PH/31 Added sender_verify_fail logging option.
360 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
361 needed by RFC 822 but not by RFC 2822 was commented out. I have now
362 tidied the source and removed it altogether.
364 PH/33 When a queue run was abandoned because the load average was too high, a
365 log line was always written; now it is written only if the queue_run log
366 selector is set. In addition, the log line for abandonment now contains
367 information about the queue run such as the pid. This is always present
368 in "start" and "stop" lines but was omitted from the "abandon" line.
370 PH/34 Omit spaces between a header name and the colon in the error message that
371 is given when verify = headers_syntax fails (if there are lots of them,
372 the message gets confusing).
374 PH/35 Change the default for dns_check_names_pattern to allow slashes within
375 names, as there are now some PTR records that contain slashes. This check
376 is only to protect against broken name servers that fall over on strange
377 characters, so the fact that it applies to all lookups doesn't matter.
379 PH/36 Now that the new test suite is complete, we can remove some of the
380 special code in Exim that was needed for the old test suite. For example,
381 sorting DNS records because real resolvers return them in an arbitrary
382 order. The new test suite's fake resolver always returns records in the
385 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
386 bounces) except when queue_only is set, to avoid logging races between
387 the different processes.
389 PH/38 Panic-die if .include specifies a non-absolute path.
391 PH/39 A tweak to the "H" retry rule from its user.
393 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
394 a label. They prevented compilation on older perls.
396 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
397 a warning to be raised on newish perls.
399 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
400 on queue. Changes to match documented behaviour of showing count of
401 messages matching specified criteria.
403 PH/40 Changed the default ident timeout from 30s to 5s.
405 PH/41 Added support for the use of login_cap features, on those BSD systems
406 that have them, for controlling the resources used by pipe deliveries.
408 PH/42 The content-scanning code uses fopen() to create files in which to put
409 message data. Previously it was not paying any attention to the mode of
410 the files. Exim runs with umask(0) because the rest of the code creates
411 files with open(), and sets the required mode explicitly. Thus, these
412 files were ending up world-writeable. This was not a big issue, because,
413 being within the spool directory, they were not world-accessible. I have
414 created a function called modefopen, which takes an additional mode
415 argument. It sets umask(777), creates the file, chmods it to the required
416 mode, then resets the umask. All the relevant calls to fopen() in the
417 content scanning code have been changed to use this function.
419 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
420 to 24 hours. This avoids potential overflow problems when processing G
421 and H retry rules. I suspect nobody ever tinkers with this value.
423 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
425 PH/45 When the plaintext authenticator is running as a client, the server's
426 challenges are checked to ensure they are valid base64 strings. By
427 default, the authentication attempt is cancelled if an invalid string is
428 received. Setting client_ignore_invalid_base64 true ignores these errors.
429 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
430 they are received. Thus, the responses can be made to depend on the
431 challenges. If an invalid string is ignored, an empty string is placed in
434 PH/46 Messages that are created by the autoreply transport now contains a
435 References: header, in accordance with RFCs 2822 and 3834.
437 PH/47 Added authenticated_sender_force to the smtp transport.
439 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
441 PH/49 Installed latest patch from the Sieve maintainer.
443 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
444 was also set, the appendfile transport was unnecessarily scanning a
445 directory of message files (e.g. for maildir delivery) to find the count
446 of files (along with the size), even though it did not need this
447 information. It now does the scan only if it needs to find either the
448 size of the count of files.
450 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
452 PH/52 Two bugs concerned with error handling when the smtp transport is
455 (i) Exim was not creating retry information for temporary errors given
456 for individual recipients after the DATA command when the smtp transport
457 was used in LMTP mode. This meant that they could be retried too
458 frequently, and not timed out correctly.
460 (ii) Exim was setting the flag that allows error details to be returned
461 for LMTP errors on RCPT commands, but not for LMTP errors for individual
462 recipients that were returned after the DATA command.
464 PH/53 This is related to PH/52, but is more general: for any failing address,
465 when detailed error information was permitted to be returned to the
466 sender, but the error was temporary, then after the final timeout, only
467 "retry timeout exceeded" was returned. Now it returns the full error as
468 well as "retry timeout exceeded".
470 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
471 do this, and (what is worse) MTAs that accept it.
473 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
474 will now be deprecated.
476 PH/56 New os.c-cygwin from the Cygwin maintainer.
478 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
479 formats (previously only available in exim formats via -bpr, -bpru,
480 and -bpra. Now also available in native and exiqgrep formats)
482 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
483 with very large, slow to parse queues
485 JJ/08 exipick: added ! as generic prefix to negate any criteria format
487 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
489 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
490 responses to authentication challenges, though it was showing the
491 challenges; (ii) I've removed the CR characters from the debug output for
494 PH/58 Allow for the insertion of a newline as well as a space when a string
495 is turned into more than one encoded-word during RFC 2047 encoding. The
496 Sieve code now uses this.
498 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
499 data_4xx, lost_connection, tls_required.
501 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
502 message was being sent as an SMTP response.
504 PH/61 Add -l and -k options to exicyclog.
506 PH/62 When verifying, if an address was redirected to one new address, so that
507 verification continued, and the new address failed or deferred after
508 having set something in $address_data, the value of $address_data was not
509 passed back to the ACL. This was different to the case when no
510 redirection occurred. The value is now passed back in both cases.
512 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
513 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
514 use login_cap.h, so on its own it isn't the distinguishing feature. The
515 new name refers directly to the setclassresources() function.
517 PH/65 Added configuration files for NetBSD3.
519 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
521 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
522 is preferred over IPv4.
524 PH/68 The bounce_return_message and bounce_return_body options were not being
525 honoured for bounces generated during the reception of non-SMTP messages.
526 In particular, this applied to messages rejected by the ACL. This bug has
527 been fixed. However, if bounce_return_message is true and bounce_return_
528 body is false, the headers that are returned for a non-SMTP message
529 include only those that have been read before the error was detected.
530 (In the case of an ACL rejection, they have all been read.)
532 PH/69 The HTML version of the specification is now built in a directory called
533 spec_html instead of spec.html, because the latter looks like a path with
534 a MIME-type, and this confuses some software.
536 PH/70 Catch two compiler warnings in sieve.c.
538 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
539 function verify_get_ident() calls ip_connect() to connect a socket, but
540 if the "connect()" function timed out, ip_connect() used to close the
541 socket. However, verify_get_ident() also closes the socket later, and in
542 between Exim writes to the log, which may get opened at this point. When
543 the socket was closed in ip_connect(), the log could get the same file
544 descriptor number as the socket. This naturally causes chaos. The fix is
545 not to close the socket in ip_connect(); the socket should be closed by
546 the function that creates it. There was only one place in the code where
547 this was missing, in the iplookup router, which I don't think anybody now
548 uses, but I've fixed it anyway.
550 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
551 well as to direct DNS lookups. Otherwise the handling of names in host
552 lists is inconsistent and therefore confusing.
558 PH/01 Two changes to the default runtime configuration:
560 (1) Move the checks for relay_from_hosts and authenticated clients from
561 after to before the (commented out) DNS black list checks.
563 (2) Add control=submission to the relay_from_hosts and authenticated
564 clients checks, on the grounds that messages accepted by these
565 statements are most likely to be submissions.
567 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
569 (1) Generate an error if the third argument for the ${prvs expansion is
572 (2) Treat a missing third argument of ${prvscheck as if it were an empty
575 (3) Reset the variables that are obtained from the first argument of
576 ${prvscheck and used in the second argument before leaving the code,
577 because their memory is reclaimed, so using them afterwards may do
580 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
581 one (it's much easier than Tom thought :-).
583 (5) Because of (4), we can now allow for the use of $prvscheck_result
584 inside the third argument.
586 PH/03 For some reason, the default setting of PATH when running a command from
587 a pipe transport was just "/usr/bin". I have changed it to
590 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
591 anything to be listed in the output from -bV.
593 PH/05 When a filter generated an autoreply, the entire To: header line was
594 quoted in the delivery log line, like this:
596 => >A.N.Other <ano@some.domain> <original@ddress> ...
598 This has been changed so that it extracts the operative address. There
599 may be more than one such address. If so, they are comma-separated, like
602 => >ano@some.domain,ona@other.domain <original@ddress> ...
604 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
605 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
606 not being looked up in the rDNS to get a host name, Exim was showing the
607 IP address twice in Received: lines, even though the IP addresses were
608 identical. For example:
610 Received: from [1.2.3.4] (helo=[1.2.3.4])
612 However, if the real host name was known, it was omitting the HELO data
613 if it matched the actual IP address. This has been tidied up so that it
614 doesn't show the same IP address twice.
616 PH/07 When both +timestamp and +memory debugging was on, the value given by
617 $tod_xxx expansions could be wrong, because the tod_stamp() function was
618 called by the debug printing, thereby overwriting the timestamp buffer.
619 Debugging no longer uses the tod_stamp() function when +timestamp is set.
621 PH/08 When the original message was included in an autoreply transport, it
622 always said "this is a copy of the message, including all the headers",
623 even if body_only or headers_only was set. It now gives an appropriate
626 PH/09 Applied a patch from the Sieve maintainer which:
628 o fixes some comments
629 o adds the (disabled) notify extension core
630 o adds some debug output for the result of if/elsif tests
631 o points to the current vacation draft in the documentation
632 and documents the missing references header update
636 o fixes a bug in processing the envelope test (when testing
637 multiple envelope elements, the last element determinted the
640 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
641 Electronic Mail") by including:
643 Auto-submitted: auto-generated
645 in the messages that it generates (bounce messages and others, such as
646 warnings). In the case of bounce messages for non-SMTP mesages, there was
647 also a typo: it was using "Auto_submitted" (underscore instead of
648 hyphen). Since every message generated by Exim is necessarily in response
649 to another message, thes have all been changed to:
651 Auto-Submitted: auto-replied
653 in accordance with these statements in the RFC:
655 The auto-replied keyword:
657 - SHOULD be used on messages sent in direct response to another
658 message by an automatic process,
660 - MUST NOT be used on manually-generated messages,
662 - MAY be used on Delivery Status Notifications (DSNs) and Message
663 Disposition Notifications (MDNs),
665 - MUST NOT be used on messages generated by automatic or periodic
666 processes, except for messages which are automatic responses to
669 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
670 to the default Received: header definition.
672 PH/12 Added log selector acl_warn_skipped (default on).
674 PH/13 After a successful wildlsearch lookup, discard the values of numeric
675 variables because (a) they are in the wrong storage pool and (b) even if
676 they were copied, it wouldn't work properly because of the caching.
678 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
679 checking when decoding. Apparently there are clients that generate
680 overlong encoded strings. Why am I not surprised?
682 PH/15 If the first argument of "${if match_address" was not empty, but did not
683 contain an "@" character, Exim crashed. Now it writes a panic log message
684 and treats the condition as false.
686 PH/16 In autoreply, treat an empty string for "once" the same as unset.
688 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
689 extension "envelope-auth". The code is finished and in agreement with
690 other implementations, but there is no documentation so far and in fact,
691 nobody wrote the draft yet. This extension is currently #undef'ed, thus
692 not changing the active code.
694 Print executed "if" and "elsif" statements when debugging is used. This
695 helps a great deal to understand what a filter does.
697 Document more things not specified clearly in RFC3028. I had all this
698 sorted out, when out of a sudden new issues came to my mind. Oops."
700 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
703 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
704 canonical form (as documented). However, after a host name lookup from
705 the IP address, check_host() was doing a simple string comparison with
706 addresses acquired from the DNS when checking that the found name did
707 have the original IP as one of its addresses. Since any found IPv6
708 addresses are likely to be in abbreviated form, the comparison could
709 fail. Luckily, there already exists a function for doing the comparison
710 by converting both addresses to binary, so now that is used instead of
713 PH/20 There was another similar case to PH/19, when a complete host name was
714 given in a host list; looking up its IP address could give an abbreviated
715 form, whereas the current host's name might or might not be abbreviated.
716 The same fix has been applied.
722 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
723 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
726 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
727 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
729 PH/03 Typo: missing ".o" in src/pcre/Makefile.
731 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
732 header line, restrict the check to what is listed in RFCs 2369 and 2929.
733 Also, for "Auto-Submitted", treat anything other than "no" as
734 non-personal, in accordance with RFC 3834. (Previously it treated
735 anything starting "auto-" as non-personal.)
737 TF/01 The control=submission/name=... option had a problem with syntax
738 errors if the name included a slash character. The /name= option
739 now slurps the rest of the string, so it can include any characters
740 but it must come last in the list of options (after /sender_retain
743 PH/05 Some modifications to the interface to the fake nameserver for the new
751 TK/01 Added the "success_on_redirect" address verification option. See
752 NewStuff for rationale and an example.
754 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
756 PH/02 Patch to exigrep to allow it to work on syslog lines.
758 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
759 fread() to skip over the body file's header line, because in Cygwin the
760 header line is locked and is inaccessible.
762 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
763 co-exist for some time) to make it clear that it is the Exim ID that is
764 referenced, not the Message-ID: header line.
766 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
767 string_format() function, because snprintf() does not exist on all
770 PH/06 The use of forbid_filter_existstest now also locks out the use of the
771 ${stat: expansion item.
773 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
774 protocol synchronization error", to keep the pedants happy.
776 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
777 well as for IRIX systems, when gcc is being used. See the host.c source
780 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
782 PH/10 Named domain lists were not working if used in a queue_smtp_domains
785 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
786 transport and to the smtp transport in LMTP mode.
788 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
790 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
791 run a filter in a subprocess. This could lead to confusion in subsequent
792 lookups in the parent process. There should also be a search_tidyup() at
793 the end of the subprocess.
795 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
796 only if the host matched helo_try_verify_hosts, which caused the
797 verification to occur when the EHLO/HELO command was issued. The ACL just
798 tested the remembered result. Now, if a previous verification attempt has
799 not happened, "verify = helo" does it there and then.
801 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
803 TK/03 Fix log output including CR from clamd.
805 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
806 exist provoked a memory error which could cause a segfault.
808 PH/15 Installed PCRE 6.2
810 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
812 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
813 of the problem. Specifically, suggested +O2 rather than +O1 for the
816 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
818 PH/20 If a delivery was routed to a non-standard port by means of an SRV
819 record, the port was not correctly logged when the outgoing_port log
820 selector was set (it logged the transort's default port).
822 PH/21 Added support for host-specific ports to manualroute, queryprogram,
823 fallback_hosts, and "hosts" in the smtp transport.
825 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
826 host errors such as "Connection refused".
828 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
829 authentication with radiusclient 0.4.9:
831 - Error returned from rc_read_config was caught wrongly
832 - Username/password not passed on to radius server due to wrong length.
834 The presumption is that some radiusclient API changes for 4.51/PH/17
835 were not taken care of correctly. The code is still untested by me (my
836 Linux distribution still has 0.3.2 of radiusclient), but it was
837 contributed by a Radius user.
839 PH/24 When doing a callout, the value of $domain wasn't set correctly when
840 expanding the "port" option of the smtp transport.
842 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
843 while reading a MIME header. Thanks to Tom Hughes for a patch.
845 PH/24 Include config.h inside local_scan.h so that configuration settings are
848 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
849 that in an ACL for RCPT (for example), you can examine exactly what was
852 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
853 commands, but it was not correctly comparing the address with the actual
854 client host address. Thus, it would show the EHLO address in Received:
855 header lines when this was not necessary.
857 PH/27 Added the % operator to ${eval:}.
859 PH/28 Exim tries to create and chdir to its spool directory when it starts;
860 it should be ignoring failures (because with -C, for example, it has lost
861 privilege). It wasn't ignoring creation failures other than "already
864 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
867 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
868 because an input line was too long, either on its own, or by virtue of
869 too many continuations, the temporary file was not being removed, and the
870 return code was incorrect.
872 PH/31 Missing "BOOL" in function definition in filtertest.c.
874 PH/32 Applied Sieve patches from the maintainer.
876 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
878 PH/33 Added "verify = not_blind".
880 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
881 Local/Makefile (with some defaults set). These are used in built scripts
882 such as exicyclog, but they have never been used in the exim_install
883 script (though there are many overriding facilities there). I have
884 arranged that the exim_install script now takes note of these two
887 PH/35 Installed configuration files for Dragonfly.
889 PH/36 When a locally submitted message by a trusted user did not contain a
890 From: header, and the sender address was obtained from -f or from an SMTP
891 MAIL command, and the trusted user did not use -F to supply a sender
892 name, $originator_name was incorrectly used when constructing a From:
893 header. Furthermore, $originator_name was used for submission mode
894 messages from external hosts without From: headers in a similar way,
895 which is clearly wrong.
897 PH/37 Added control=suppress_local_fixups.
899 PH/38 When log_selector = +received_sender was set, and the addition of the
900 sender made the log line's construction buffer exactly full, or one byte
901 less than full, an overflow happened when the terminating "\n" was
904 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
905 when the result of a list match is failure because a DNS lookup failed.
907 PH/40 RM_COMMAND is now used in the building process.
909 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
910 the "build-* directories that it finds.
912 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
913 address in a domain literal was a prefix of an interface address.
915 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
916 when verifying a sender address, unless rewrite_headers is false.
918 PH/44 Wrote a long comment about why errors_to addresses are verified as
919 recipients, not senders.
921 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
922 the ratelimit ACL was added.
924 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
926 PH/46 Added extra information about PostgreSQL errors to the error string.
928 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
929 avoiding the need to install special zones in a real server. This is
930 backwards compatible; if it can't find the fake resolver, it drops back.
931 Thus, both old and new test suites can be run.
933 TF/02 Added util/ratelimit.pl
935 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
936 clock is set back in time.
938 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
939 Candler <B.Candler@pobox.com>.
941 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
942 OK for addresses that are the result of redirections.
944 PH/48 A number of further additions for the benefit of the new test suite,
945 including a fake gethostbyname() that interfaces to the fake DNS resolver
948 TF/06 The fix for widen_domains has also been applied to qualify_single and
949 search_parents which are the other dnslookup options that can cause
952 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
955 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
957 TF/07 Exim produced the error message "an SRV record indicated no SMTP
958 service" if it encountered an MX record with an empty target hostname.
959 The message is now "an MX or SRV record indicated no SMTP service".
961 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
962 if the DNS of the sending site is misconfigured. This is quite a
963 common situation. This change restores the behaviour of treating a
964 helo verification defer as a failure.
966 PH/51 If self=fail was set on a router, the bounce message did not include the
967 actual error message.
973 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
975 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
976 command itself ended in error, the underlying message about the transport
977 filter timeout was being overwritten with the pipe command error. Now the
978 underlying error message should be appended to the second error message.
980 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
982 PH/02 Reduce the amount of output that "make" produces by default. Full output
983 can still be requested.
985 PH/03 The warning log line about a condition test deferring for a "warn" verb
986 was being output only once per connection, rather than after each
987 occurrence (because it was using the same function as for successful
988 "warn" verbs). This seems wrong, so I have changed it.
990 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
991 it should not have, which might have caused a crash in the right
992 circumstances, but probably never did.
994 PH/04 Installed a modified version of Tony Finch's patch to make submission
995 mode fix the return path as well as the Sender: header line, and to
996 add a /name= option so that you can make the user's friendly name appear
999 TF/03 Added the control = fakedefer ACL modifier.
1001 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1002 Mark Lowes for thorough testing.
1004 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1006 TK/03 Merged latest SRS patch from Miles Wilton.
1008 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1009 with the definition in sysexits.h (which is #included earlier).
1010 Fortunately, Exim does not actually use EX_OK. The code used to try to
1011 preserve the sysexits.h value, by assumimg that macro definitions were
1012 scanned for macro replacements. I have been disabused of this notion,
1013 so now the code just undefines EX_OK before #including unistd.h.
1015 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1016 in the smtp transport. When a block could not be written in a single
1017 write() function, the timeout was being re-applied to each part-write.
1018 This seems wrong - if the receiver was accepting one byte at a time it
1019 would take for ever. The timeout is now adjusted when this happens. It
1020 doesn't have to be particularly precise.
1022 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1023 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1025 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1026 without a domain if the check to <postmaster@domain> fails.
1028 SC/01 Eximstats: added -xls and the ability to specify output files
1029 (patch written by Frank Heydlauf).
1031 SC/02 Eximstats: use FileHandles for outputing results.
1033 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1035 SC/04 Eximstats: fixed display of large numbers with -nvr option
1037 SC/05 Eximstats: fixed merging of reports with empty tables.
1039 SC/06 Eximstats: added the -include_original_destination flag
1041 SC/07 Eximstats: removed tabs and trailing whitespace.
1043 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1045 TK/06 MBOX spool code: Add real "From " MBOX separator line
1046 so the .eml file is really in mbox format (even though
1047 most programs do not really care). Patch from Alex Miller.
1049 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1050 The latter is generated from $received_to and is only set if the
1051 message has one envelope recipient. SA can use these headers,
1052 obviously out-of-the-box. Patch from Alex Miller.
1054 PH/08 The ${def test on a variable was returning false if the variable's
1055 value was "0", contrary to what the specification has always said!
1056 The result should be true unless the variable is empty.
1058 PH/09 The syntax error of a character other than { following "${if
1059 def:variable_name" (after optional whitespace) was not being diagnosed.
1060 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1061 accidental colon was present, for example, could give incorrect results.
1063 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1064 result is used (not including appendfile, where other changes are about
1067 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1068 This involved changing a lot of size variables from int to off_t. It
1069 should work with maildirs and everything.
1071 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1072 spamd dying while we are connected to it.
1074 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1077 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1078 function for BDB 4.3.
1080 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1082 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1083 the sources for PCRE have been re-arranged and more files are now
1086 PH/15 The code I had for printing potentially long long variables in PH/11
1087 above was not the best (it lost precision). The length of off_t variables
1088 is now inspected at build time, and an appropriate printing format (%ld
1089 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1090 to be "long long int" or "long int". This is needed for the internal
1091 formatting function string_vformat().
1093 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1094 the configuration file to be ":syslog", then the script "guesses" where
1095 the logs files are, rather than using the compiled in default. In our
1096 case the guess is not the same as the compiled default, so the script
1097 suddenly stopped working when I started to use syslog. The patch checks
1098 to see if log_file_path is "". If so, it attempts to read it from exim
1099 with no configuration file to get the compiled in version, before it
1100 falls back to the previous guessing code."
1102 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1103 implementing BATV in an Exim configuration. See NewStuff for the gory
1106 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1107 Makefile that are specific to HP-UX.
1109 PH/18 If the "use_postmaster" option was set for a recipient callout together
1110 with the "random" option, the postmaster address was used as the MAIL
1111 FROM address for the random test, but not for the subsequent recipient
1112 test. It is now used for both.
1114 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1115 patch removes a few documentation additions to RFC 3028, because the
1116 latest draft now contains them. It adds the new en;ascii-case comparator
1117 and a new error check for 8bit text in MIME parts. Comparator and
1118 require names are now matched exactly. I enabled the subaddress
1119 extension, but it is not well tested yet (read: it works for me)."
1121 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1122 rework some of the code of TK/09 above to avoid the hardwired use of
1123 "%lld" and "long long". Replaced the call to snprintf() with a call to
1126 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1127 records point to non-existent hosts", "retry timeout exceeded", and
1128 "retry time not reached for any host after a long failure period".
1130 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1131 experimental DomainKeys support:
1133 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1134 (2) On an error such as an illegally used "control", the wrong name for
1135 the control was given.
1137 These problems did NOT occur unless DomainKeys support was compiled.
1139 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1141 PH/24 Added ${if match_ip condition.
1143 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1144 databases so that it will be absolutely obvious if a crash occurs in the
1145 DB library. This is a regular occurrence (often caused by mis-matched
1148 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1149 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1150 picked up on a user's system that detects such things. There doesn't seem
1151 to be a gcc warning option for this - only an attribute that has to be
1152 put on the function's prototype. It seems that in Fedora Core 4 they have
1153 set this on a number of new functions. No doubt there will be more in due
1156 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1157 specify a transport. However, if an address that was verified by such a
1158 router was the subject of a callout, Exim crashed because it tried to
1159 read the rcpt_include_affixes from the non-existent transport. Now it
1160 just assumes that the setting of that option is false. This bug was
1161 introduced by 4.51/PH/31.
1163 PH/28 Changed -d+all to exclude +memory, because that information is very
1164 rarely of interest, but it makes the output a lot bigger. People tend to
1165 do -d+all out of habit.
1167 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1168 code in os-type was giving problems when libc.so lives in lib64, like on
1171 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1172 aren't the modern standard, and it seems that some systems' include files
1173 don't always have them. Exim was already checking for some of the newer
1174 ones like T_AAAA, and defining it itself. I've added checks for all the
1175 record types that Exim uses.
1177 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1178 not automatically generating a new one, as it is supposed to. This
1179 prevented TLS from working. If the file did exist, but contained invalid
1180 data, a new version was generated, as expected. It was only the case of a
1181 non-existent file that was broken.
1183 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1184 with a change in libdomainkeys > 0.64.
1186 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1187 from DNS. If the selector record carries the flag, it now has
1188 precedence over the domain-wide flag.
1190 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1192 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1193 the use of a port name that isn't defined in /etc/services) occurred, the
1194 message was deferred as in a normal delivery, and thus remained on the
1195 spool, instead of being failed because of the mua_wrapper setting. This
1196 is now fixed, and I tidied up some of the mua_wrapper messages at the
1199 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1200 the messages in a hash of arrays rather than using individual hashes.
1201 This is a bit cleaner and results in dramatic memory savings, albeit
1202 at a slight CPU cost.
1204 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1205 as requested by Marc Sherman.
1207 SC/10 Eximstats: added histograms for user specified patterns as requested
1210 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1212 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1213 fopen() in the content-scanning modules that did not already have it.
1219 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1220 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1222 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1224 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1226 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1227 to test Sieve filters that use "vacation".
1229 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1230 that changes the way the GnuTLS parameters are stored in the cache file.
1231 The new format can be generated externally. For backward compatibility,
1232 if the data in the cache doesn't make sense, Exim assumes it has read an
1233 old-format file, and it generates new data and writes a new file. This
1234 means that you can't go back to an older release without removing the
1237 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1238 work if there are any delivery delays because "one_time" forces the
1239 parent to be marked "delivered", so its unseen clone is never tried
1240 again. For this reason, Exim now forbids the simultaneous setting of
1243 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1244 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1245 entry if you want to know the details. The fix, however, neglected to
1246 consider the case when local delivery batching is involved. The test for
1247 "previously delivered" was not happening when checking to see if an
1248 address could be batched with a previous (undelivered) one; under
1249 certain circumstances this could lead to multiple deliveries to the same
1252 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1253 in its include files, and this causes problems building Exim.
1255 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1256 header_syntax) but Exim was just ignoring anything given after a slash.
1257 In particular, this caused confusion with an attempt to use "verify =
1258 reverse_host_lookup/defer_ok". An error is now given when options are
1259 supplied for verify items that do not have them. (Maybe reverse_host_
1260 lookup should have a defer_ok option, but that's a different point.)
1262 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1263 defined by RFC 821) to 2048, because there were problems with some AUTH
1264 commands, and RFC 1869 says the size should be increased for extended
1265 SMTP commands that take arguments.
1267 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1270 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1271 "unknown" error; now it says that the functionality isn't in the binary.
1273 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1274 an address' error message when a string expansion fails (syntax or
1275 whatever). Otherwise the password may appear in the log. Following change
1276 PH/42 below, there is no longer a chance of it appearing in a bounce
1279 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1281 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1282 of its addresses was ever tried. (Bugzilla bug #2.)
1284 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1285 the result incorrectly in the debug output. (It correctly added a newline
1286 to what was transported.)
1288 TF/01 Added $received_time.
1290 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1291 commented out examples of how to interface to a virus scanner and to
1292 SpamAssassin. Also added commented examples of av_scanner and
1293 spamd_address settings.
1295 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1296 and controls are allowed in which ACLs. There were a couple of minor
1297 errors. Some of the entries in the conditions table (which is a table of
1298 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1299 negation of where the condition IS allowed.
1301 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1303 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1304 header file does not have a version number, so I've had to invent a new
1305 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1306 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1307 radiusclient), but it was contributed by a Radius user.
1309 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1310 files or directories, for OpenSSL.
1312 PH/19 When an Exim process that is running as root has to create an Exim log
1313 file, it does so in a subprocess that runs as exim:exim so as to get the
1314 ownership right at creation (otherwise, other Exim processes might see
1315 the file with the wrong ownership). There was no test for failure of this
1316 fork() call, which would lead to the process getting stuck as it waited
1317 for a non-existent subprocess. Forks do occasionally fail when resources
1318 run out. I reviewed all the other calls to fork(); they all seem to check
1321 PH/20 When checking for unexpected SMTP input at connect time (before writing
1322 the banner), Exim was not dealing correctly with a non-positive return
1323 from the read() function. If the client had disconnected by this time,
1324 the result was a log entry for a synchronization error with an empty
1325 string after "input=" when read() returned zero. If read() returned -1
1326 (an event I could not check), uninitialized data bytes were printed.
1327 There were reports of junk text (parts of files, etc) appearing after
1330 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1332 PH/22 Added support for macro redefinition, and (re)definition in between
1333 driver and ACL definitions.
1335 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1336 forgetting to use the resulting value; it was using the unexpanded value.
1338 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1339 hadn't been configured. The fix is from Juergen Kreileder, who
1340 understands it better than I do:
1342 "Here's what I see happening with three configured cyrus_sasl
1343 authenticators configured (plain, login, cram-md5):
1345 On startup auth_cyrus_sasl_init() gets called for each of these.
1346 This means three calls to sasl_listmech() without a specified mech_list.
1347 => SASL tests which mechs of all available mechs actually work
1348 => three warnings about OTP not working
1349 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1351 With the patch, sasl_listmech() also gets called three times. But now
1352 SASL's mech_list option is set to the server_mech specified in the the
1353 authenticator. Or in other words, the answer from sasl_listmech()
1354 gets limited to just the mech you're testing for (which is different
1356 => the return list contains just 'plain' or 'login', 'cram-md5' or
1357 nothing depending on the value of ob->server_mech.
1359 I've just tested the patch: Authentication still works fine,
1360 unavailable mechs specified in the exim configuration are still
1361 caught, and the auth.log warnings about OTP are gone."
1363 PH/25 When debugging is enabled, the contents of the command line are added
1364 to the debugging output, even when log_selector=+arguments is not
1367 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1368 answer is "GNU", and only if the return is "GNU/something" is the answer
1371 PH/27 $acl_verify_message is now set immediately after the failure of a
1372 verification in an ACL, and so is available in subsequent modifiers. In
1373 particular, the message can be preserved by coding like this:
1375 warn !verify = sender
1376 set acl_m0 = $acl_verify_message
1378 Previously, $acl_verify_message was set only while expanding "message"
1379 and "log_message" when a very denied access.
1381 PH/28 Modified OS/os.c-Linux with
1383 -#ifndef OS_LOAD_AVERAGE
1384 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1386 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1387 nomenclature these days.)
1389 PH/29 Installed patch from the Sieve maintainer that adds the options
1390 sieve_useraddress and sieve_subaddress to the redirect router.
1392 PH/30 In these circumstances:
1393 . Two addresses routed to the same list of hosts;
1394 . First host does not offer TLS;
1395 . First host accepts first address;
1396 . First host gives temporary error to second address;
1397 . Second host offers TLS and a TLS session is established;
1398 . Second host accepts second address.
1399 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1400 and peerdn, if requested) that were in fact used only for the second
1403 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1404 attention to any local part prefix or suffix that was matched by the
1405 router that accepted the address. It now behaves in the same way as it
1406 does for delivery: the affixes are removed from the local part unless
1407 rcpt_include_affixes is set on the transport.
1409 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1410 timeout during the DATA phase of an incoming message.
1412 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1413 applied a patch sanctioned by the Sieve maintainer.
1415 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1416 the uid or gid is negative. A case of a negative gid caused this to be
1417 noticed. The fix allows for either to be negative.
1419 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1420 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1423 PH/36 The change PH/12 above was broken. Fixed it.
1425 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1426 the grounds that routing the same address twice would always produce the
1427 same answer. This might have been true once, but it is certainly no
1428 longer true now. Routing a child address may depend on the previous
1429 routing that produced that child. Some complicated redirection strategies
1430 went wrong when messages had multiple recipients, and made Exim's
1431 behaviour dependent on the order in which the addresses were given.
1433 I have moved the duplicate checking until after the routing is complete.
1434 Exim scans the addresses that are assigned to local and remote
1435 transports, and removes any duplicates. This means that more work will be
1436 done, as duplicates will always all be routed, but duplicates are
1437 presumably rare, so I don't expect this is of any significance.
1439 For deliveries to pipes, files, and autoreplies, the duplicate checking
1440 still happens during the routing process, since they are not going to be
1443 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1444 It corrects a timeout issue with spamd. This is Ian's comment: "The
1445 background is that sometimes spamd either never reads data from a
1446 connection it has accepted, or it never writes response data. The exiscan
1447 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1448 blindly assumes that writes won't block so it may never time out."
1450 PH/39 Allow G after quota size as well as K and M.
1452 PH/40 The value set for $authenticated_id in an authenticator may not contain
1453 binary zeroes or newlines because the value is written to log lines and
1454 to spool files. There was no check on this. Now the value is run through
1455 the string_printing() function so that such characters are converted to
1456 printable escape sequences.
1458 PH/41 $message_linecount is a new variable that contains the total number of
1459 lines in the message. Compare $body_linecount, which is the count for the
1462 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1463 bounce and delay warning messages, except in certain special cases, which
1466 (a) An SMTP error message from a remote host;
1467 (b) A message specified in a :fail: redirection;
1468 (c) A message specified in a "fail" command in a system filter;
1469 (d) A message specified in a FAIL return from the queryprogram router;
1470 (e) A message specified by the cannot_route_message router option.
1472 In these cases only, Exim does include the error details in bounce and
1473 warning messages. There are also a few cases where bland messages such
1474 as "unrouteable address" or "local delivery error" are given.
1476 PH/43 $value is now also set for the "else" part of a ${run expansion.
1478 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1479 being worked on, but at least Exim now implements the latest version to
1482 PH/45 In a pipe transport, although a timeout while waiting for the pipe
1483 process to complete was treated as a delivery failure, a timeout while
1484 writing the message to the pipe was logged, but erroneously treated as a
1485 successful delivery. Such timeouts include transport filter timeouts. For
1486 consistency with the overall process timeout, these timeouts are now
1487 treated as errors, giving rise to delivery failures by default. However,
1488 there is now a new Boolean option for the pipe transport called
1489 timeout_defer, which, if set TRUE, converts the failures into defers for
1490 both kinds of timeout. A transport filter timeout is now identified in
1493 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1494 systems where "make" and "gmake" are different, calling "gmake" at top
1495 level broke things. I've arranged for the value of $(MAKE) to be passed
1496 from the Makefile to this script so that it can call the same version of
1500 A note about Exim versions 4.44 and 4.50
1501 ----------------------------------------
1503 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1504 changes of various kinds. As a consequence, a big documentation update was
1505 needed. This delayed the release for rather longer than seemed good, especially
1506 in the light of a couple of (minor) security issues. Therefore, the changes
1507 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1508 release. So 4.44 and 4.50 are in effect two different branches that both start
1511 I have left the 4.50 change log unchanged; it contains all the changes since
1512 4.43. The change log for 4.44 is below; many of its items are identical to
1513 those for 4.50. This seems to be the most sensible way to preserve the
1514 historical information.
1520 1. Minor wording change to the doc/README.SIEVE file.
1522 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1523 computation of the current number of files was incorrect.
1525 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1526 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1529 4. Give more explanation in the error message when the command for a transport
1530 filter fails to execute.
1532 5. There are several places where Exim runs a non-Exim command in a
1533 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1534 was being done only for the command run by the queryprogram router. It is
1535 now done for all such subprocesses. The other cases are: ${run, transport
1536 filters, and the commands run by the lmtp and pipe transports.
1538 6. Added CONFIGURE_GROUP build-time option.
1540 7. Some older OS have a limit of 256 on the maximum number of file
1541 descriptors. Exim was using setrlimit() to set 1000 as a large value
1542 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1543 systems. I've change it so that if it can't get 1000, it tries for 256.
1545 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1546 was an oversight, and furthermore, ever since the addition of extra
1547 controls (e.g. 4.43/32), the checks on when to allow different forms of
1548 "control" were broken. There should now be diagnostics for all cases when a
1549 control that does not make sense is encountered.
1551 9. Added the /retain_sender option to "control=submission".
1553 10. $recipients is now available in the predata ACL (oversight).
1555 11. Tidy the search cache before the fork to do a delivery from a message
1556 received from the command line. Otherwise the child will trigger a lookup
1557 failure and thereby defer the delivery if it tries to use (for example) a
1558 cached ldap connection that the parent has called unbind on.
1560 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1561 of $address_data from the recipient verification was clobbered by the
1562 sender verification.
1564 13. The value of address_data from a sender verification is now available in
1565 $sender_address_data in subsequent conditions in the ACL statement.
1567 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1569 15. Added a new option "connect=<time>" to callout options, to set a different
1572 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1573 was its contents. (It was OK if the option was not defined at all.)
1575 17. A "Completed" log line is now written for messages that are removed from
1576 the spool by the -Mrm option.
1578 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1579 information about exactly what failed.
1581 19. Added -dd to debug only the daemon process.
1583 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1584 handles timeouts, both on the server side and network timeouts. Renamed the
1585 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1587 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1588 It is now set to "smtps".
1590 22. $host_address is now set to the target address during the checking of
1591 ignore_target_hosts.
1593 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1594 being passed; this would have caused $sender_host_name to have been used if
1595 matching the list had actually called for a host name (not very likely,
1596 since this list is usually IP addresses). A host name is now passed as
1599 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1600 code that specifies a non-restarting handler (typically sigaction() in
1601 modern systems) in an attempt to fix a rare and obscure crash bug.
1603 25. Narrowed the window for a race in the daemon that could cause it to ignore
1604 SIGCHLD signals. This is not a major problem, because they are used only to
1605 wake it up if nothing else does.
1607 26. A malformed maildirsize file could cause Exim to calculate negative values
1608 for the mailbox size or file count. Odd effects could occur as a result.
1609 The maildirsize information is now recalculated if the size or filecount
1612 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1613 support for a long time. Removed HAVE_SYS_VFS_H.
1615 28. Installed the latest version of exipick from John Jetmore.
1617 29. In an address list, if the pattern was not a regular expression, an empty
1618 subject address (from a bounce message) matched only if the pattern was an
1619 empty string. Non-empty patterns were not even tested. This was the wrong
1620 because it is perfectly reasonable to use an empty address as part of a
1621 database query. An empty address is now tested by patterns that are
1622 lookups. However, all the other forms of pattern expect the subject to
1623 contain a local part and a domain, and therefore, for them, an empty
1624 address still always fails if the pattern is not itself empty.
1626 30. Exim went into a mad DNS loop when attempting to do a callout where the
1627 host was specified on an smtp transport, and looking it up yielded more
1628 than one IP address.
1630 31. Re-factored the code for checking spool and log partition space into a
1631 function that finds that data and another that does the check. The former
1632 is then used to implement four new variables: $spool_space, $log_space,
1633 $spool_inodes, and $log_inodes.
1635 32. The RFC2047 encoding function was originally intended for short strings
1636 such as real names; it was not keeping to the 75-character limit for
1637 encoded words that the RFC imposes. It now respects the limit, and
1638 generates multiple encoded words if necessary. To be on the safe side, I
1639 have increased the buffer size for the ${rfc2047: expansion operator from
1642 33. It is now permitted to omit both strings after an "if" condition; if the
1643 condition is true, the result is "true". As before, when the second string
1644 is omitted, a false condition yields an empty string. This makes it less
1645 cumbersome to write custom ACL and router conditions.
1647 34. Failure to deliver a bounce message always caused it to be frozen, even if
1648 there was an errors_to setting on the router. The errors_to setting is now
1651 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1652 canonical form (fully expanded) before being placed in
1653 $sender_host_address.
1655 36. The table in the code that translates DNS record types into text (T_A to
1656 "A" for instance) was missing entries for NS and CNAME. It is just possible
1657 that this could have caused confusion if both these types were looked up
1658 for the same domain, because the text type is used as part of Exim's
1659 per-process caching. But the chance of anyone hitting this buglet seems
1662 37. The dnsdb lookup has been extended in a number of ways.
1664 (1) There is a new type, "zns", which walks up the domain tree until it
1665 finds some nameserver records. It should be used with care.
1667 (2) There is a new type, "mxh", which is like "mx" except that it returns
1668 just the host names, not the priorities.
1670 (3) It is now possible to give a list of domains (or IP addresses) to be
1671 looked up. The behaviour when one of the lookups defers can be
1672 controlled by a keyword.
1674 (4) It is now possible to specify the separator character for use when
1675 multiple records are returned.
1677 38. The dnslists ACL condition has been extended: it is now possible to supply
1678 a list of IP addresses and/or domains to be looked up in a particular DNS
1681 39. Added log_selector=+queue_time_overall.
1683 40. When running the queue in the test harness, wait just a tad after forking a
1684 delivery process, to get repeatability of debugging output.
1686 41. Include certificate and key file names in error message when GnuTLS fails
1687 to set them up, because the GnuTLS error message doesn't include the name
1688 of the failing file when there is a problem reading it.
1690 42. Allow both -bf and -bF in the same test run.
1692 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1694 44. The "Exiscan patch" is now merged into the mainline Exim source.
1696 45. Sometimes the final signoff response after QUIT could fail to get
1697 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1698 < 0 before doing a fflush(). This bug looks as though it goes back to the
1699 introduction of TLS in release 3.20, but "sometimes" must have been rare
1700 because the tests only now provoked it.
1702 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1703 (this can affect the format of dates).
1705 47. exim_tidydb, when checking for the continued existence of a message for
1706 which it has found a message-specific retry record, was not finding
1707 messages that were in split spool directories. Consequently, it was
1708 deleting retry records that should have stayed in existence.
1710 48. Steve fixed some bugs in eximstats.
1712 49. The SPA authentication driver was not abandoning authentication and moving
1713 on to the next authenticator when an expansion was forced to fail,
1714 contradicting the general specification for all authenticators. Instead it
1715 was generating a temporary error. It now behaves as specified.
1717 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1718 (the order specifies the preference for clients). The order is now AES256,
1719 AES128, 3DES, ARCFOUR128.
1721 51. Small patch to Sieve code - explicitly set From: when generating an
1724 52. Exim crashed if a remote delivery caused a very long error message to be
1725 recorded - for instance if somebody sent an entire SpamAssassin report back
1726 as a large number of 550 error lines. This bug was coincidentally fixed by
1727 increasing the size of one of Exim's internal buffers (big_buffer) that
1728 happened as part of the Exiscan merge. However, to be on the safe side, I
1729 have made the code more robust (and fixed the comments that describe what
1732 53. Now that there can be additional text after "Completed" in log lines (if
1733 the queue_time_overall log selector is set), a one-byte patch to exigrep
1734 was needed to allow it to recognize "Completed" as not the last thing in
1737 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1738 patch that reportedly fixes this has been added. I am not expert enough to
1739 create a test for it. This is what the patch creator wrote:
1741 "I found a little strange behaviour of ldap code when working with
1742 Windows 2003 AD Domain, where users was placed in more than one
1743 Organization Units. When I tried to give exim partial DN, the exit code
1744 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1745 But simultaneously result of request was absolutely normal ldap result,
1746 so I produce this patch..."
1748 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1749 REFERENCE, so I have modified the code to exclude the patch when that macro
1752 55. Some experimental protocols are using DNS PTR records for new purposes. The
1753 keys for these records are domain names, not reversed IP addresses. The
1754 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1755 leaves it alone. Component reversal etc. now happens only for IP addresses.
1758 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1760 57. Double the size of the debug message buffer (to 2048) so that more of very
1761 long debug lines gets shown.
1763 58. The exicyclog utility now does better if the number of log files to keep
1764 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1766 59. Two changes related to the smtp_active_hostname option:
1768 (1) $smtp_active_hostname is now available as a variable.
1769 (2) The default for smtp_banner uses $smtp_active_hostname instead
1770 of $primary_hostname.
1772 60. The host_aton() function is supposed to be passed a string that is known
1773 to be a valid IP address. However, in the case of IPv6 addresses, it was
1774 not checking this. This is a hostage to fortune. Exim now panics and dies
1775 if the condition is not met. A case was found where this could be provoked
1776 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1777 components; fortuitously, this particular loophole had already been fixed
1778 by change 4.50/55 above.
1780 If there are any other similar loopholes, the new check in host_aton()
1781 itself should stop them being exploited. The report I received stated that
1782 data on the command line could provoke the exploit when Exim was running as
1783 exim, but did not say which command line option was involved. All I could
1784 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1785 running as the user.
1788 61. There was a buffer overflow vulnerability in the SPA authentication code
1789 (which came originally from the Samba project). I have added a test to the
1790 spa_base64_to_bits() function which I hope fixes it.
1793 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1794 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1796 63. The daemon start-up calls getloadavg() while still root for those OS that
1797 need the first call to be done as root, but it missed one case: when
1798 deliver_queue_load_max is set with deliver_drop_privilege. This is
1799 necessary for the benefit of the queue runner, because there is no re-exec
1800 when deliver_drop_privilege is set.
1802 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1803 This has been fixed.
1805 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1806 was in use, was not putting the data itself into the right store pool;
1807 consequently, it could be overwritten for a subsequent message in the same
1808 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1811 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1813 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1814 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1815 the function were treating the return as a boolean value, which happened to
1816 work because 0=false and not-0=true, but is not correct code.
1818 68. The host_aton() function was not handling scoped IPv6 addresses (those
1819 with, for example, "%eth0" on the end) correctly.
1821 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1822 negated items (that is, ~something) in unsigned ints. Some compilers
1823 apparently mutter when there is no cast.
1825 70. If an address verification called from an ACL failed, and did not produce a
1826 user-specific message (i.e. there was only a "system" message), nothing was
1827 put in $acl_verify_message. In this situation, it now puts the system
1830 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1831 session; change 4.31/43 added the unwanted input to the log line - except
1832 that it did not do this in the start of session case. It now does.
1834 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1835 This is wrong and can cause the other end to generate a synchronization
1836 error if it is another Exim or anything else that does the synchronization
1837 check. A QUIT command is no longer sent after a timeout.
1839 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1840 during host lookups.
1842 74. The defer_ok option of callout verification was not working if it was used
1843 when verifying addresses in header lines, that is, for this case:
1845 verify = header_sender/callout=defer_ok
1847 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1848 those file descriptors could be used for SMTP connections. If anything
1849 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1850 could be sent to the SMTP client, causing chaos. The daemon now opens
1851 stdin, stdout, and stderr to /dev/null when it puts itself into the
1854 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1855 log by default. The user can override this with suitable Perl magic.
1857 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1858 the log message when discard triggers, was not working for the DATA ACL or
1859 for the non-SMTP ACL.
1861 78. Error message wording change in sieve.c.
1863 79. If smtp_accept_max_per_host was set, the number of connections could be
1864 restricted to fewer than expected, because the daemon was trying to set up
1865 a new connection before checking whether the processes handling previous
1866 connections had finished. The check for completed processes is now done
1867 earlier. On busy systems, this bug wouldn't be noticed because something
1868 else would have woken the daemon, and it would have reaped the completed
1871 80. If a message was submitted locally by a user whose login name contained one
1872 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1873 It caused a spool format error. I have fixed the spool reading code. A
1874 related problem was that the "from" clause in the Received: line became
1875 illegal because of the space(s). It is now covered by ${quote_local_part.
1877 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
1880 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
1882 83. Patch from Sieve maintainer to fix unterminated string problem in
1883 "vacation" handling.
1885 84. Some minor changes to the Linux configuration files to help with other
1886 OS variants using glibc.
1888 85. One more patch for Sieve to update vacation handling to latest spec.
1891 ----------------------------------------------------
1892 See the note above about the 4.44 and 4.50 releases.
1893 ----------------------------------------------------
1899 1. Change 4.43/35 introduced a bug that caused file counts to be
1900 incorrectly computed when quota_filecount was set in an appendfile
1903 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1904 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1907 3. Give more explanation in the error message when the command for a transport
1908 filter fails to execute.
1910 4. There are several places where Exim runs a non-Exim command in a
1911 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1912 was being done only for the command run by the queryprogram router. It is
1913 now done for all such subprocesses. The other cases are: ${run, transport
1914 filters, and the commands run by the lmtp and pipe transports.
1916 5. Some older OS have a limit of 256 on the maximum number of file
1917 descriptors. Exim was using setrlimit() to set 1000 as a large value
1918 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1919 systems. I've change it so that if it can't get 1000, it tries for 256.
1921 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1922 was an oversight, and furthermore, ever since the addition of extra
1923 controls (e.g. 4.43/32), the checks on when to allow different forms of
1924 "control" were broken. There should now be diagnostics for all cases when a
1925 control that does not make sense is encountered.
1927 7. $recipients is now available in the predata ACL (oversight).
1929 8. Tidy the search cache before the fork to do a delivery from a message
1930 received from the command line. Otherwise the child will trigger a lookup
1931 failure and thereby defer the delivery if it tries to use (for example) a
1932 cached ldap connection that the parent has called unbind on.
1934 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1935 of $address_data from the recipient verification was clobbered by the
1936 sender verification.
1938 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1939 was its contents. (It was OK if the option was not defined at all.)
1941 11. A "Completed" log line is now written for messages that are removed from
1942 the spool by the -Mrm option.
1944 12. $host_address is now set to the target address during the checking of
1945 ignore_target_hosts.
1947 13. When checking ignore_target_hosts for an ipliteral router, no host name was
1948 being passed; this would have caused $sender_host_name to have been used if
1949 matching the list had actually called for a host name (not very likely,
1950 since this list is usually IP addresses). A host name is now passed as
1953 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1954 code that specifies a non-restarting handler (typically sigaction() in
1955 modern systems) in an attempt to fix a rare and obscure crash bug.
1957 15. Narrowed the window for a race in the daemon that could cause it to ignore
1958 SIGCHLD signals. This is not a major problem, because they are used only to
1959 wake it up if nothing else does.
1961 16. A malformed maildirsize file could cause Exim to calculate negative values
1962 for the mailbox size or file count. Odd effects could occur as a result.
1963 The maildirsize information is now recalculated if the size or filecount
1966 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1967 support for a long time. Removed HAVE_SYS_VFS_H.
1969 18. Updated exipick to current release from John Jetmore.
1971 19. Allow an empty sender to be matched against a lookup in an address list.
1972 Previously the only cases considered were a regular expression, or an
1975 20. Exim went into a mad DNS lookup loop when doing a callout where the
1976 host was specified on the transport, if the DNS lookup yielded more than
1979 21. The RFC2047 encoding function was originally intended for short strings
1980 such as real names; it was not keeping to the 75-character limit for
1981 encoded words that the RFC imposes. It now respects the limit, and
1982 generates multiple encoded words if necessary. To be on the safe side, I
1983 have increased the buffer size for the ${rfc2047: expansion operator from
1986 22. Failure to deliver a bounce message always caused it to be frozen, even if
1987 there was an errors_to setting on the router. The errors_to setting is now
1990 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1991 canonical form (fully expanded) before being placed in
1992 $sender_host_address.
1994 24. Updated eximstats to version 1.33
1996 25. Include certificate and key file names in error message when GnuTLS fails
1997 to set them up, because the GnuTLS error message doesn't include the name
1998 of the failing file when there is a problem reading it.
2000 26. Expand error message when OpenSSL has problems setting up cert/key files.
2003 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2004 (this can affect the format of dates).
2006 28. exim_tidydb, when checking for the continued existence of a message for
2007 which it has found a message-specific retry record, was not finding
2008 messages that were in split spool directories. Consequently, it was
2009 deleting retry records that should have stayed in existence.
2011 29. eximstats updated to version 1.35
2012 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2013 1.35 - bugfix such that pie charts by volume are generated correctly
2015 30. The SPA authentication driver was not abandoning authentication and moving
2016 on to the next authenticator when an expansion was forced to fail,
2017 contradicting the general specification for all authenticators. Instead it
2018 was generating a temporary error. It now behaves as specified.
2020 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2021 (the order specifies the preference for clients). The order is now AES256,
2022 AES128, 3DES, ARCFOUR128.
2024 31. Small patch to Sieve code - explicitly set From: when generating an
2027 32. Exim crashed if a remote delivery caused a very long error message to be
2028 recorded - for instance if somebody sent an entire SpamAssassin report back
2029 as a large number of 550 error lines. This bug was coincidentally fixed by
2030 increasing the size of one of Exim's internal buffers (big_buffer) that
2031 happened as part of the Exiscan merge. However, to be on the safe side, I
2032 have made the code more robust (and fixed the comments that describe what
2035 33. Some experimental protocols are using DNS PTR records for new purposes. The
2036 keys for these records are domain names, not reversed IP addresses. The
2037 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2038 leaves it alone. Component reversal etc. now happens only for IP addresses.
2041 34. The host_aton() function is supposed to be passed a string that is known
2042 to be a valid IP address. However, in the case of IPv6 addresses, it was
2043 not checking this. This is a hostage to fortune. Exim now panics and dies
2044 if the condition is not met. A case was found where this could be provoked
2045 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2046 components; fortuitously, this particular loophole had already been fixed
2047 by change 4.50/55 or 4.44/33 above.
2049 If there are any other similar loopholes, the new check in host_aton()
2050 itself should stop them being exploited. The report I received stated that
2051 data on the command line could provoke the exploit when Exim was running as
2052 exim, but did not say which command line option was involved. All I could
2053 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2054 running as the user.
2057 35. There was a buffer overflow vulnerability in the SPA authentication code
2058 (which came originally from the Samba project). I have added a test to the
2059 spa_base64_to_bits() function which I hope fixes it.
2062 36. The daemon start-up calls getloadavg() while still root for those OS that
2063 need the first call to be done as root, but it missed one case: when
2064 deliver_queue_load_max is set with deliver_drop_privilege. This is
2065 necessary for the benefit of the queue runner, because there is no re-exec
2066 when deliver_drop_privilege is set.
2068 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2069 was in use, was not putting the data itself into the right store pool;
2070 consequently, it could be overwritten for a subsequent message in the same
2071 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2074 38. Sometimes the final signoff response after QUIT could fail to get
2075 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2076 < 0 before doing a fflush(). This bug looks as though it goes back to the
2077 introduction of TLS in release 3.20, but "sometimes" must have been rare
2078 because the tests only now provoked it.
2084 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2085 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2086 Now it returns an integer. A number of calls were still expecting a T/F
2087 return. Fortuitously, in all cases, the tests worked in OK situations,
2088 which is the norm. However, things would have gone wrong on any write
2089 failures on the smtp file descriptor. This function is used when sending
2090 messages over SMTP and also when doing verify callouts.
2092 2. When Exim is called to do synchronous delivery of a locally submitted
2093 message (the -odf or -odi options), it no longer closes stderr before doing
2096 3. Implemented the mua_wrapper option.
2098 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2100 5. Implemented the functions header_remove(), header_testname(),
2101 header_add_at_position(), and receive_remove_recipient(), and exported them
2104 6. If an ACL "warn" statement specified the addition of headers, Exim already
2105 inserted X-ACL-Warn: at the start if there was no header name. However, it
2106 was not making this test for the second and subsequent header lines if
2107 there were newlines in the string. This meant that an invalid header could
2108 be inserted if Exim was badly configured.
2110 7. Allow an ACL "warn" statement to add header lines at the start or after all
2111 the Received: headers, as well as at the end.
2113 8. Added the rcpt_4xx retry error code.
2115 9. Added postmaster_mailfrom=xxx to callout verification option.
2117 10. Added mailfrom=xxxx to the callout verification option, for verify=
2120 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2121 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2123 12. Inserted some casts to stop certain compilers warning when using pointer
2124 differences as field lengths or precisions in printf-type calls (mostly
2125 affecting debugging statements).
2127 13. Added optional readline() support for -be (dynamically loaded).
2129 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2130 same clock tick as a message's arrival, so that its received time was the
2131 same as the "first fail" time on the retry record, and that message
2132 remained on the queue past the ultimate address timeout, every queue runner
2133 would try a delivery (because it was past the ultimate address timeout) but
2134 after another failure, the ultimate address timeout, which should have then
2135 bounced the address, did not kick in. This was a "< instead of <=" error;
2136 in most cases the first failure would have been in the next clock tick
2137 after the received time, and all would be well.
2139 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2140 being recognized when the domain list was tested by the match_domain
2141 condition in an expansion string.
2143 16. Added the ${str2b64: operator.
2145 17. Exim was always calling setrlimit() to set a large limit for the number of
2146 processes, without checking whether the existing limit was already
2147 adequate. (It did check for the limit on file descriptors.) Furthermore,
2148 errors from getrlimit() and setrlimit() were being ignored. Now they are
2149 logged to the main and panic logs, but Exim does carry on, to try to do its
2150 job under whatever limits there are.
2152 18. Imported PCRE 5.0.
2154 19. Trivial typo in log message " temporarily refused connection" (the leading
2157 20. If the log selector return_path_on_delivery was set and an address was
2158 redirected to /dev/null, the delivery process crashed because it assumed
2159 that a return path would always be set for a "successful" delivery. In this
2160 case, the whole delivery is bypassed as an optimization, and therefore no
2163 21. Internal re-arrangement: the function for sending a challenge and reading
2164 a response while authentication was assuming a zero-terminated challenge
2165 string. It's now changed to take a pointer and a length, to allow for
2166 binary data in such strings.
2168 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2170 23. Exim was not respecting finduser_retries when seeking the login of the
2171 uid under which it was called; it was always trying 10 times. (The default
2172 setting of finduser_retries is zero.) Also, it was sleeping after the final
2173 failure, which is pointless.
2175 24. Implemented tls_on_connect_ports.
2177 25. Implemented acl_smtp_predata.
2179 26. If the domain in control=submission is set empty, Exim assumes that the
2180 authenticated id is a complete email address when it generates From: or
2181 Sender: header lines.
2183 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2184 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2185 chown and chgrp in /bin and hostname in /usr/bin.
2187 28. Exim was keeping the "process log" file open after each use, just as it
2188 does for the main log. This opens the possibility of it remaining open for
2189 long periods when the USR1 signal hits a daemon. Occasional processlog
2190 errors were reported, that could have been caused by this. Anyway, it seems
2191 much more sensible not to leave this file open at all, so that is what now
2194 29. The long-running daemon process does not normally write to the log once it
2195 has entered its main loop, and it closes the log before doing so. This is
2196 so that log files can straightforwardly be renamed and moved. However,
2197 there are a couple of unusual error situations where the daemon does write
2198 log entries, and I had neglected to close the log afterwards.
2200 30. The text of an SMTP error response that was received during a remote
2201 delivery was being truncated at 512 bytes. This is too short for some of
2202 the long messages that one sometimes sees. I've increased the limit to
2205 31. It is now possible to make retry rules that apply only when a message has a
2206 specific sender, in particular, an empty sender.
2208 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2209 it possible to be selective about when SMTP synchronization is enforced.
2211 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2213 32. Implemented hosts_connection_nolog.
2215 33. Added an ACL for QUIT.
2217 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2220 35. Added mailbox_size and mailbox_filecount to appendfile.
2222 36. Added control = no_multiline_responses to ACLs.
2224 37. There was a bug in the logic of the code that waits for the clock to tick
2225 in the case where the clock went backwards by a substantial amount such
2226 that the microsecond fraction of "now" was more than the microsecond
2227 fraction of "then" (but the whole seconds number was less).
2229 38. Added support for the libradius Radius client library this is found on
2230 FreeBSD (previously only the radiusclient library was supported).
2236 1. When certain lookups returned multiple values in the form name=value, the
2237 quoting of the values was not always being done properly. Specifically:
2238 (a) If the value started with a double quote, but contained no whitespace,
2240 (b) If the value contained whitespace other than a space character (i.e.
2241 tabs or newlines or carriage returns) it was not quoted.
2242 This fix has been applied to the mysql and pgsql lookups by writing a
2243 separate quoting function and calling it from the lookup code. The fix
2244 should probably also be applied to nisplus, ibase and oracle lookups, but
2245 since I cannot test any of those, I have not disturbed their existing code.
2247 2. A hit in the callout cache for a specific address caused a log line with no
2248 reason for rejecting RCPT. Now it says "Previous (cached) callout
2249 verification failure".
2251 3. There was an off-by-one bug in the queryprogram router. An over-long
2252 return line was truncated at 256 instead of 255 characters, thereby
2253 overflowing its buffer with the terminating zero. As well as fixing this, I
2254 have increased the buffer size to 1024 (and made a note to document this).
2256 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2257 when Exim is waiting for an SMTP response from a remote server, Exim
2258 restarts its select() call on the socket, thereby resetting its timeout.
2259 This is not a problem when such interrupts are rare. Somebody set up a cron
2260 job to run exiwhat every 2 minutes, which is less than the normal select()
2261 timeout (5 or 10 minutes). This meant that the select() timeout never
2262 kicked in because it was always reset. I have fixed this by comparing the
2263 time when an interrupt arrives with the time at the start of the first call
2264 to select(). If more time than the timeout has elapsed, the interrupt is
2265 treated as a timeout.
2267 5. Some internal re-factoring in preparation for the addition of Sieve
2268 extensions (by MH). In particular, the "personal" test is moved to a
2269 separate function, and given an option for scanning Cc: and Bcc: (which is
2270 not set for Exim filters).
2272 6. When Exim created an email address using the login of the caller as the
2273 local part (e.g. when creating a From: or Sender: header line), it was not
2274 quoting the local part when it contained special characters such as @.
2276 7. Installed new OpenBSD configuration files.
2278 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2279 try to make them clearer.
2281 9. Callout options, other than the timeout value, were being ignored when
2282 verifying sender addresses in header lines. For example, when using
2284 verify = header_sender/callout=no_cache
2286 the cache was (incorrectly) being used.
2288 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2289 only the Cygwin environment.
2291 11. When return_path_on_delivery was set as a log selector, if different remote
2292 addresses in the same message used different return paths and parallel
2293 remote delivery occurred, the wrong values would sometimes be logged.
2294 (Whenever a remote delivery process finished, the return path value from
2295 the most recently started remote delivery process was logged.)
2297 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2298 lines when AUTH and/or TLS are in use. This is the "received protocol"
2299 field. Exim used to use "asmtp" for authenticated SMTP, without any
2300 indication (in the protocol name) for TLS use. Now it follows the RFC and
2301 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2302 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2303 names appear in log lines as well as in Received: header lines.
2305 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2306 extensions, and comparison tests, and to fix some bugs.
2308 14. Changes to the "personal" filter test:
2310 (1) The test was buggy in that it was just doing the equivalent of
2311 "contains" tests on header lines. For example, if a user's address was
2312 anne@some.where, the "personal" test would incorrectly be true for
2314 To: susanne@some.where
2316 This test is now done by extracting each address from the header in turn,
2317 and checking the entire address. Other tests that are part of "personal"
2318 are now done using regular expressions (for example, to check local parts
2319 of addresses in From: header lines).
2321 (2) The list of non-personal local parts in From: addresses has been
2322 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2323 taken from the Sieve specification recommendations.
2325 (3) If the message contains any header line starting with "List-" it is
2326 treated as non-personal.
2328 (4) The test for "circular" in the Subject: header line has been removed
2329 because it now seems ill-conceived.
2331 15. Minor typos in src/EDITME comments corrected.
2333 16. Installed latest exipick from John Jetmore.
2335 17. If headers_add on a router specified a text string that was too long for
2336 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2337 of string_sprintf() is now avoided.
2339 18. $message_body_size was not set (it was always zero) when running the DATA
2340 ACL and the local_scan() function.
2342 19. For the "mail" command in an Exim filter, no default was being set for
2343 the once_repeat time, causing a random time value to be used if "once" was
2344 specified. (If the value happened to be <= 0, no repeat happened.) The
2345 default is now 0s, meaning "never repeat". The "vacation" command was OK
2346 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2347 (I found it when inspecting the code).
2349 20. There is now an overall timeout for performing a callout verification. It
2350 defaults to 4 times the callout timeout, which applies to individual SMTP
2351 commands during the callout. The overall timeout applies when there is more
2352 than one host that can be tried. The timeout is checked before trying the
2353 next host. This prevents very long delays if there are a large number of
2354 hosts and all are timing out (e.g. when the network connections are timing
2355 out). The value of the overall timeout can be changed by specifying an
2356 additional sub-option for "callout", called "maxwait". For example:
2358 verify = sender/callout=5s,maxwait=20s
2360 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2361 to the end before writing, but this should make it even safer).
2363 22. Exim was forgetting that it had advertised PIPELINING for the second and
2364 subsequent messages on an SMTP connection. It was also not resetting its
2365 memory on STARTTLS and an internal HELO.
2367 23. When Exim logs an SMTP synchronization error within a session, it now
2368 records whether PIPELINING has been advertised or not.
2370 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2371 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2372 rather than long int.
2374 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2376 26. Added the never_mail option to autoreply.
2382 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2383 crash if the getsockname() call failed; this can happen if a connection is
2384 closed very soon after it is established. The problem was simply in the
2385 order in which certain operations were done, causing Exim to try to write
2386 to the SMTP stream before it had set up the file descriptor. The bug has
2387 been fixed by making things happen in the correct order.
2393 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2394 before the connection was closed, thus losing the rejection response.
2396 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2397 some early Solaris releases, but causes trouble in current releases where
2398 socklen_t is defined.
2400 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2403 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2406 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2407 that releasing the top of it at the end releases what was used for sub-
2408 expansions (unless the block got too big). However, discard this block if
2409 the first thing is a variable or header, so that we can use its block when
2410 it is dynamic (useful for very large $message_headers, for example).
2412 6. Lookups now cache *every* query, not just the most recent. A new, separate
2413 store pool is used for this. It can be recovered when all lookup caches are
2414 flushed. Lookups now release memory at the end of their result strings.
2415 This has involved some general refactoring of the lookup sources.
2417 7. Some code has been added to the store_xxx() functions to reduce the amount
2418 of flapping under certain conditions.
2420 8. log_incoming_interface used to affect only the <= reception log lines. Now
2421 it causes the local interface and port to be added to several more SMTP log
2422 lines, for example "SMTP connection from", and rejection lines.
2424 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2426 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2428 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2429 could be overwritten at the end of the current message (or the start of a
2430 new message if it was set in a HELO ACL). The value is now preserved for
2431 the duration of the SMTP connection.
2433 12. If a transport had a headers_rewrite setting, and a matching header line
2434 contained an unqualified address, that address was qualified, even if it
2435 did not match any rewriting rules. The underlying bug was that the values
2436 of the flags that permit the existence of unqualified sender and recipient
2437 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2438 non-local messages, and by -bnq for local messages) were not being
2439 preserved with the message after it was received.
2441 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2442 "next input=" as part of the text comprising the host identity instead of
2443 the correct text. The code was using the same buffer for two different
2444 strings. However, depending on which order the printing function evaluated
2445 its arguments, the bug did not always show up. Under Linux, for example, my
2446 test suite worked just fine.
2448 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2449 doesn't work with some older versions of Perl. It has been changed to "my",
2450 which in any case is probably the better facility to use.
2452 15. A really picky compiler found some instances of statements for creating
2453 error messages that either had too many or two few arguments for the format
2456 16. The size of the buffer for calls to the DNS resolver has been increased
2457 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2458 for addresses that have a lot of PTR records. This alleviates a problem; it
2459 does not fully solve it.
2461 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2462 buffer now truncates the list and logs the incident, which is the same
2463 action as happens when Exim is looking up a host name and its aliases.
2464 Previously in this situation something unpredictable would happen;
2465 sometimes it was "internal error: store_reset failed".
2467 18. If a server dropped the connection unexpectedly when an Exim client was
2468 using GnuTLS and trying to read a response, the client delivery process
2469 crashed while trying to generate an error log message.
2471 19. If a "warn" verb in an ACL added multiple headers to a message in a single
2472 string, for example:
2474 warn message = H1: something\nH2: something
2476 the text was added as a single header line from Exim's point of view
2477 though it ended up OK in the delivered message. However, searching for the
2478 second and subsequent header lines using $h_h2: did not work. This has been
2479 fixed. Similarly, if a system filter added multiple headers in this way,
2480 the routers could not see them.
2482 20. Expanded the error message when iplsearch is called with an invalid key to
2483 suggest using net-iplsearch in a host list.
2485 21. When running tests using -bh, any delays imposed by "delay" modifiers in
2486 ACLs are no longer actually imposed (and a message to that effect is
2489 22. If a "gecos" field in a passwd entry contained escaped characters, in
2490 particular, if it contained a \" sequence, Exim got it wrong when building
2491 a From: or a Sender: header from that name. A second bug also caused
2492 incorrect handling when an unquoted " was present following a character
2493 that needed quoting.
2495 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2496 was not being matched caselessly.
2498 24. Arranged for all hyphens in the exim.8 source to be escaped with
2501 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2502 properly. Recipient callout cache records were still being keyed to include
2503 the sender, even when use_sender was set false. This led to far more
2504 callouts that were necessary. The sender is no longer included in the key
2505 when use_sender is false.
2507 26. Added "control = submission" modifier to ACLs.
2509 27. Added the ${base62d: operator to decode base 62 numbers.
2511 28. dnsdb lookups can now access SRV records.
2513 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2514 the configuration file.
2516 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2517 (-v) mode. This makes the output for a verbose queue run more intelligible.
2519 31. Added a use_postmaster feature to recipient callouts.
2521 32. Added the $body_zerocount variable, containing the number of binary zero
2522 bytes in the message body.
2524 33. The time of last modification of the "new" subdirectory is now used as the
2525 "mailbox time last read" when there is a quota error for a maildir
2528 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2530 35. Added +ignore_unknown as a special item in host lists.
2532 36. Code for decoding IPv6 addresses in host lists is now included, even if
2533 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2534 address was recognized as an IP address, but was then not correctly decoded
2535 into binary, causing unexpected and incorrect effects when compared with
2542 1. Very minor rewording of debugging text in manualroute to say "list of
2543 hosts" instead of "hostlist".
2545 2. If verify=header_syntax was set, and a header line with an unqualified
2546 address (no domain) and a large number of spaces between the end of the
2547 name and the colon was received, the reception process suffered a buffer
2548 overflow, and (when I tested it) crashed. This was caused by some obsolete
2549 code that should have been removed. The fix is to remove it!
2551 3. When running in the test harness, delay a bit after writing a bounce
2552 message to get a bit more predictability in the log output.
2554 4. Added a call to search_tidyup() just before forking a reception process. In
2555 theory, someone could use a lookup in the expansion of smtp_accept_max_
2556 per_host which, without the tidyup, could leave open a database connection.
2558 5. Added the variables $recipient_data and $sender_data which get set from a
2559 lookup success in an ACL "recipients" or "senders" condition, or a router
2560 "senders" option, similar to $domain_data and $local_part_data.
2562 6. Moved the writing of debug_print from before to after the "senders" test
2565 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2566 problems for message scanning, either using a data ACL, or using
2567 local_scan() because the Received: header was not generated till after they
2568 were called (in order to set the time as the time of reception completion).
2569 I have revised the way this works. The header is now generated after the
2570 body is received, but before the ACL or local_scan() are called. After they
2571 are run, the timestamp in the header is updated.
2577 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2578 before starting a queue runner without re-exec. This happened only when
2579 deliver_drop_privilege was set or when the Exim user was set to root. The
2580 effect of the bug was that timeouts during subsequent deliveries caused
2581 crashes instead of being properly handled. The handler is now left at its
2582 default (and expected) setting.
2584 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2585 message, again when deliver_drop_privilege is set or Exim is run as root.
2586 The bug described in (1) was not present in this case, but the tidying up
2587 of the other signals was missing. I have made the two cases consistent.
2589 3. The ignore_target_hosts setting on a manualroute router was being ignored
2590 for hosts that were looked up using the /MX notation.
2592 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2595 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2596 operated on the sender address. After changing the $sender_address to <>
2597 for the sender address verify, Exim was re-instated it as the original
2598 (before rewriting) address, but remembering that it had rewritten it, so it
2599 wasn't rewriting it again. This bug also had the effect of breaking the
2600 sender address verification caching when the sender address was rewritten.
2602 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2603 This has been changed so that if the ip literal address matches
2604 ignore_target_hosts, the router declines.
2606 7. Added expansion conditions match_domain, match_address, and match_local_
2607 part (NOT match_host).
2609 8. The placeholder for the Received: header didn't have a length field set.
2611 9. Added code to Exim itself and to exim_lock to test for a specific race
2612 condition that could lead to file corruption when using MBX delivery. The
2613 issue is with the lockfile that is created in /tmp. If this file is removed
2614 after a process has opened it but before that process has acquired a lock,
2615 there is the potential for a second process to recreate the file and also
2616 acquire a lock. This could lead to two Exim processes writing to the file
2617 at the same time. The added code performs the same test as UW imapd; it
2618 checks after acquiring the lock that its file descriptor still refers to
2619 the same named file.
2621 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2622 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2625 11. Added the smtp_active_hostname option. If used, this will typically be made
2626 to depend on the incoming interface address. Because $interface_address is
2627 not set up until the daemon has forked a reception process, error responses
2628 that can happen earlier (such as "too many connections") no longer contain
2631 12. If an expansion in a condition on a "warn" statement fails because a lookup
2632 defers, the "warn" statement is abandoned, and the next ACL statement is
2633 processed. Previously this caused the whole ACL to be aborted.
2635 13. Added the iplsearch lookup type.
2637 14. Added ident_timeout as a log selector.
2639 15. Added tls_certificate_verified as a log selector.
2641 16. Added a global option tls_require_ciphers (compare the smtp transport
2642 option of the same name). This controls incoming TLS connections.
2644 17. I finally figured out how to make tls_require_ciphers do a similar thing
2645 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2646 before starting the TLS session.
2648 18. Tabs are now shown as \t in -bP output.
2650 19. If the log selector return_path_on_delivery was set, Exim crashed when
2651 bouncing a message because it had too many Received: header lines.
2653 20. If two routers both had headers_remove settings, and the first one included
2654 a superfluous trailing colon, the final name in the first list and the
2655 first name in the second list were incorrectly joined into one item (with a
2656 colon in the middle).
2662 1. Added -C and -D options to the exinext utility, mainly to make it easier
2663 to include in the automated testing, but these could be helpful when
2664 multiple configurations are in use.
2666 2. The exinext utility was not formatting the output nicely when there was
2667 an alternate port involved in the retry record key, nor when there was a
2668 message id as well (for retries that were specific to a specific message
2669 and a specific host). It was also confused by IPv6 addresses, because of
2670 the additional colons they contain. I have fixed the IPv4 problem, and
2671 patched it up to do a reasonable job for IPv6.
2673 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2674 delivery, the log line now contains "pipelined" if PIPELINING was used.
2676 4. An SMTP transport process used to panic and die if the bind() call to set
2677 an explicit outgoing interface failed. This has been changed; it is now
2678 treated in the same way as a connect() failure.
2680 5. A reference to $sender_host_name in the part of a conditional expansion
2681 that was being skipped was still causing a DNS lookup. This no longer
2684 6. The def: expansion condition was not recognizing references to header lines
2685 that used bh_ and bheader_.
2687 7. Added the _cache feature to named lists.
2689 8. The code for checking quota_filecount in the appendfile transport was
2690 allowing one more file than it should have been.
2692 9. For compatibility with Sendmail, the command line option
2700 and sets the incoming protocol and host name (for trusted callers). The
2701 host name and its colon can be omitted when only the protocol is to be set.
2702 Note the Exim already has two private options, -pd and -ps, that refer to
2703 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2704 "s", but I don't think that's a major issue.
2706 10. A number of refactoring changes to the code, none of which should affect
2709 (a) The number of logging options was getting close to filling up the
2710 32-bit word that was used as a bit map. I have split them into two classes:
2711 those that are passed in the argument to log_write(), and those that are
2712 only ever tested independently outside of that function. These are now in
2713 separate 32-bit words, so there is plenty of room for expansion again.
2714 There is no change in the user interface or the logging behaviour.
2716 (b) When building, for example, log lines, the code previously used a
2717 macro that called string_cat() twice, in order to add two strings. This is
2718 not really sufficiently general. Furthermore, there was one instance where
2719 it was actually wrong because one of the argument was used twice, and in
2720 one call a function was used. (As it happened, calling the function twice
2721 did not affect the overall behaviour.) The macro has been replaced by a
2722 function that can join an arbitrary number of extra strings onto a growing
2725 (c) The code for expansion conditions now uses a table and a binary chop
2726 instead of a serial search (which was left over from when there were very
2727 few conditions). Also, it now recognizes conditions like "pam" even when
2728 the relevant support is not compiled in: a suitably worded error message is
2729 given if an attempt is made to use such a condition.
2731 11. Added ${time_interval:xxxxx}.
2733 12. A bug was causing one of the ddress fields not to be passed back correctly
2734 from remote delivery subprocesses. The field in question was not being
2735 subsequently used, so this caused to problems in practice.
2737 13. Added new log selectors queue_time and deliver_time.
2739 14. Might have fixed a bug in maildirsizefile handling that threw up
2740 "unexpected character" debug warnings, and recalculated the data
2741 unnecessarily. In any case, I expanded the warning message to give more
2744 15. Added the message "Restricted characters in address" to the statements in
2745 the default ACL that block characters like @ and % in local parts.
2747 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2748 Three changes have been made:
2750 (a) There was a serious bug; a negative response to MAIL caused the whole
2751 recipient domain to be cached as invalid, thereby blocking all messages
2752 to all local parts at the same domain, from all senders. This bug has
2753 been fixed. The domain is no longer cached after a negative response to
2754 MAIL if the sender used is not empty.
2756 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2759 (c) A new callout option, "use_sender" has been added for people who want
2760 the modified behaviour.
2766 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2769 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2770 indeed breaks things for older releases.
2772 3. Added additional logging to the case where there is a problem reading data
2773 from a filter that is running in a subprocess using a pipe, in order to
2774 try to track down a specific problem.
2776 4. Testing facility fudge: when running in the test harness and attempting
2777 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2778 getting "No route to host". Convert this to a timeout.
2780 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2783 6. Some OS don't have socklen_t but use size_t instead. This affects the
2784 fifth argument of getsockopt() amongst other things. This is now
2785 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2786 can be set for individual OS. I have set it for SunOS5, OSF1, and
2787 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2788 some earlier ones do not.
2790 7. Change 4.30/15 was not doing the test caselessly.
2792 8. The standard form for an IPv6 address literal was being rejected by address
2793 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2794 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2795 this, as well as the form without the "IPv6" on the front (but only when
2796 address literals are enabled, of course).
2798 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2800 10. Exim crashed if a message with an empty sender address specified by -f
2801 encountered a router with an errors_to setting. This could be provoked only
2802 by a command such as
2806 where an empty string was supplied; "<>" did not hit this bug.
2808 11. Installed PCRE release 4.5.
2810 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2811 remained set. It is now erased.
2813 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2814 times from message ids (which are base 36 rather than the normal 62).
2816 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2817 were being counted as actual protocol errors, and logged if the log
2818 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2819 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2820 and DATA following a set of rejected RCPTs do not count as protocol errors.
2821 In other words, Exim assumes they were pipelined, though this may not
2822 actually be the case. Of course, in all cases the client gets an
2823 appropriate error code.
2825 15. If a lookup fails in an ACL condition, a message about the failure may
2826 be available; it is used if testing the ACL cannot continue, because most
2827 such messages specify what the cause of the deferral is. However, some
2828 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2829 that caused an old message to be retained and used if a later statement
2830 caused a defer, replacing the real cause of the deferral.
2832 16. If an IP address had so many PTR records that the DNS lookup buffer
2833 was not large enough to hold them, Exim could crash while trying to process
2834 the truncated data. It now detects and logs this case.
2836 17. Further to 4.21/58, another change has been made: if (and only if) the
2837 first line of a message (the first header line) ends with CRLF, a bare LF
2838 in a subsequent header line has a space inserted after it, so as not to
2839 terminate the header.
2841 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2842 unnecessarily, used atoi() instead of strtol(), and didn't check the
2843 termination when getting file sizes from file names by regex.
2845 19. Completely re-implemented the support for maildirsize files, in the light
2846 of a number of problems with the previous contributed implementation
2847 (4.30/29). In particular:
2849 . If the quota is zero, the maildirsize file is maintained, but no quota is
2852 . If the maildir directory does not exist, it is created before any attempt
2853 to write a maildirsize file.
2855 . The quota value in the file is just a cache; if the quota is changed in
2856 the transport, the new value overrides.
2858 . A regular expression is available for excluding directories from the
2861 20. The autoreply transport checks the characters in options that define the
2862 message's headers; it allows continued headers, but it was checking with
2863 isspace() after an embedded newline instead of explicitly looking for a
2866 21. If all the "regular" hosts to which an address was routed had passed their
2867 expiry times, and had not reached their retry times, the address was
2868 bounced, even if fallback hosts were defined. Now Exim should go on to try
2871 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2872 equivalent code in the SMTP transport. Some hosts send humungous responses
2873 to HELO/EHLO, more than 1024 it seems.
2875 23. Refactoring: code in filter.c used (void *) for "any old type" but this
2876 gives compiler warnings in some environments. I've now done it "properly",
2879 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2880 (because of problems with the built-in one) was declared to return uschar *
2881 instead of char *, causing compiler failure.
2883 25. Fixed a file descriptor leak when processing alias/forward files.
2885 26. Fixed a minor format string issue in dbfn.c.
2887 27. Typo in exim.c: ("dmbnz" for "dbmnz").
2889 28. If a filter file refered to $h_xxx or $message_headers, and the headers
2890 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2893 29. When a sender address is verified, it is cached, to save repeating the test
2894 when there is more than one recipient in a message. However, when the
2895 verification involves a callout, it is possible for different callout
2896 options to be set for different recipients. It is too complicated to keep
2897 track of this in the cache, so now Exim always runs a verification when a
2898 callout is required, relying on the callout cache for the optimization.
2899 The overhead is duplication of the address routing, but this should not be
2902 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
2903 to be verified with callout=postmaster, and the main callout worked but the
2904 postmaster check failed, the verification correctly failed. However, if a
2905 subsequent RCPT command asked for sender verification *without* the
2906 postmaster check, incorrect caching caused this verification also to fail,
2909 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2910 it was not caching the DNS options (qualify_single, search_parents) that
2911 were used when the lookup failed. A subsequent lookup with different
2912 options therefore always gave the same answer, though there were cases
2913 where it should not have. (Example: a "domains = !$mx_any" option on a
2914 dnslookup router: the "domains" option is always processed without any
2915 widening, but the router might have qualify_single set.) Now Exim uses the
2916 cached value only when the same options are set.
2918 32. Added John Jetmore's "exipick" utility to the distribution.
2920 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
2921 than a timeout (e.g. a certificate is required, and is not provided), an
2922 Exim server now closes the connection immediately. Previously it waited for
2923 the client to close - but if the client is SSL, it seems that they each
2924 wait for each other, leading to a delay before one of them times out.
2926 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2927 maintained 0.8.x compatibility because I don't think many are using it, and
2928 it is clearly obsolete.
2930 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2933 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2934 client certificate was expired. A simple patch fixes this, though I don't
2935 understand the full logic of why the verify callback is called multiple
2938 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2939 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2940 which causes problems with some clients (such as the Certicom SSL Plus
2941 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2942 disables the coutermeasure allowing Eudora to connect."
2944 38. Exim was not checking that a write() to a log file succeeded. This could
2945 lead to Bad Things if a log got too big, in particular if it hit a file
2946 size limit. Exim now panics and dies if it cannot write to a log file, just
2947 as it does if it cannot open a log file.
2949 39. Modified OS/Makefile-Linux so that it now contains
2951 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2953 The two -D definitions ensure that Exim is compiled with large file
2954 support, which makes it possible to handle log files that are bigger than
2957 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2958 instance) a domain was checked against a named list that involved a lookup,
2959 causing $domain_data to be set, then another domain was checked against the
2960 same list, then the first domain was re-checked, the value of $domain_data
2961 after the final check could be wrong. In particular, if the second check
2962 failed, it could be set empty. This bug probably also applied to
2965 41. The strip_trailing_dot option was not being applied to the address given
2966 with the -f command-line option.
2968 42. The code for reading a message's header from the spool was incrementing
2969 $received_count, but never initializing it. This meant that the value was
2970 incorrect (doubled) while delivering a message in the same process in which
2971 it was received. In the most common configuration of Exim, this never
2972 happens - a fresh exec is done - but it can happen when
2973 deliver_drop_privilege is set.
2975 43. When Exim logs an SMTP synchronization error - client data sent too soon -
2976 it now includes up to 150 characters of the unexpected data in the log
2979 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2980 and building data strings. The size of both of these buffers was 10 000
2981 bytes - far larger than anybody would *ever* want, thought I. Needless to
2982 say, somebody hit the limit. I have increased the maximum line length to
2983 20 000 and the maximum data length of concatenated lines to 100 000. I have
2984 also fixed two bugs, because there was no checking on these buffers. Tsk,
2985 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2986 buffer is too small.
2988 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2989 lsearch lookups. Now it does.
2991 46. When parsing a route_list item in a manualroute router, a fixed-length
2992 buffer was used for the list of hosts. I made this 1024 bytes long,
2993 thinking that nobody would ever have a list of hosts that long. Wrong.
2994 Somebody had a whole pile of complicated expansion conditions, and the
2995 string was silently truncated, leading to an expansion error. It turns out
2996 that it is easier to change to an unlimited length (owing to other changes
2997 that have happened since this code was originally written) than to build
2998 structure for giving a limitation error. The length of the item that
2999 expands into the list of hosts is now unlimited.
3001 47. The lsearch lookup could not handle data where the length of text line was
3002 more than 4095 characters. Such lines were truncated, leading to shortened
3003 data being returned. It should now handle lines of any length.
3005 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3006 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3009 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3010 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3011 no longer happens in comment lines. A list of replacements is now placed
3012 at the head of all of the source files, except those whose only change is
3013 to replace PERL_COMMAND in the very first #! line.
3015 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3016 messages on the queue, with a bottom-up merge sort, using code contributed
3017 by Michael Haardt. This should make operations like -bp somewhat faster on
3018 large queues. It won't affect queue runners, except when queue_run_in_order
3021 51. Installed eximstats 1.31 in the distribution.
3023 52. Added support for SRV lookups to the dnslookup router.
3025 53. If an ACL referred to $message_body or $message_body_end, the value was not
3026 reset for any messages that followed in the same SMTP session.
3028 54. The store-handling optimization for building very long strings was not
3029 differentiating between the different store pools. I don't think this
3030 actually made any difference in practice, but I've tidied it.
3032 55. While running the routers to verify a sender address, $sender_address
3033 was still set to the sender address. This is wrong, because when routing to
3034 send a bounce to the sender, it would be empty. Therefore, I have changed
3035 it so that, while verifying a sender address, $sender_address is set to <>.
3036 (There is no change to what happens when verifying a recipient address.)
3038 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3039 target A or AAAA records (if not already returned) without resetting the
3040 qualify_single or search_parents options of the DNS resolver. These are
3041 inappropriate in this case because the targets of MX and SRV records must
3042 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3043 target that, when qualified, matched something in the local domain. These
3044 two options are now turned off when doing these lookups.
3046 57. It seems that at least some releases of Reiserfs (which does not have the
3047 concept of a fixed number of inodes) returns zero and not -1 for the
3048 number of available inodes. This interacted badly with check_spool_inodes,
3049 which assumed that -1 was the "no such thing" setting. What I have done is
3050 to check that the total number of inodes is greater than zero before doing
3051 the test of how many are available.
3053 58. When a "warn" ACL statement has a log_message modifier, the message is
3054 remembered, and not repeated. This is to avoid a lot of repetition when a
3055 message has many recipients that cause the same warning to be written.
3056 Howewer, Exim was preserving the list of already written lines for an
3057 entire SMTP session, which doesn't seem right. The memory is now reset if a
3058 new message is started.
3060 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3061 debugging output unless log_rewrite was also set.
3063 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3064 of (int)(handle) when we know that handle contains (void *)(-1).
3066 61. The Exim daemon panic-logs an error return when it closes the incoming
3067 connection. However "connection reset by peer" seems to be common, and
3068 isn't really an error worthy of noting specially, so that particular error
3071 62. When Exim is trying to find all the local interfaces, it used to panic and
3072 die if the ioctl to get the interface flags failed. However, it seems that
3073 on at least one OS (Solaris 9) it is possible to have an interface that is
3074 included in the list of interfaces, but for which you get a failure error
3075 for this call. This happens when the interface is not "plumbed" into a
3076 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3077 failure of the "get flags" call assumes that the interface is down.
3079 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3080 makes life easier for people who are doing arithmetic on fields extracted
3081 from dates, where you often get leading zeros that should not be
3082 interpreted as octal.
3084 64. Added qualify_domain to the redirect router, to override the global
3087 65. If a pathologically long header line contained very many addresses (the
3088 report of this problem mentioned 10 000) and each of them was rewritten,
3089 Exim could use up a very large amount of memory. (It kept on making new
3090 copies of the header line as it rewrote, and never released the old ones.)
3091 At the expense of a bit more processing, the header rewriting function has
3092 been changed so that it no longer eats memory in this way.
3094 66. The generation of the Received: header has been moved from the time that a
3095 message starts to be received, to the time that it finishes. The timestamp
3096 in the Received: header should now be very close to that of the <= log
3097 line. There are two side-effects of this change:
3099 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3100 logged header lines no longer include the local Received: line, because
3101 it has not yet been created. The same applies to a copy of the message
3102 that is returned to a non-SMTP sender when a message is rejected.
3104 (b) When a filter file is tested using -bf, no additional Received: header
3105 is added to the test message. After some thought, I decided that this
3108 This change does not affect the value of $received_for. It is still set
3109 after address rewriting, but before local_scan() is called.
3111 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3113 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3114 gave an unhelpful panic error message, and a defer error. I have managed to
3115 change this behaviour so that it now rejects any supplied certificate,
3116 which seems right, as the list of acceptable certificates is empty.
3118 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3119 gave an unhelpful defer error. I have not managed to make this reject any
3120 supplied certificates, but the error message it gives is "no certificate
3121 supplied", which is not helpful.
3123 70. exigrep's output now also includes lines that are not associated with any
3124 message, but which match the given pattern. Implemented by a patch from
3125 Martin Sluka, which also tidied up the Perl a bit.
3127 71. Recipient callout verification, like sender verification, was using <> in
3128 the MAIL FROM command. This isn't really the right thing, since the actual
3129 sender may affect whether the remote host accepts the recipient or not. I
3130 have changed it to use the actual sender in the callout; this means that
3131 the cache record is now keyed on a recipient/sender pair, not just the
3132 recipient address. There doesn't seem to be a real danger of callout loops,
3133 since a callout by the remote host to check the sender would use <>.
3134 [SEE ABOVE: changed after hitting problems.]
3136 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3137 temporary errors. However, in the case of such a code being given after
3138 the end of a data transmission (i.e. after ".") Exim was failing to write
3139 a retry record for the message. (Yes, there was some broken host that was
3140 actually sending 8xx at this point.)
3142 73. An unknown lookup type in a host list could cause Exim to panic-die when
3143 the list was checked. (An example that provoked this was putting <; in the
3144 middle of a list instead of at the start.) If this happened during a DATA
3145 ACL check, a -D file could be left lying around. This kind of configuration
3146 error no longer causes Exim to die; instead it causes a defer errror. The
3147 incident is still logged to the main and panic logs.
3149 74. Buglet left over from Exim 3 conversion. The message "too many messages
3150 in one connection" was written to the rejectlog but not the mainlog, except
3151 when address rewriting (yes!) was being logged.
3153 75. Added write_rejectlog option.
3155 76. When a system filter was run not as root (that is, when system_filter_user
3156 was set), the values of the $n variables were not being returned to the
3157 main process; thus, they were not subsequently available in the $sn
3160 77. Added +return_path_on_delivery log selector.
3162 78. A connection timeout was being treated differently from recipients deferred
3163 when testing hosts_max_try with a message that was older than the host's
3164 retry timeout. (The host should not be counted, thus allowing all hosts to
3165 be tried at least once before bouncing.) This may have been the cause of an
3166 occasionally reported bug whereby a message would remain on the queue
3167 longer than the retry timeout, but would be bounced if a delivery was
3168 forced. I say "may" because I never totally pinned down the problem;
3169 setting up timeout/retry tests is difficult. See also the next item.
3171 79. The ultimate address timeout was not being applied to errors that involved
3172 a combination of host plus message (for example, a timeout on a MAIL
3173 command). When an address resolved to a number of possible hosts, and they
3174 were not all tried for each delivery (e.g. because of hosts_max_try), a
3175 message could remain on the queue longer than the retry timeout.
3177 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3180 81. Fixed an obscure SMTP outgoing bug which required at least the following
3181 conditions: (a) there was another message waiting for the same server;
3182 (b) the server returned 5xx to all RCPT commands in the first message so
3183 that the message was not completed; (c) the server dropped the connection
3184 or gave a negative response to the RSET that Exim sends to abort the
3185 transaction. The observed case was a dropped connection after DATA that had
3186 been sent in pipelining mode. That is, the server had advertised PIPELINING
3187 but was not implementing it correctly. The effect of the bug was incorrect
3188 behaviour, such as trying another host, and this could lead to a crash.
3194 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3195 and daemon.c were passed as pointers to ints; they should have been
3196 pointers to socklen_t variables (which are typically unsigned ints).
3198 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3201 3. Fixed a really odd bug that affected only the testing scheme; patching a
3202 certain fixed string in the binary changed the value of another string that
3203 happened to be identical to the end of the original first string.
3205 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3206 name", it returns that address as the IP address. On some operating
3207 systems (e.g. Solaris), it also passes back the IP address string as the
3208 "host name". However, on others (e.g. Linux), it passes back an empty
3209 string. Exim wasn't checking for this, and was changing the host name to an
3210 empty string, assuming it had been canonicized.
3212 5. Although rare, it is permitted to have more than one PTR record for a given
3213 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3214 all the names associated with an address, because they do in Solaris.
3215 However, it seems that they do not in Linux for data that comes from the
3216 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3217 I found this out when I moved to a new Linux workstation and tried to run
3218 the Exim test suite.
3220 To get round this problem I have changed the code so that it now does its
3221 own call to the DNS to look up PTR records when searching for a host name.
3222 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3223 addresses that are only in /etc/hosts are still found.
3225 This behaviour is, however, controlled by an option called host_lookup_
3226 order, which defaults to "bydns:byaddr". If people want to use the other
3227 order, or indeed, just use one or the other means of lookup, they can
3228 specify it in this variable.
3230 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3231 some operating systems, this comes back from gethostbyaddr() as an empty
3232 string, and this is what Exim used to test for. However, it seems that in
3233 other systems, "." is yielded. Exim now tests for this case too.
3235 7. The values of check_spool_space and check_log_space are now held internally
3236 as a number of kilobytes instead of an absolute number of bytes. If a
3237 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3238 kilobyte. This means that much larger values can be stored.
3240 8. Exim monitor: an attempt to get the action menu when not actually pointing
3241 at a message produces an empty menu entitled "No message selected". This
3242 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3243 no entries in it ("Shell widget menu has zero width and/or height"). So I
3244 have added a single, blank menu entry in this case.
3246 9. Added ${quote_local_part.
3248 10. MIME decoding is now applied to the contents of Subject: header lines when
3251 11. Now that a reference to $sender_host_address automatically causes a reverse
3252 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3253 host lookup before query-style lookups in lists that might use this
3254 variable. This has therefore been abolished, and the "net-" prefix is no
3255 longer necessary for query-style lookups.
3257 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3258 have been a typo for LFLAGS, so it has been changed.
3260 13. The install script calls Exim with "-C /dev/null" in order to find the
3261 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3262 to be output. Howeve, since Exim outputs its version number before the
3263 error, it didn't break the script. It just looked ugly. I fixed this by
3264 always allowing "-C /dev/null" if the caller is root.
3266 14. Ignore overlarge ACL variable number when reading spool file - insurance
3267 against a later release with more variables having written the file.
3269 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3270 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3271 this, as well as the form without the "IPv6" on the front.
3273 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3274 OS/Makefile-Darwin file.
3276 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3277 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3279 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3280 variable. However, some OS use ints for inodes. I've added cast to long int
3281 to get rid of the compiler warning.
3283 19. I had forgotten to lock out "/../" in configuration file names when
3284 ALT_CONFIG_PREFIX was set.
3286 20. Routers used for verification do not need to specify transports. However,
3287 if such a router generated a host list, and callout was configured, Exim
3288 crashed, because it could not find a port number from the (non-existent)
3289 transport. It now assumes port 25 in this circumstance.
3291 21. Added the -t option to exigrep.
3293 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3294 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3295 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3296 them is set, LOOKUP_LSEARCH is forced.
3298 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3300 24. Added sender and host information to the "rejected by local_scan()" log
3301 line; previously there was no indication of these.
3303 25. Added .include_if_exists.
3305 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3306 Linux. It turns out that not all file systems support this. Apparently some
3307 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3308 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3309 not supported on the file descriptor, is now ignored when Exim is trying to
3310 sync a directory. This applies only to Linux.
3312 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3314 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3315 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3318 29. Added support for maildirsize files from supplied patch (modified a bit).
3320 30. The use of :fail: followed by an empty string could lead Exim to respond to
3321 sender verification failures with (e.g.):
3323 550 Verification failed for <xxx>
3324 550 Sender verify failed
3326 where the first response line was missing the '-' that indicates it is not
3327 the final line of the response.
3329 31. The loop for finding the name of the user that called Exim had a hardwired
3330 limit of 10; it now uses the value of finduser_retries, which is used for
3331 all other user lookups.
3333 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3336 33. Exim was neglecting to zero errno before one call of strtol() when
3337 expanding a string and expecting an integer value. On some systems this
3338 resulted in spurious "integer overflow" errors. Also, it was casting the
3339 result into an int without checking.
3341 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3342 did not work. The code looks as if it has *never* worked, though it appears
3343 to have been documented since at least releast 1.62. I have made it work.
3345 35. The "timeout_DNS" error in retry rules, also documented since at least
3346 1.62, also never worked. As it isn't clear exactly what this means, and
3347 clearly it isn't a major issue, I have abolished the feature by treating it
3348 as "timeout", and writing a warning to the main and panic logs.
3350 36. The display of retry rules for -brt wasn't always showing the error code
3353 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3354 timeout_connect_A, timeout_connect_MX.
3356 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3357 to the empty sender.
3359 39. The daemon was not analysing the content of -oX till after it had closed
3360 stderr and disconnected from the controlling terminal. This meant that any
3361 syntax errors were only noted on the panic log, and the return code from
3362 the command was 0. By re-arranging the code a little, I've made the
3363 decoding happen first, so such errors now appear on stderr, and the return
3364 code is 1. However, the actual setting up of the sockets still happens in
3365 the disconnected process, so errors there are still only recorded on the
3368 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3369 connections (as happens on some IP stacks) was logged at start up time as
3370 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3371 it from "IPv6 and IPv4", which means that two separate sockets are being
3374 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3375 says whether AF_INET or AF_INET6 was passed as an argument.
3377 42. Exiwhat output was messed up when time zones were included in log
3380 43. Exiwhat now gives more information about the daemon's listening ports,
3381 and whether -tls-on-connect was used.
3383 44. The "port" option of the smtp transport is now expanded.
3385 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3386 silently ignored. Now an error message is written to the main and panic
3389 46. There's a new ACL modifier called "logwrite" which writes to a log file
3390 as soon as it is encountered.
3392 47. Added $local_user_uid and $local_user_gid at routing time.
3394 48. Exim crashed when trying to verify a sender address that was being
3397 49. Exim was recognizing only a space character after ".include". It now also
3398 recognizes a tab character.
3400 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3401 extracting the relevant information from the specification. The man page no
3402 longer contains scrambled data for the -d option, and I've added a section
3403 at the front about calling Exim under different names.
3405 51. Added "extra_headers" argument to the "mail" command in filter files.
3407 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3410 53. Installed eximstats 1.29.
3412 54. Added transport_filter_timeout as a generic transport option.
3414 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3415 Cc: header lines. This was required by RFC 822, but it not required by RFC
3418 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3419 incoming messages that did not have them. Now it does so only if the
3420 message originates locally, that is, if there is no associated remote host
3421 address. When Resent- header lines are present, this applies to the Resent-
3422 lines rather than the non-Resent- lines.
3424 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3425 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3427 58. Messages for configuration errors now include the name of the main
3428 configuration file - useful now that there may be more than one file in a
3429 list (.included file names were always shown).
3431 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3432 for those rare installations that do not start the daemon as root or run it
3433 setuid root. I've cut out the call to initgroups() if the daemon is not
3436 60. The Exim user and group can now be bound into the binary as text strings
3437 that are looked up at the start of Exim's processing.
3439 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3441 62. Added $mailstore_basename variable.
3443 63. Installed patch to sieve.c from Michael Haardt.
3445 64. When Exim failed to open the panic log after failing to open the main log,
3446 the original message it was trying to log was written to stderr and debug
3447 output, but if they were not available (the usual case in production), it
3448 was lost. Now it is written to syslog before the two lines that record the
3449 failures to open the logs.
3451 65. Users' Exim filters run in subprocesses under the user's uid. It is
3452 possible for a "deliver" command or an alias in a "personal" command to
3453 provoke an address rewrite. If logging of address rewriting is configured,
3454 this fails because the process is not running as root or exim. There may be
3455 a better way of dealing with this, but for the moment (because 4.30 needs
3456 to be released), I have disabled address rewrite logging when running a
3457 filter in a non-root, non-exim process.
3463 1. The buildconfig auxiliary program wasn't quoting the value set for
3464 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3465 not defined. This bug was masked in 4.22 by the effect that was fixed in
3468 2. Some messages that were rejected after a message id was allocated were
3469 shown as "incomplete" by exigrep. It no longer does this for messages that
3470 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3472 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3473 have allow_domain_literals set, the ID did not get logged in the <= line.
3474 Domain literals are now always recognized in Message-ID: header lines.
3476 4. The first argument for a ${extract expansion item is the key name or field
3477 number. Leading and trailing spaces in this item were not being ignored,
3478 causing some misleading effects.
3480 5. When deliver_drop_privilege was set, single queue runner processes started
3481 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3482 same command in the process it spins off) were not dropping privilege.
3484 6. When the daemon running as "exim" started a queue runner, it always
3485 re-executed Exim in the spun-off process. This is a waste of effort when
3486 deliver_drop_privilege is set. The new process now just calls the
3487 queue-runner function directly.
3493 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3496 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3497 ignored. Though the use of -oP was forcing the writing of a pid file, it
3498 was always written to the default place.
3500 3. If the message "no IP address found for host xxxx" is generated during
3501 incoming verification, it is now followed by identification of the incoming
3502 connection (so you can more easily find what provoked it).
3504 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3506 5. Added some features to "harden" Exim a bit more against certain attacks:
3508 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3509 be put in Local/Makefile. This is like the never_users runtime option,
3510 but it cannot be overridden. The default setting is "root".
3512 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3513 prefix string with which any file named in a -C command line option
3516 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3517 is retained for -C and -D only if the caller of Exim is root. Without
3518 it, the exim user may also use -C and -D and retain privilege.
3520 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3521 command line option is disabled.
3523 6. Macro names set by the -D option must start with an upper case letter, just
3524 like macro names defined in the configuration file.
3526 7. Added "dereference=" facility to LDAP.
3528 8. Two instances of the typo "uknown" in the source files are fixed.
3530 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3531 the Configure-Makefile script screwed up while processing it.
3533 10. Incorporated PCRE 4.4.
3535 11. The SMTP synchronization check was not operating right at the start of an
3536 SMTP session. For example, it could not catch a HELO sent before the client
3537 waited for the greeting. There is now a check for outstanding input at the
3538 point when the greeting is written. Because of the duplex, asynchronous
3539 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3540 way, but not yet received, when the check is performed.
3542 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3543 on TCP/IP sockets, because this apparently causes some broken clients to
3546 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3547 unchanged) from the Cygwin maintainer.
3549 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3550 was supported instead of testing for mbx. Effectively a typo.
3552 15. The spa authenticator server code was not checking that the input it
3553 received was valid base64.
3555 16. The debug output line for the "set" modifier in ACLs was not showing the
3556 name of the variable that was being set.
3558 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3560 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3561 lookup on its own. Something else was needed to trigger the lookup. For
3562 example, a match in host_lookup or the need for a host name in a host list.
3563 Now, if $sender_host_name is referenced and the host name has not yet been
3564 looked up, a lookup is performed. If the lookup fails, the variable remains
3565 empty, and $host_lookup_failed is set to "1".
3567 19. Added "eqi" as a case-independent comparison operator.
3569 20. The saslauthd authentication condition could segfault if neither service
3570 nor realm was specified.
3572 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3573 error message that was logged was misleading, and incoming SMTP
3574 connections were dropped. The message is now more accurate, and temporary
3575 errors are given to SMTP connections.
3577 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3578 (except RSET) with a 421 error, until QUIT is received. However, it was
3579 failing to send a response to QUIT.
3581 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3582 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3583 ACL is now run after the helo_try_verify_hosts code.
3585 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3586 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3587 case-independent, so other case variants are also recognized). Apparently
3588 some systems use these upper case variants.
3590 25. If more than two messages were waiting for the same host, and a transport
3591 filter was specified for the transport, Exim sent two messages over the
3592 same TCP/IP connection, and then failed with "socket operation on non-
3593 socket" when it tried to send the third.
3595 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3597 27. The extern definition of crypt16() in expand.c was not being excluded when
3598 the OS had its own crypt16() function.
3600 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3601 as a preferred synonym for return_size_limit, both as an option and as an
3604 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3606 30. Changed the default configuration ACL to relax the local part checking rule
3607 for addresses that are not in any local domains. For these addresses,
3608 slashes and pipe symbols are allowed within local parts, but the sequence
3609 /../ is explicitly forbidden.
3611 31. SPA server authentication was not clearing the challenge buffer before
3614 32. log_message in a "warn" ACL statement was writing to the reject log as
3615 well as to the main log, which contradicts the documentation and doesn't
3616 seem right (because no rejection is happening). So I have stopped it.
3618 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3619 However, I am unable to do any testing of this.
3621 34. Fixed an infelicity in the appendfile transport. When checking directories
3622 for a mailbox, to see if any needed to be created, it was accidentally
3623 using path names with one or more superfluous leading slashes; tracing
3624 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3626 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3627 contents are added to the log line that is written for every discarded
3628 recipient. (Previously a log_message setting was ignored.)
3630 36. The ${quote: operator now quotes the string if it is empty.
3632 37. The install script runs exim in order to find its version number. If for
3633 some reason other than non-existence or emptiness, which it checks, it
3634 could not run './exim', it was installing it with an empty version number,
3635 i.e. as "exim-". This error state is now caught, and the installation is
3638 38. An argument was missing from the function that creates an error message
3639 when Exim fails to connect to the socket for saslauthd authentication.
3640 This could cause Exim to crash, or give a corrupted message.
3642 39. Added isip, isip4, and isip6 to ${if conditions.
3644 40. The ACL variables $acl_xx are now saved with the message, and can be
3645 accessed later in routers, transports, and filters.
3647 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3648 strings in the file are not string-expanded.
3650 42. If a MAIL command specified a SIZE value that was too large to fit into an
3651 int variable, the check against message_size_limit failed. Such values are
3652 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3653 day this will have to be increased, but I don't think I want to be around
3654 when emails are that large.
3661 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3662 iconv() is not standard in FreeBSD.
3664 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3665 IPv6 enabled. The observed symptom was a segmentation fault on return from
3666 the function os_common_find_running_interfaces() in src/os.c.
3668 3. In the check_special_case() function in daemon.c I had used "errno" as an
3669 argument name, which causes warnings on some systems. This was basically a
3670 typo, since it was named "eno" in the comments!
3672 4. The code that waits for the clock to tick (at a resolution of some fraction
3673 of a second) so as to ensure message-id uniqueness was always waiting for
3674 at least one whole tick, when it could have waited for less. [This is
3675 almost certainly not relevant at current processor speeds, where it is
3676 unlikely to ever wait at all. But we try to future-proof.]
3678 5. The function that sleeps for a time interval that includes fractions of a
3679 second contained a race. It did not block SIGALRM between setting the
3680 timer, and suspending (a couple of lines later). If the interval was short
3681 and the sigsuspend() was delayed until after it had expired, the suspension
3682 never ended. On busy systems this could lead to processes getting stuck for
3685 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3686 process, before it forks any delivery processes. The open lookup caching
3687 mechanism meant that the open file or database connection was passed into
3688 the delivery process. The problem was that delivery processes always tidy
3689 up cached lookup data. This could cause a problem for the next delivery
3690 process started by the queue runner, because the external queue runner
3691 process does not know about the closure. So the next delivery process
3692 still has data in the lookup cache. In the case of a file lookup, there was
3693 no problem because closing a file descriptor in a subprocess doesn't affect
3694 the parent. However, if the lookup was caching a connection to a database,
3695 the connection was closed, and the second delivery process was likely to
3696 see errors such as "PGSQL: query failed: server closed the connection
3697 unexpectedly". The problem has been fixed by closing all cached lookups
3698 in a queue runner before running a delivery process.
3700 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3701 seem to have the "const" qualifier which it has on other OS. I've
3704 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3705 *of the same type* (client or server) with the same public name that an
3706 error should be diagnosed.
3708 9. When Exim looked up a host name for an IP address, but failed to find the
3709 original IP address when looking up the host name (a safety check), it
3710 output the message "<ip address> does not match any IP for NULL", which was
3711 confusing, to say the least. The bug was that the host name should have
3712 appeared instead of "NULL".
3714 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3715 user that is built into the binary, and the -C or -D options is used, root
3716 privilege is dropped before the configuration file is read. In addition,
3717 logging is switched to stderr instead of the normal log files. If the
3718 configuration then re-defines the Exim user, the unprivileged environment
3719 is probably not what is expected, so Exim logs a panic warning message (but
3722 However, if deliver_drop_privilege is set, the unprivileged state may well
3723 be exactly what is intended, so the warning has been cut out in that case,
3724 and Exim is allowed to try to write to its normal log files.
3730 1. smtp_return_error_details was not giving details for temporary sender
3731 or receiver verification errors.
3733 2. Diagnose a configuration error if two authenticators have the same public
3736 3. Exim used not to create the message log file for a message until the first
3737 delivery attempt. This could be confusing when incoming messages were held
3738 for policy or load reasons. The message log file is now created at the time
3739 the message is received, and an initial "Received" line is written to it.
3741 4. The automatically generated man page for command line options had a minor
3742 bug that caused no ill effects; however, a more serious problem was that
3743 the procedure for building the man page automatically didn't always
3744 operate. Consequently, release 4.20 contains an out-of-date version. This
3745 shouldn't happen again.
3747 5. When building Exim with embedded Perl support, the script that builds the
3748 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3749 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3751 6. The freeze_tell option was not being used for messages that were frozen on
3752 arrival, either by an ACL or by local_scan().
3754 7. Added the smtp_incomplete_transaction log selector.
3756 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3757 was accepting AUTH without a new EHLO.
3759 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3760 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3763 10. Logging of TCP/IP connections (when configured) now happens in the main
3764 daemon process instead of the child process, so that the TCP/IP connection
3765 count is more accurate (but it can never be perfect).
3767 11. The use of "drop" in a nested ACL was not being handled correctly in the
3768 outer ACL. Now, if condition failure induced by the nested "drop" causes
3769 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3770 or "require"), the connection is dropped.
3772 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3773 that yield "discard" can now be used with an "accept" or a "discard" verb,
3774 but an error is generated for any others (because I can't see a useful way
3775 to define what should happen).
3777 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3778 are now processed in the same way as lines in the Exim configuration file.
3779 In particular, continuation lines are supported.
3781 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3783 15. Added -ti meaning -t -i.
3785 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3786 domains, and warn by logging if others are found.
3788 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3789 returned from GIFCONF when Exim is trying to find the list of interfaces on
3790 a host. The code in os.c has been modified to copy each ifreq to an aligned
3791 structure in all cases.
3793 Also, in some cases, the returned ifreq's were being copied to a 'struct
3794 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3795 means the last couple of bytes of an IPv6 address could be chopped if the
3796 ifreq contained only a normal sockaddr (14 bytes storage).
3798 18. Named domain lists were not supported in the hosts_treat_as_local option.
3799 An entry such as +xxxx was not recognized, and was treated as a literal
3802 19. Ensure that header lines added by a DATA ACL are included in the reject log
3803 if the ACL subsequently rejects the message.
3805 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3806 MD5 (which is deprecated).
3808 21. When testing a filter file using -bf, Exim was writing a message when it
3809 took the sender from a "From " line in the message, but it was not doing so
3810 when it took $return_path from a Return-Path: header line. It now does.
3812 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3813 with a valid header line field name (a series of printing characters
3814 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3816 23. Changed "disc" in the source to "disk" to conform to the documentation and
3817 the book and for uniformity.
3819 24. Ignore Sendmail's -Ooption=value command line item.
3821 25. When execve() failed while trying to run a command in a pipe transport,
3822 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3823 could be confused with a return value of 69 from the command itself. This
3824 has been changed to 127, the value the shell returns if it is asked to run
3825 a non-existent command. The wording for the related log line suggests a
3826 non-existent command as the problem.
3828 26. If received_header_text expands to an empty string, do not add a Received:
3829 header line to the message. (Well, it adds a token one on the spool, but
3830 marks it "old" so that it doesn't get used or transmitted.)
3832 27. Installed eximstats 1.28 (addition of -nt option).
3834 28. There was no check for failure on the call to getsockname() in the daemon
3835 code. This can fail if there is a shortage of resources on the system, with
3836 ENOMEM, for example. A temporary error is now given on failure.
3838 29. Contrary to the C standard, it seems that in some environments, the
3839 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3840 program. Exim now does this explicitly; it affects the formatting of
3841 timestamps using strftime().
3843 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3844 complaints. I've now initialized all the variables, to avoid this.
3846 32. Header lines added by a system filter were not being "seen" during
3847 transport-time rewrites.
3849 33. The info_callback() function passed to OpenSSL is set up with type void
3850 (*)(SSL *, int, int), as described somewhere. However, when calling the
3851 function (actually a macro) that sets it up, the type void(*)() is
3852 expected. I've put in a cast to prevent warnings from picky compilers.
3854 34. If a DNS black list lookup found a CNAME record, but there were no A
3855 records associated with the domain it pointed at, Exim crashed.
3857 35. If a DNS black list lookup returned more than one A record, Exim ignored
3858 all but the first. It now scans all returned addresses if a particular IP
3859 value is being sought. In this situation, the contents of the
3860 $dnslist_value variable are a list of all the addresses, separated by a
3863 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3864 to accept a host name and all its aliases if the forward lookup for any of
3865 them yielded the IP address of the incoming connection. Now it accepts only
3866 those names whose forward lookup yields the correct IP address. Any other
3867 names are discarded. This closes a loophole whereby a rogue DNS
3868 administrator could create reverse DNS records to break through a
3869 wildcarded host restriction in an ACL.
3871 37. If a user filter or a system filter that ran in a subprocess used any of
3872 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3873 the wrong values were passed to the pipe command ($thisaddress had the
3874 value of $0, $0 had the value of $1, etc). This bug was introduced by
3875 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3877 38. Improved the line breaking for long SMTP error messages from ACLs.
3878 Previously, if there was no break point between 40 and 75 characters, Exim
3879 left the rest of the message alone. Two changes have been made: (a) I've
3880 reduced the minimum length to 35 characters; (b) if it can't find a break
3881 point between 35 and 75 characters, it looks ahead and uses the first one
3882 that it finds. This may give the occasional overlong line, but at least the
3883 remaining text gets split now.
3885 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3886 file descriptors might be low, and that setting 1000 would always raise it.
3887 It turns out that in some environments, the limit is already over 1000 and
3888 that lowering it causes trouble. So now Exim takes care not to decrease it.
3890 40. When delivering a message, the value of $return_path is set to $sender_
3891 address at the start of routing (routers may change the value). By an
3892 oversight, this default was not being set up when an address was tested by
3893 -bt or -bv, which affected the outcome if any router or filter referred to
3896 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
3897 log only when "message" or "log_message" are set. However, if one of the
3898 conditions was an address verification, or a call to a nested ACL, the
3899 messages generated by the underlying test were being passed through. This
3900 no longer happens. The underlying message is available in $acl_verify_
3901 message for both "message" and "log_message" expansions, so it can be
3902 passed through if needed.
3904 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3905 new expansion $bh_ to give the encoded byte string without charset
3906 translation. Translation happens only if iconv() is available; HAVE_ICONV
3907 indicates this at build time. HEADERS_CHARSET gives the charset to
3908 translate to; headers_charset can change it in the configuration, and
3909 "headers charset" can change it in an individual filter file.
3911 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
3912 that creates RFC 2047 encoded "words" labels them as that charset instead
3913 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3914 expansion operator; (ii) when Exim creates a From: line for a local
3915 message; (iii) when a header line is rewritten to include a "phrase" part.
3917 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3918 buggy, causing it to skip the first lines of messages whose message ID
3919 ended in 'D'. This would not have bitten before Exim release 4.14, because
3920 message IDs were unlikely to end in 'D' before then. The effect was to have
3921 incorrect size information for certain domains.
3923 45. #include "config.h" was missing at the start of the crypt16.c module. This
3924 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3927 46. If there was a timeout during a "random" callout check, Exim treated it as
3928 a failure of the random address, and carried on sending RSET and the real
3929 address. If the delay was just some slowness somewhere, the response to the
3930 original RCPT would be taken as a response to RSET and so on, causing
3931 mayhem of various kinds.
3933 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3934 when I implemented it. It didn't allow for the fact that some option values
3935 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3936 the right test for positive values.
3938 48. Domain names in DNS records are case-independent. Exim always looks them up
3939 in lower case. Some resolvers return domain names in exactly the case they
3940 appear in the zone file, that is, they may contain uppercase letters. Not
3941 all resolvers do this - some return always lower case. Exim was treating a
3942 change of case by a resolver as a change of domain, similar to a widening
3943 of a domain abbreviation. This triggered its re-routing code and so it was
3944 trying to route what was effectively the same domain again. This normally
3945 caused routing to fail (because the router wouldn't handle the domain
3946 twice). Now Exim checks for this case specially, and just changes the
3947 casing of the domain that it ultimately uses when it transmits the message
3950 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3953 50. If a filter generated a file delivery with a non-absolute name (possible if
3954 no home directory exists for the router), the forbid_file option was not
3957 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
3958 the existing equality matching.
3960 52. Exim was using ints instead of ino_t variables in some places where it was
3961 dealing with inode numbers.
3963 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3964 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3965 called TMPDIR, and if it finds it is different, it changes its value.
3967 54. The smtp_printf() function is now made available to local_scan() so
3968 additional output lines can be written before returning. There is also an
3969 smtp_fflush() function to enable the detection of a dropped connection.
3970 The variables smtp_input and smtp_batched_input are exported to
3973 55. Changed the default runtime configuration: the message "Unknown user"
3974 has been removed from the ACL, and instead placed on the localuser router,
3975 using the cannot_route_message feature. This means that any verification
3976 failures that generate their own messages won't get overridden. Similarly,
3977 the "Unrouteable address" message that was in the ACL for unverifiable
3978 relay addresses has also been removed.
3980 56. Added hosts_avoid_esmtp to the smtp transport.
3982 57. The exicyclog script was not checking for the esoteric option
3983 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
3984 will work only if exicyclog is run under the appropriate euid.
3986 58. Following a discussion on the list, the rules by which Exim recognises line
3987 endings on incoming messages have been changed. The -dropcr and drop_cr
3988 options are now no-ops, retained only for backwards compatibility. The
3989 following line terminators are recognized: LF CRLF CR. However, special
3990 processing applies to CR:
3992 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
3993 nor a local message in the state where . is a terminator.
3995 (ii) If a bare CR is encountered in a header line, an extra space is added
3996 after the line terminator so as not to end the header. The reasoning
3997 behind this is that bare CRs in header lines are most likely either
3998 to be mistakes, or people trying to play silly games.
4000 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4001 was being incorrectly given as 18 bytes larger than it should have been.
4002 This is a VOB (very old bug).
4004 60. This may never have affected anything current, but just in case it has:
4005 When the local host is found other than at the start of a list of hosts,
4006 the local host, those with the same MX, and any that follow, are discarded.
4007 When the list in question was part of a longer list of hosts, the following
4008 hosts (not currently being processed) were also being discarded. This no
4009 longer happens. I'm not sure if this situation could ever has previously
4012 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4015 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4016 header. This is something that is recommended in a new Internet Draft, and
4017 is something that is documented as being done by Sendmail. There are two
4018 possible values. For messages generated by the autoreply transport, Exim
4021 Auto-Submitted: auto-replied
4023 whereas for all other generated messages (e.g. bounces) it adds
4025 Auto-Submitted: auto-generated
4027 63. The "personal" condition in filters now includes a test for the
4028 Auto-Submitted: header. If it contains the string "auto-" the message it
4029 not considered personal.
4031 64. Added rcpt_include_affixes as a generic transport option.
4033 65. Added queue_only_override (default true).
4035 66. Added the syslog_duplication option.
4037 67. If what should have been the first header line of a message consisted of
4038 a space followed by a colon, Exim was mis-interpreting it as a header line.
4039 It isn't of course - it is syntactically invalid and should therefore be
4040 treated as the start of the message body. The misbehaviour could have
4041 caused a number of strange effects, including loss of data in subsequent
4042 header lines, and spool format errors.
4044 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4045 client host had authenticated. This control can now be exercised by an ACL
4046 for more flexibility.
4048 69. By default, callouts do not happen when testing with -bh. There is now a
4049 variant, -bhc, which does actually run the callout code, including
4050 consulting and updating the callout cache.
4052 70. Added support for saslauthd authentication, courtesy of Alexander
4055 71. If statvfs() failed on the spool or log directories while checking their
4056 size for availability, Exim confusingly gave the error "space shortage".
4057 Furthermore, in debugging mode it crashed with a floating point exception.
4058 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4059 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4060 serious problem, Exim now writes to the main and panic logs when this
4061 happens, with details of the failure. It then refuses to accept the
4062 incoming message, giving the message "spool directory problem" or "log
4063 directory problem" with a 421 code for SMTP messages.
4065 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4066 0, 1, and 2 exist, because some OS complain for execs without them (see
4067 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4068 descriptors. However, the code omitted to check that the open succeeded,
4069 causing mysterious errors if for some reason the permissions on /dev/null
4070 got screwed. Now Exim writes a message to the main and panic logs, and
4071 bombs out if it can't open /dev/null.
4073 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4074 interact so that it is all more flexible. It is supposed to remain
4075 backwards compatible. Also added extra_local_interfaces.
4077 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4078 to bomb out with an assertion failure - to the client this appears as a
4079 connection drop. This problem occurs in the part of the code that was taken
4080 from the Samba project. Fortunately, the assertion is in a very simple
4081 function, so I have fixed this by reproducing the function inline in the
4082 one place where it is called, and arranging for authentication to fail
4083 instead of killing the process with assert().
4085 75. The SPA client code was not working when the server requested OEM rather
4086 than Unicode encoding.
4088 76. Added code to make require_files with a specific uid setting more usable in
4089 the case where statting the file as root fails - usually a non-root-mounted
4090 NFS file system. When this happens and the failure is EACCES, Exim now
4091 forks a subprocess and does the per-uid checking as the relevant uid.
4093 77. Added process_log_path.
4095 78. If log_file_path was not explicitly set, a setting of check_log_space or
4096 check_log_inodes was ignored.
4098 79. If a space check for the spool or log partitions fails, the incident is now
4099 logged. Of course, in the latter case the data may get lost...
4101 80. Added the %p formatting code to string_format() so that it can be used to
4102 print addresses in debug_print(). Adjusted all the address printing in the
4103 debugging in store.c to use %p rather than %d.
4105 81. There was a concern that a line of code in smtp_in.c could overflow a
4106 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4107 initially expressed, the concern was not well-founded, because trailing
4108 spaces are removed early. However, if the trailing spaces were followed by
4109 a NULL, they did not get removed, so the overflow was possible. Two fixes
4112 (a) I re-wrote the offending code in a cleaner fashion.
4113 (b) If an incoming SMTP command contains a NULL character, it is rejected
4116 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4117 runs initgroups(), so that if the Exim user is in any additional groups,
4118 they will be used during message reception.
4124 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob