1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.409 2006/10/16 13:43:21 ph10 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 TF/01 In the add_headers option to the mail command in an Exim filter,
10 there was a bug that Exim would claim a syntax error in any
11 header after the first one which had an odd number of characters
14 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
15 callout verification, Exim cached a "reject" for the entire domain. This
16 is correct for most verifications, but it is not correct for a recipient
17 verification with use_sender or use_postmaster set, because in that case
18 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
19 case of MAIL FROM:<> rejection from other early rejections (e.g.
20 rejection of HELO). When verifying a recipient using a non-null MAIL
21 address, the cache is ignored if it shows MAIL FROM:<> rejection.
22 Whatever the result of the callout, the value of the domain cache is
23 left unchanged (for any other kind of callout, getting as far as trying
24 RCPT means that the domain itself is ok).
26 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
29 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
30 manifest itself as EPIPE rather than ECONNECT. When tidying away a
31 session, the daemon ignores ECONNECT errors and logs others; it now
32 ignores EPIPE as well.
34 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
35 (quoted-printable decoding).
37 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
38 later the small subsequent patch to fix an introduced bug.
40 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
42 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
44 PH/08 An error is now given if message_size_limit is specified negative.
46 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
47 to be given (somewhat) arbitrary names.
49 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
52 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
53 miscellaneous code fixes
55 PH/10 Added the log_reject_target ACL modifier to specify where to log
58 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
59 hostname. This is wrong, because it relates to the incoming message (and
60 probably the interface on which it is arriving) and not to the outgoing
61 callout (which could be using a different interface). This has been
62 changed to use the value of the helo_data option from the smtp transport
63 instead - this is what is used when a message is actually being sent. If
64 there is no remote transport (possible with a router that sets up host
65 addresses), $smtp_active_hostname is used.
67 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
68 tweaks were necessary in order to get it to work (see also 21 below):
69 (a) The code assumed that strncpy() returns a negative number on buffer
70 overflow, which isn't the case. Replaced with Exim's string_format()
72 (b) There were several signed/unsigned issues. I just did the minimum
73 hacking in of casts. There is scope for a larger refactoring.
74 (c) The code used strcasecmp() which is not a standard C function.
75 Replaced with Exim's strcmpic() function.
76 (d) The code set only $1; it now sets $auth1 as well.
77 (e) A simple test gave the error "authentication client didn't specify
78 service in request". It would seem that Dovecot has changed its
79 interface. Fortunately there's a specification; I followed it and
80 changed what the client sends and it appears to be working now.
82 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
85 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
86 address A is aliased to B and C, where B exists and C does not. Without
87 -v the output is "A verified" because verification stops after a
88 successful redirection if more than one address is generated. However,
89 with -v the child addresses are also verified. Exim was outputting "A
90 failed to verify" and then showing the successful verification for C,
91 with its parentage. It now outputs "B failed to verify", showing B's
92 parentage before showing the successful verification of C.
94 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
95 look up a TXT record in a specific list after matching in a combined
98 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
99 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
100 they consult the DNS. I had assumed they would set it the way they
101 wanted; and indeed my experiments on Linux seem to show that in some
102 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
103 To be on the safe side, however, I have now made the interface to
104 host_find_byname() similar to host_find_bydns(), with an argument
105 containing the DNS resolver options. The host_find_byname() function now
106 sets these options at its start, just as host_find_bydns() does. The smtp
107 transport options dns_qualify_single and dns_search_parents are passed to
108 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
109 of host_find_byname() use the default settings of RES_DEFNAMES
110 (qualify_single) but not RES_DNSRCH (search_parents).
112 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
113 spool_read_header() do less string testing, by means of a preliminary
114 switch on the second character of optional "-foo" lines. (This is
115 overdue, caused by the large number of possibilities that now exist.
116 Originally there were few.) While I was there, I also converted the
117 str(n)cmp tests so they don't re-test the leading "-" and the first
118 character, in the hope this might squeeze out yet more improvement.
120 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
121 flag allowing group syntax was set by the header_syntax check but not
122 turned off, possible causing trouble later; (2) The flag was not being
123 set at all for the header_verify test, causing "group"-style headers to
124 be rejected. I have now set it in this case, and also caused header_
125 verify to ignore an empty address taken from a group. While doing this, I
126 came across some other cases where the code for allowing group syntax
127 while scanning a header line wasn't quite right (mostly, not resetting
128 the flag correctly in the right place). These bugs could have caused
129 trouble for malformed header lines. I hope it is now all correct.
131 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
132 with the "reply" argument non-NULL. The code, however (which originally
133 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
134 but it didn't always do it. This confused somebody who was copying the
135 code for some other use. I have removed all the tests.
137 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
138 feature that was used to support insecure browsers during the U.S. crypto
139 embargo. It requires special client support, and Exim is probably the
140 only MTA that supported it -- and would never use it because real RSA is
141 always available. This code has been removed, because it had the bad
142 effect of slowing Exim down by computing (never used) parameters for the
143 RSA_EXPORT functionality.
145 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
146 authenticator to fail if there's a tab character in the incoming data
147 (there should never be unless someone is messing about, as it's supposed
148 to be base64-encoded). Also added, on Timo's advice, the "secured" option
149 if the connection is using TLS or if the remote IP is the same as the
150 local IP, and the "valid-client-cert option" if a client certificate has
157 SC/01 Use a glob alias rather than an array ref in eximstats generated
158 parser. This improves both readability and performance.
160 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
161 Don't display local sender or destination tables in eximstats unless
162 there is data to show.
163 Added average volumes into the eximstats top table text output.
165 SC/03 Collect data on the number of addresses (recipients) as well
166 as the number of messages in eximstats.
168 TF/01 Correct an error in the documentation for the redirect router. Exim
169 does (usually) call initgroups() when daemonizing.
171 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
172 with consistent privilege compared to when running as a daemon.
174 TF/03 Note in the spec that $authenticated_id is not set for local
175 submissions from trusted users.
177 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
178 Thanks to Dean Brooks <dean@iglou.com> for the patch.
180 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
181 by adding some example configuration directives to the default
182 configuration file. A little bit of work is required to uncomment the
183 directives and define how usernames and passwords are checked, but
184 there is now a framework to start from.
186 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
187 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
188 without this. I don't know how relevant this is to other LDAP libraries.
190 PH/02 Add the verb name to the "unknown ACL verb" error.
192 PH/03 Magnus Holmgren's patch for filter_prepend_home.
194 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
196 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
197 directory not expanded when it should be if an expanded home directory
198 was set for the address (which is overridden by the transport).
200 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
203 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
204 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
205 because it is too late at that time, and has no effect.
207 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
208 security issue with \' (bugzilla #107). I could not use the
209 PQescapeStringConn() function, because it needs a PGconn value as one of
212 PH/08 When testing addresses using -bt, indicate those final addresses that
213 are duplicates that would not cause an additional delivery. At least one
214 person was confused, thinking that -bt output corresponded to deliveries.
215 (Suppressing duplicates isn't a good idea as you lose the information
216 about possibly different redirections that led to the duplicates.)
218 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
219 systems where poll() doesn't work, in particular OS X.
221 PH/10 Added more information to debugging output for retry time not reached.
223 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
224 operations in malware.c.
226 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
229 PH/13 If write_rejectlog was set false when logging was sent to syslog with
230 syslog_duplication set false, log lines that would normally be written
231 both the the main log and to the reject log were not written to syslog at
234 PH/14 In the default configuration, change the use of "message" in ACL warn
235 statements to "add_header".
237 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
238 not followed by a command (e.g. "seen endif").
240 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
241 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
244 PH/17 Added extra conditions to the default value of delay_warning_condition
248 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
249 { match{$h_precedence:}{(?i)bulk|list|junk} } \
250 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
253 The Auto-Submitted: and various List- headers are standardised, whereas I
254 don't think Precedence: ever was.
256 PH/18 Refactored debugging code in route_finduser() to show more information,
257 in particular, the error code if getpwnam() issues one.
259 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
260 This is apparently needed in addition to the PH/07 change above to avoid
261 any possible encoding problems.
263 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
264 but not after initializing Perl.
266 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
267 output them only if debugging. By default they are written stderr,
268 apparently, which is not desirable.
270 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
273 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
276 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
278 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
279 authenticated or an ident call has been made. Suppress the default
280 values for $authenticated_id and $authenticated_sender (but permit -oMai
281 and -oMas) when testing with -bh.
283 PH/24 Re-jigged the order of the tests in the default configuration so that the
284 tests for valid domains and recipients precede the DNS black list and CSA
285 tests, on the grounds that those ones are more expensive.
287 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
288 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
289 This bug exists in every version of Exim that I still have, right back to
292 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
293 However, an attempt to turn on case-sensitivity in a regex key by
294 including (?-i) didn't work because the subject string was already
295 lowercased, and the effects were non-intuitive. It turns out that a
296 one-line patch can be used to allow (?-i) to work as expected.
302 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
303 other effects) broke the use of negated acl sub-conditions.
305 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
308 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
309 "Deny" causes Exim to reject the incoming connection with a 554 error.
310 Unfortunately, if there is a major crisis, such as a disk failure,
311 tcp-wrappers gives "deny", whereas what one would like would be some
312 kind of temporary error. A kludge has been added to help with this.
313 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
314 554 error is used if errno is still zero or contains ENOENT (which occurs
315 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
318 PH/03 Add -lutil to the default FreeBSD LIBS setting.
320 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
321 errors. Otherwise a message that provokes a temporary error (when other
322 messages do not) can cause a whole host to time out.
324 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
325 addresses were routed directly to files or pipes from a redirect router.
326 File deliveries just didn't batch; pipe deliveries might have suffered
329 PH/06 A failure to get a lock for a hints database would erroneously always say
330 "Failed to get write lock", even when it was really a read lock.
332 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
333 of 0600. This has been changed to use the value of the lockfile_mode
334 option (which defaults to 0600).
336 PH/08 Applied small patch from the Sieve maintainer.
338 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
339 folder from quota calculations, a direct delivery into this folder messed
340 up the contents of the maildirsize file. This was because the regex was
341 used only to exclude .Trash (or whatever) when the size of the mailbox
342 was calculated. There was no check that a delivery was happening into an
343 excluded directory. This bug has been fixed by ignoring all quota
344 processing for deliveries into excluded directories.
346 PH/10 Added the maildirfolder_create_regex option to appendfile.
352 PH/01 The code for finding all the local interface addresses on a FreeBSD
353 system running IPv6 was broken. This may well have applied to all BSD
354 systems, as well as to others that have similar system calls. The broken
355 code found IPv4 interfaces correctly, but gave incorrect values for the
356 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
357 that it would not match correctly against @[] and not recognize the IPv6
360 PH/02 The ipliteral router was not recognizing addresses of the form user@
361 [ipv6:....] because it didn't know about the "ipv6:" prefix.
363 PH/03 Added disable_ipv6.
365 PH/04 Changed $reply_address to use the raw form of the headers instead of the
366 decoded form, because it is most often used to construct To: headers
367 lines in autoreplies, and the decoded form may well be syntactically
368 invalid. However, $reply_address has leading white space removed, and all
369 newlines turned into spaces so that the autoreply transport does not
372 PH/05 If group was specified without a user on a router, and no group or user
373 was specified on a transport, the group from the router was ignored.
375 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
376 for visible compile-time settings that can be used to change these
377 numbers, for those that want even more. Backwards compatibility with old
378 spool files has been maintained. However, going back to a previous Exim
379 release will lost any variables that are in spool files.
381 PH/07 Two small changes when running in the test harness: increase delay when
382 passing a TCP/IP connection to a new process, in case the original
383 process has to generate a bounce, and remove special handling of
384 127.0.0.2 (sic), which is no longer necessary.
386 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
387 be the same on different OS.
389 PH/09 Moved a debug statement in filter processing to avoid a race problem when
392 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
393 whether --show-vars was specified or not
395 JJ/02 exipick: Added support for new ACL variable spool format introduced
398 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
399 syntactically invalid From: or Reply-to: line, and a filter used this to
400 generate an autoreply, and therefore failed to obtain an address for the
401 autoreply, Exim could try to deliver to a non-existent relative file
402 name, causing unrelated and misleading errors. What now happens is that
403 it logs this as a hard delivery error, but does not attempt to create a
406 PH/11 The exinext utility has a -C option for testing purposes, but although
407 the given file was scanned by exinext itself; it wasn't being passed on
408 when Exim was called.
410 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
411 an end-of-file indication when reading a command response.
413 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
414 compiled. In many other places in Exim, IPv6 addresses are always
415 recognized, so I have changed this. It also means that IPv4 domain
416 literals of the form [IPV4:n.n.n.n] are now always recognized.
418 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
419 used if the router is not running as root, for example, when verifying at
420 ACL time, or when using -bh. The debugging output from this situation was
421 non-existent - all you got was a failure to exec. I have made two
424 (a) Failures to set uid/gid, the current directory, or a process leader
425 in a subprocess such as that created by queryprogram now generate
426 suitable debugging ouput when -d is set.
428 (b) The queryprogram router detects when it is not running as root,
429 outputs suitable debugging information if -d is set, and then runs
430 the subprocess without attempting to change uid/gid.
432 PH/15 Minor change to Makefile for building test_host (undocumented testing
435 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
436 additional section of a DNS packet that returns MX or SRV records.
437 Instead, it always explicitly searches for A/AAAA records. This avoids
438 major problems that occur when a DNS server includes only records of one
439 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
440 fixed another bug: if SRV records were looked up and the corresponding
441 address records were *not* found in the additional section, the port
442 values from the SRV records were lost.
444 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
445 using the correct key (the original address) when searching the retry
446 rules in order to find which one to use for generating the retry hint.
448 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
449 adding the default one. Similarly, if it contains a Reply-To: header, the
450 errors_reply_to option, if set, is not used.
452 PH/19 When calculating a retry time, Exim used to measure the "time since
453 failure" by looking at the "first failed" field in the retry record. Now
454 it does not use this if it is later than than the arrival time of the
455 message. Instead it uses the arrival time. This makes for better
456 behaviour in cases where some deliveries succeed, thus re-setting the
457 "first failed" field. An example is a quota failure for a huge message
458 when small messages continue to be delivered. Without this change, the
459 "time since failure" will always be short, possible causing more frequent
460 delivery attempts for the huge message than are intended.
461 [Note: This change was subsequently modified - see PH/04 for 4.62.]
463 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
464 $1, $2, $3) because the numerical variables can be reset during some
465 expansion items (e.g. "match"), thereby losing the authentication data.
467 PH/21 Make -bV show the size of off_t variables so that the test suite can
468 decide whether to run tests for quotas > 2G.
470 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
471 mailbox_size, and mailbox_filecount in the appendfile transport. If a
472 filecount value is greater than 2G or if a quota value is greater than 2G
473 on a system where the size of off_t is not greater than 4, a panic error
476 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
477 never match. The debug and -bh output now contains an explicit error
478 message indicating a malformed IPv4 address or mask.
480 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
481 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
484 PH/25 Do not write to syslog when running in the test harness. The only
485 occasion when this arises is a failure to open the main or panic logs
486 (for which there is an explicit test).
488 PH/26 Added the /no_tell option to "control=freeze".
490 PH/27 If a host name lookup failed very early in a connection, for example, if
491 the IP address matched host_lookup and the reverse lookup yielded a name
492 that did not have a forward lookup, an error message of the form "no IP
493 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
494 could be logged. Now it outputs the IP address instead of "NULL".
496 PH/28 An enabling patch from MH: add new function child_open_exim2() which
497 allows the sender and the authenticated sender to be set when
498 submitting a message from within Exim. Since child_open_exim() is
499 documented for local_scan(), the new function should be too.
501 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
502 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
503 results in an empty string is now treated as unset.
505 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
507 PH/31 Added sender_verify_fail logging option.
509 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
510 needed by RFC 822 but not by RFC 2822 was commented out. I have now
511 tidied the source and removed it altogether.
513 PH/33 When a queue run was abandoned because the load average was too high, a
514 log line was always written; now it is written only if the queue_run log
515 selector is set. In addition, the log line for abandonment now contains
516 information about the queue run such as the pid. This is always present
517 in "start" and "stop" lines but was omitted from the "abandon" line.
519 PH/34 Omit spaces between a header name and the colon in the error message that
520 is given when verify = headers_syntax fails (if there are lots of them,
521 the message gets confusing).
523 PH/35 Change the default for dns_check_names_pattern to allow slashes within
524 names, as there are now some PTR records that contain slashes. This check
525 is only to protect against broken name servers that fall over on strange
526 characters, so the fact that it applies to all lookups doesn't matter.
528 PH/36 Now that the new test suite is complete, we can remove some of the
529 special code in Exim that was needed for the old test suite. For example,
530 sorting DNS records because real resolvers return them in an arbitrary
531 order. The new test suite's fake resolver always returns records in the
534 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
535 bounces) except when queue_only is set, to avoid logging races between
536 the different processes.
538 PH/38 Panic-die if .include specifies a non-absolute path.
540 PH/39 A tweak to the "H" retry rule from its user.
542 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
543 a label. They prevented compilation on older perls.
545 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
546 a warning to be raised on newish perls.
548 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
549 on queue. Changes to match documented behaviour of showing count of
550 messages matching specified criteria.
552 PH/40 Changed the default ident timeout from 30s to 5s.
554 PH/41 Added support for the use of login_cap features, on those BSD systems
555 that have them, for controlling the resources used by pipe deliveries.
557 PH/42 The content-scanning code uses fopen() to create files in which to put
558 message data. Previously it was not paying any attention to the mode of
559 the files. Exim runs with umask(0) because the rest of the code creates
560 files with open(), and sets the required mode explicitly. Thus, these
561 files were ending up world-writeable. This was not a big issue, because,
562 being within the spool directory, they were not world-accessible. I have
563 created a function called modefopen, which takes an additional mode
564 argument. It sets umask(777), creates the file, chmods it to the required
565 mode, then resets the umask. All the relevant calls to fopen() in the
566 content scanning code have been changed to use this function.
568 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
569 to 24 hours. This avoids potential overflow problems when processing G
570 and H retry rules. I suspect nobody ever tinkers with this value.
572 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
574 PH/45 When the plaintext authenticator is running as a client, the server's
575 challenges are checked to ensure they are valid base64 strings. By
576 default, the authentication attempt is cancelled if an invalid string is
577 received. Setting client_ignore_invalid_base64 true ignores these errors.
578 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
579 they are received. Thus, the responses can be made to depend on the
580 challenges. If an invalid string is ignored, an empty string is placed in
583 PH/46 Messages that are created by the autoreply transport now contains a
584 References: header, in accordance with RFCs 2822 and 3834.
586 PH/47 Added authenticated_sender_force to the smtp transport.
588 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
590 PH/49 Installed latest patch from the Sieve maintainer.
592 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
593 was also set, the appendfile transport was unnecessarily scanning a
594 directory of message files (e.g. for maildir delivery) to find the count
595 of files (along with the size), even though it did not need this
596 information. It now does the scan only if it needs to find either the
597 size of the count of files.
599 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
601 PH/52 Two bugs concerned with error handling when the smtp transport is
604 (i) Exim was not creating retry information for temporary errors given
605 for individual recipients after the DATA command when the smtp transport
606 was used in LMTP mode. This meant that they could be retried too
607 frequently, and not timed out correctly.
609 (ii) Exim was setting the flag that allows error details to be returned
610 for LMTP errors on RCPT commands, but not for LMTP errors for individual
611 recipients that were returned after the DATA command.
613 PH/53 This is related to PH/52, but is more general: for any failing address,
614 when detailed error information was permitted to be returned to the
615 sender, but the error was temporary, then after the final timeout, only
616 "retry timeout exceeded" was returned. Now it returns the full error as
617 well as "retry timeout exceeded".
619 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
620 do this, and (what is worse) MTAs that accept it.
622 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
623 will now be deprecated.
625 PH/56 New os.c-cygwin from the Cygwin maintainer.
627 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
628 formats (previously only available in exim formats via -bpr, -bpru,
629 and -bpra. Now also available in native and exiqgrep formats)
631 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
632 with very large, slow to parse queues
634 JJ/08 exipick: added ! as generic prefix to negate any criteria format
636 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
638 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
639 responses to authentication challenges, though it was showing the
640 challenges; (ii) I've removed the CR characters from the debug output for
643 PH/58 Allow for the insertion of a newline as well as a space when a string
644 is turned into more than one encoded-word during RFC 2047 encoding. The
645 Sieve code now uses this.
647 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
648 data_4xx, lost_connection, tls_required.
650 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
651 message was being sent as an SMTP response.
653 PH/61 Add -l and -k options to exicyclog.
655 PH/62 When verifying, if an address was redirected to one new address, so that
656 verification continued, and the new address failed or deferred after
657 having set something in $address_data, the value of $address_data was not
658 passed back to the ACL. This was different to the case when no
659 redirection occurred. The value is now passed back in both cases.
661 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
662 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
663 use login_cap.h, so on its own it isn't the distinguishing feature. The
664 new name refers directly to the setclassresources() function.
666 PH/65 Added configuration files for NetBSD3.
668 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
670 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
671 is preferred over IPv4.
673 PH/68 The bounce_return_message and bounce_return_body options were not being
674 honoured for bounces generated during the reception of non-SMTP messages.
675 In particular, this applied to messages rejected by the ACL. This bug has
676 been fixed. However, if bounce_return_message is true and bounce_return_
677 body is false, the headers that are returned for a non-SMTP message
678 include only those that have been read before the error was detected.
679 (In the case of an ACL rejection, they have all been read.)
681 PH/69 The HTML version of the specification is now built in a directory called
682 spec_html instead of spec.html, because the latter looks like a path with
683 a MIME-type, and this confuses some software.
685 PH/70 Catch two compiler warnings in sieve.c.
687 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
688 function verify_get_ident() calls ip_connect() to connect a socket, but
689 if the "connect()" function timed out, ip_connect() used to close the
690 socket. However, verify_get_ident() also closes the socket later, and in
691 between Exim writes to the log, which may get opened at this point. When
692 the socket was closed in ip_connect(), the log could get the same file
693 descriptor number as the socket. This naturally causes chaos. The fix is
694 not to close the socket in ip_connect(); the socket should be closed by
695 the function that creates it. There was only one place in the code where
696 this was missing, in the iplookup router, which I don't think anybody now
697 uses, but I've fixed it anyway.
699 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
700 well as to direct DNS lookups. Otherwise the handling of names in host
701 lists is inconsistent and therefore confusing.
707 PH/01 Two changes to the default runtime configuration:
709 (1) Move the checks for relay_from_hosts and authenticated clients from
710 after to before the (commented out) DNS black list checks.
712 (2) Add control=submission to the relay_from_hosts and authenticated
713 clients checks, on the grounds that messages accepted by these
714 statements are most likely to be submissions.
716 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
718 (1) Generate an error if the third argument for the ${prvs expansion is
721 (2) Treat a missing third argument of ${prvscheck as if it were an empty
724 (3) Reset the variables that are obtained from the first argument of
725 ${prvscheck and used in the second argument before leaving the code,
726 because their memory is reclaimed, so using them afterwards may do
729 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
730 one (it's much easier than Tom thought :-).
732 (5) Because of (4), we can now allow for the use of $prvscheck_result
733 inside the third argument.
735 PH/03 For some reason, the default setting of PATH when running a command from
736 a pipe transport was just "/usr/bin". I have changed it to
739 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
740 anything to be listed in the output from -bV.
742 PH/05 When a filter generated an autoreply, the entire To: header line was
743 quoted in the delivery log line, like this:
745 => >A.N.Other <ano@some.domain> <original@ddress> ...
747 This has been changed so that it extracts the operative address. There
748 may be more than one such address. If so, they are comma-separated, like
751 => >ano@some.domain,ona@other.domain <original@ddress> ...
753 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
754 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
755 not being looked up in the rDNS to get a host name, Exim was showing the
756 IP address twice in Received: lines, even though the IP addresses were
757 identical. For example:
759 Received: from [1.2.3.4] (helo=[1.2.3.4])
761 However, if the real host name was known, it was omitting the HELO data
762 if it matched the actual IP address. This has been tidied up so that it
763 doesn't show the same IP address twice.
765 PH/07 When both +timestamp and +memory debugging was on, the value given by
766 $tod_xxx expansions could be wrong, because the tod_stamp() function was
767 called by the debug printing, thereby overwriting the timestamp buffer.
768 Debugging no longer uses the tod_stamp() function when +timestamp is set.
770 PH/08 When the original message was included in an autoreply transport, it
771 always said "this is a copy of the message, including all the headers",
772 even if body_only or headers_only was set. It now gives an appropriate
775 PH/09 Applied a patch from the Sieve maintainer which:
777 o fixes some comments
778 o adds the (disabled) notify extension core
779 o adds some debug output for the result of if/elsif tests
780 o points to the current vacation draft in the documentation
781 and documents the missing references header update
785 o fixes a bug in processing the envelope test (when testing
786 multiple envelope elements, the last element determinted the
789 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
790 Electronic Mail") by including:
792 Auto-submitted: auto-generated
794 in the messages that it generates (bounce messages and others, such as
795 warnings). In the case of bounce messages for non-SMTP mesages, there was
796 also a typo: it was using "Auto_submitted" (underscore instead of
797 hyphen). Since every message generated by Exim is necessarily in response
798 to another message, thes have all been changed to:
800 Auto-Submitted: auto-replied
802 in accordance with these statements in the RFC:
804 The auto-replied keyword:
806 - SHOULD be used on messages sent in direct response to another
807 message by an automatic process,
809 - MUST NOT be used on manually-generated messages,
811 - MAY be used on Delivery Status Notifications (DSNs) and Message
812 Disposition Notifications (MDNs),
814 - MUST NOT be used on messages generated by automatic or periodic
815 processes, except for messages which are automatic responses to
818 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
819 to the default Received: header definition.
821 PH/12 Added log selector acl_warn_skipped (default on).
823 PH/13 After a successful wildlsearch lookup, discard the values of numeric
824 variables because (a) they are in the wrong storage pool and (b) even if
825 they were copied, it wouldn't work properly because of the caching.
827 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
828 checking when decoding. Apparently there are clients that generate
829 overlong encoded strings. Why am I not surprised?
831 PH/15 If the first argument of "${if match_address" was not empty, but did not
832 contain an "@" character, Exim crashed. Now it writes a panic log message
833 and treats the condition as false.
835 PH/16 In autoreply, treat an empty string for "once" the same as unset.
837 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
838 extension "envelope-auth". The code is finished and in agreement with
839 other implementations, but there is no documentation so far and in fact,
840 nobody wrote the draft yet. This extension is currently #undef'ed, thus
841 not changing the active code.
843 Print executed "if" and "elsif" statements when debugging is used. This
844 helps a great deal to understand what a filter does.
846 Document more things not specified clearly in RFC3028. I had all this
847 sorted out, when out of a sudden new issues came to my mind. Oops."
849 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
852 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
853 canonical form (as documented). However, after a host name lookup from
854 the IP address, check_host() was doing a simple string comparison with
855 addresses acquired from the DNS when checking that the found name did
856 have the original IP as one of its addresses. Since any found IPv6
857 addresses are likely to be in abbreviated form, the comparison could
858 fail. Luckily, there already exists a function for doing the comparison
859 by converting both addresses to binary, so now that is used instead of
862 PH/20 There was another similar case to PH/19, when a complete host name was
863 given in a host list; looking up its IP address could give an abbreviated
864 form, whereas the current host's name might or might not be abbreviated.
865 The same fix has been applied.
871 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
872 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
875 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
876 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
878 PH/03 Typo: missing ".o" in src/pcre/Makefile.
880 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
881 header line, restrict the check to what is listed in RFCs 2369 and 2929.
882 Also, for "Auto-Submitted", treat anything other than "no" as
883 non-personal, in accordance with RFC 3834. (Previously it treated
884 anything starting "auto-" as non-personal.)
886 TF/01 The control=submission/name=... option had a problem with syntax
887 errors if the name included a slash character. The /name= option
888 now slurps the rest of the string, so it can include any characters
889 but it must come last in the list of options (after /sender_retain
892 PH/05 Some modifications to the interface to the fake nameserver for the new
900 TK/01 Added the "success_on_redirect" address verification option. See
901 NewStuff for rationale and an example.
903 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
905 PH/02 Patch to exigrep to allow it to work on syslog lines.
907 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
908 fread() to skip over the body file's header line, because in Cygwin the
909 header line is locked and is inaccessible.
911 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
912 co-exist for some time) to make it clear that it is the Exim ID that is
913 referenced, not the Message-ID: header line.
915 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
916 string_format() function, because snprintf() does not exist on all
919 PH/06 The use of forbid_filter_existstest now also locks out the use of the
920 ${stat: expansion item.
922 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
923 protocol synchronization error", to keep the pedants happy.
925 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
926 well as for IRIX systems, when gcc is being used. See the host.c source
929 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
931 PH/10 Named domain lists were not working if used in a queue_smtp_domains
934 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
935 transport and to the smtp transport in LMTP mode.
937 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
939 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
940 run a filter in a subprocess. This could lead to confusion in subsequent
941 lookups in the parent process. There should also be a search_tidyup() at
942 the end of the subprocess.
944 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
945 only if the host matched helo_try_verify_hosts, which caused the
946 verification to occur when the EHLO/HELO command was issued. The ACL just
947 tested the remembered result. Now, if a previous verification attempt has
948 not happened, "verify = helo" does it there and then.
950 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
952 TK/03 Fix log output including CR from clamd.
954 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
955 exist provoked a memory error which could cause a segfault.
957 PH/15 Installed PCRE 6.2
959 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
961 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
962 of the problem. Specifically, suggested +O2 rather than +O1 for the
965 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
967 PH/20 If a delivery was routed to a non-standard port by means of an SRV
968 record, the port was not correctly logged when the outgoing_port log
969 selector was set (it logged the transort's default port).
971 PH/21 Added support for host-specific ports to manualroute, queryprogram,
972 fallback_hosts, and "hosts" in the smtp transport.
974 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
975 host errors such as "Connection refused".
977 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
978 authentication with radiusclient 0.4.9:
980 - Error returned from rc_read_config was caught wrongly
981 - Username/password not passed on to radius server due to wrong length.
983 The presumption is that some radiusclient API changes for 4.51/PH/17
984 were not taken care of correctly. The code is still untested by me (my
985 Linux distribution still has 0.3.2 of radiusclient), but it was
986 contributed by a Radius user.
988 PH/24 When doing a callout, the value of $domain wasn't set correctly when
989 expanding the "port" option of the smtp transport.
991 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
992 while reading a MIME header. Thanks to Tom Hughes for a patch.
994 PH/24 Include config.h inside local_scan.h so that configuration settings are
997 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
998 that in an ACL for RCPT (for example), you can examine exactly what was
1001 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1002 commands, but it was not correctly comparing the address with the actual
1003 client host address. Thus, it would show the EHLO address in Received:
1004 header lines when this was not necessary.
1006 PH/27 Added the % operator to ${eval:}.
1008 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1009 it should be ignoring failures (because with -C, for example, it has lost
1010 privilege). It wasn't ignoring creation failures other than "already
1013 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1016 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1017 because an input line was too long, either on its own, or by virtue of
1018 too many continuations, the temporary file was not being removed, and the
1019 return code was incorrect.
1021 PH/31 Missing "BOOL" in function definition in filtertest.c.
1023 PH/32 Applied Sieve patches from the maintainer.
1025 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1027 PH/33 Added "verify = not_blind".
1029 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1030 Local/Makefile (with some defaults set). These are used in built scripts
1031 such as exicyclog, but they have never been used in the exim_install
1032 script (though there are many overriding facilities there). I have
1033 arranged that the exim_install script now takes note of these two
1036 PH/35 Installed configuration files for Dragonfly.
1038 PH/36 When a locally submitted message by a trusted user did not contain a
1039 From: header, and the sender address was obtained from -f or from an SMTP
1040 MAIL command, and the trusted user did not use -F to supply a sender
1041 name, $originator_name was incorrectly used when constructing a From:
1042 header. Furthermore, $originator_name was used for submission mode
1043 messages from external hosts without From: headers in a similar way,
1044 which is clearly wrong.
1046 PH/37 Added control=suppress_local_fixups.
1048 PH/38 When log_selector = +received_sender was set, and the addition of the
1049 sender made the log line's construction buffer exactly full, or one byte
1050 less than full, an overflow happened when the terminating "\n" was
1053 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1054 when the result of a list match is failure because a DNS lookup failed.
1056 PH/40 RM_COMMAND is now used in the building process.
1058 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1059 the "build-* directories that it finds.
1061 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1062 address in a domain literal was a prefix of an interface address.
1064 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1065 when verifying a sender address, unless rewrite_headers is false.
1067 PH/44 Wrote a long comment about why errors_to addresses are verified as
1068 recipients, not senders.
1070 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1071 the ratelimit ACL was added.
1073 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1075 PH/46 Added extra information about PostgreSQL errors to the error string.
1077 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1078 avoiding the need to install special zones in a real server. This is
1079 backwards compatible; if it can't find the fake resolver, it drops back.
1080 Thus, both old and new test suites can be run.
1082 TF/02 Added util/ratelimit.pl
1084 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1085 clock is set back in time.
1087 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1088 Candler <B.Candler@pobox.com>.
1090 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1091 OK for addresses that are the result of redirections.
1093 PH/48 A number of further additions for the benefit of the new test suite,
1094 including a fake gethostbyname() that interfaces to the fake DNS resolver
1097 TF/06 The fix for widen_domains has also been applied to qualify_single and
1098 search_parents which are the other dnslookup options that can cause
1101 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1104 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1106 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1107 service" if it encountered an MX record with an empty target hostname.
1108 The message is now "an MX or SRV record indicated no SMTP service".
1110 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1111 if the DNS of the sending site is misconfigured. This is quite a
1112 common situation. This change restores the behaviour of treating a
1113 helo verification defer as a failure.
1115 PH/51 If self=fail was set on a router, the bounce message did not include the
1116 actual error message.
1122 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1124 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1125 command itself ended in error, the underlying message about the transport
1126 filter timeout was being overwritten with the pipe command error. Now the
1127 underlying error message should be appended to the second error message.
1129 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1131 PH/02 Reduce the amount of output that "make" produces by default. Full output
1132 can still be requested.
1134 PH/03 The warning log line about a condition test deferring for a "warn" verb
1135 was being output only once per connection, rather than after each
1136 occurrence (because it was using the same function as for successful
1137 "warn" verbs). This seems wrong, so I have changed it.
1139 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1140 it should not have, which might have caused a crash in the right
1141 circumstances, but probably never did.
1143 PH/04 Installed a modified version of Tony Finch's patch to make submission
1144 mode fix the return path as well as the Sender: header line, and to
1145 add a /name= option so that you can make the user's friendly name appear
1148 TF/03 Added the control = fakedefer ACL modifier.
1150 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1151 Mark Lowes for thorough testing.
1153 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1155 TK/03 Merged latest SRS patch from Miles Wilton.
1157 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1158 with the definition in sysexits.h (which is #included earlier).
1159 Fortunately, Exim does not actually use EX_OK. The code used to try to
1160 preserve the sysexits.h value, by assumimg that macro definitions were
1161 scanned for macro replacements. I have been disabused of this notion,
1162 so now the code just undefines EX_OK before #including unistd.h.
1164 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1165 in the smtp transport. When a block could not be written in a single
1166 write() function, the timeout was being re-applied to each part-write.
1167 This seems wrong - if the receiver was accepting one byte at a time it
1168 would take for ever. The timeout is now adjusted when this happens. It
1169 doesn't have to be particularly precise.
1171 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1172 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1174 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1175 without a domain if the check to <postmaster@domain> fails.
1177 SC/01 Eximstats: added -xls and the ability to specify output files
1178 (patch written by Frank Heydlauf).
1180 SC/02 Eximstats: use FileHandles for outputing results.
1182 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1184 SC/04 Eximstats: fixed display of large numbers with -nvr option
1186 SC/05 Eximstats: fixed merging of reports with empty tables.
1188 SC/06 Eximstats: added the -include_original_destination flag
1190 SC/07 Eximstats: removed tabs and trailing whitespace.
1192 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1194 TK/06 MBOX spool code: Add real "From " MBOX separator line
1195 so the .eml file is really in mbox format (even though
1196 most programs do not really care). Patch from Alex Miller.
1198 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1199 The latter is generated from $received_to and is only set if the
1200 message has one envelope recipient. SA can use these headers,
1201 obviously out-of-the-box. Patch from Alex Miller.
1203 PH/08 The ${def test on a variable was returning false if the variable's
1204 value was "0", contrary to what the specification has always said!
1205 The result should be true unless the variable is empty.
1207 PH/09 The syntax error of a character other than { following "${if
1208 def:variable_name" (after optional whitespace) was not being diagnosed.
1209 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1210 accidental colon was present, for example, could give incorrect results.
1212 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1213 result is used (not including appendfile, where other changes are about
1216 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1217 This involved changing a lot of size variables from int to off_t. It
1218 should work with maildirs and everything.
1220 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1221 spamd dying while we are connected to it.
1223 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1226 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1227 function for BDB 4.3.
1229 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1231 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1232 the sources for PCRE have been re-arranged and more files are now
1235 PH/15 The code I had for printing potentially long long variables in PH/11
1236 above was not the best (it lost precision). The length of off_t variables
1237 is now inspected at build time, and an appropriate printing format (%ld
1238 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1239 to be "long long int" or "long int". This is needed for the internal
1240 formatting function string_vformat().
1242 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1243 the configuration file to be ":syslog", then the script "guesses" where
1244 the logs files are, rather than using the compiled in default. In our
1245 case the guess is not the same as the compiled default, so the script
1246 suddenly stopped working when I started to use syslog. The patch checks
1247 to see if log_file_path is "". If so, it attempts to read it from exim
1248 with no configuration file to get the compiled in version, before it
1249 falls back to the previous guessing code."
1251 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1252 implementing BATV in an Exim configuration. See NewStuff for the gory
1255 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1256 Makefile that are specific to HP-UX.
1258 PH/18 If the "use_postmaster" option was set for a recipient callout together
1259 with the "random" option, the postmaster address was used as the MAIL
1260 FROM address for the random test, but not for the subsequent recipient
1261 test. It is now used for both.
1263 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1264 patch removes a few documentation additions to RFC 3028, because the
1265 latest draft now contains them. It adds the new en;ascii-case comparator
1266 and a new error check for 8bit text in MIME parts. Comparator and
1267 require names are now matched exactly. I enabled the subaddress
1268 extension, but it is not well tested yet (read: it works for me)."
1270 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1271 rework some of the code of TK/09 above to avoid the hardwired use of
1272 "%lld" and "long long". Replaced the call to snprintf() with a call to
1275 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1276 records point to non-existent hosts", "retry timeout exceeded", and
1277 "retry time not reached for any host after a long failure period".
1279 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1280 experimental DomainKeys support:
1282 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1283 (2) On an error such as an illegally used "control", the wrong name for
1284 the control was given.
1286 These problems did NOT occur unless DomainKeys support was compiled.
1288 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1290 PH/24 Added ${if match_ip condition.
1292 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1293 databases so that it will be absolutely obvious if a crash occurs in the
1294 DB library. This is a regular occurrence (often caused by mis-matched
1297 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1298 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1299 picked up on a user's system that detects such things. There doesn't seem
1300 to be a gcc warning option for this - only an attribute that has to be
1301 put on the function's prototype. It seems that in Fedora Core 4 they have
1302 set this on a number of new functions. No doubt there will be more in due
1305 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1306 specify a transport. However, if an address that was verified by such a
1307 router was the subject of a callout, Exim crashed because it tried to
1308 read the rcpt_include_affixes from the non-existent transport. Now it
1309 just assumes that the setting of that option is false. This bug was
1310 introduced by 4.51/PH/31.
1312 PH/28 Changed -d+all to exclude +memory, because that information is very
1313 rarely of interest, but it makes the output a lot bigger. People tend to
1314 do -d+all out of habit.
1316 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1317 code in os-type was giving problems when libc.so lives in lib64, like on
1320 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1321 aren't the modern standard, and it seems that some systems' include files
1322 don't always have them. Exim was already checking for some of the newer
1323 ones like T_AAAA, and defining it itself. I've added checks for all the
1324 record types that Exim uses.
1326 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1327 not automatically generating a new one, as it is supposed to. This
1328 prevented TLS from working. If the file did exist, but contained invalid
1329 data, a new version was generated, as expected. It was only the case of a
1330 non-existent file that was broken.
1332 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1333 with a change in libdomainkeys > 0.64.
1335 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1336 from DNS. If the selector record carries the flag, it now has
1337 precedence over the domain-wide flag.
1339 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1341 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1342 the use of a port name that isn't defined in /etc/services) occurred, the
1343 message was deferred as in a normal delivery, and thus remained on the
1344 spool, instead of being failed because of the mua_wrapper setting. This
1345 is now fixed, and I tidied up some of the mua_wrapper messages at the
1348 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1349 the messages in a hash of arrays rather than using individual hashes.
1350 This is a bit cleaner and results in dramatic memory savings, albeit
1351 at a slight CPU cost.
1353 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1354 as requested by Marc Sherman.
1356 SC/10 Eximstats: added histograms for user specified patterns as requested
1359 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1361 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1362 fopen() in the content-scanning modules that did not already have it.
1368 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1369 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1371 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1373 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1375 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1376 to test Sieve filters that use "vacation".
1378 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1379 that changes the way the GnuTLS parameters are stored in the cache file.
1380 The new format can be generated externally. For backward compatibility,
1381 if the data in the cache doesn't make sense, Exim assumes it has read an
1382 old-format file, and it generates new data and writes a new file. This
1383 means that you can't go back to an older release without removing the
1386 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1387 work if there are any delivery delays because "one_time" forces the
1388 parent to be marked "delivered", so its unseen clone is never tried
1389 again. For this reason, Exim now forbids the simultaneous setting of
1392 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1393 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1394 entry if you want to know the details. The fix, however, neglected to
1395 consider the case when local delivery batching is involved. The test for
1396 "previously delivered" was not happening when checking to see if an
1397 address could be batched with a previous (undelivered) one; under
1398 certain circumstances this could lead to multiple deliveries to the same
1401 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1402 in its include files, and this causes problems building Exim.
1404 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1405 header_syntax) but Exim was just ignoring anything given after a slash.
1406 In particular, this caused confusion with an attempt to use "verify =
1407 reverse_host_lookup/defer_ok". An error is now given when options are
1408 supplied for verify items that do not have them. (Maybe reverse_host_
1409 lookup should have a defer_ok option, but that's a different point.)
1411 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1412 defined by RFC 821) to 2048, because there were problems with some AUTH
1413 commands, and RFC 1869 says the size should be increased for extended
1414 SMTP commands that take arguments.
1416 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1419 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1420 "unknown" error; now it says that the functionality isn't in the binary.
1422 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1423 an address' error message when a string expansion fails (syntax or
1424 whatever). Otherwise the password may appear in the log. Following change
1425 PH/42 below, there is no longer a chance of it appearing in a bounce
1428 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1430 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1431 of its addresses was ever tried. (Bugzilla bug #2.)
1433 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1434 the result incorrectly in the debug output. (It correctly added a newline
1435 to what was transported.)
1437 TF/01 Added $received_time.
1439 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1440 commented out examples of how to interface to a virus scanner and to
1441 SpamAssassin. Also added commented examples of av_scanner and
1442 spamd_address settings.
1444 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1445 and controls are allowed in which ACLs. There were a couple of minor
1446 errors. Some of the entries in the conditions table (which is a table of
1447 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1448 negation of where the condition IS allowed.
1450 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1452 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1453 header file does not have a version number, so I've had to invent a new
1454 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1455 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1456 radiusclient), but it was contributed by a Radius user.
1458 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1459 files or directories, for OpenSSL.
1461 PH/19 When an Exim process that is running as root has to create an Exim log
1462 file, it does so in a subprocess that runs as exim:exim so as to get the
1463 ownership right at creation (otherwise, other Exim processes might see
1464 the file with the wrong ownership). There was no test for failure of this
1465 fork() call, which would lead to the process getting stuck as it waited
1466 for a non-existent subprocess. Forks do occasionally fail when resources
1467 run out. I reviewed all the other calls to fork(); they all seem to check
1470 PH/20 When checking for unexpected SMTP input at connect time (before writing
1471 the banner), Exim was not dealing correctly with a non-positive return
1472 from the read() function. If the client had disconnected by this time,
1473 the result was a log entry for a synchronization error with an empty
1474 string after "input=" when read() returned zero. If read() returned -1
1475 (an event I could not check), uninitialized data bytes were printed.
1476 There were reports of junk text (parts of files, etc) appearing after
1479 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1481 PH/22 Added support for macro redefinition, and (re)definition in between
1482 driver and ACL definitions.
1484 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1485 forgetting to use the resulting value; it was using the unexpanded value.
1487 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1488 hadn't been configured. The fix is from Juergen Kreileder, who
1489 understands it better than I do:
1491 "Here's what I see happening with three configured cyrus_sasl
1492 authenticators configured (plain, login, cram-md5):
1494 On startup auth_cyrus_sasl_init() gets called for each of these.
1495 This means three calls to sasl_listmech() without a specified mech_list.
1496 => SASL tests which mechs of all available mechs actually work
1497 => three warnings about OTP not working
1498 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1500 With the patch, sasl_listmech() also gets called three times. But now
1501 SASL's mech_list option is set to the server_mech specified in the the
1502 authenticator. Or in other words, the answer from sasl_listmech()
1503 gets limited to just the mech you're testing for (which is different
1505 => the return list contains just 'plain' or 'login', 'cram-md5' or
1506 nothing depending on the value of ob->server_mech.
1508 I've just tested the patch: Authentication still works fine,
1509 unavailable mechs specified in the exim configuration are still
1510 caught, and the auth.log warnings about OTP are gone."
1512 PH/25 When debugging is enabled, the contents of the command line are added
1513 to the debugging output, even when log_selector=+arguments is not
1516 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1517 answer is "GNU", and only if the return is "GNU/something" is the answer
1520 PH/27 $acl_verify_message is now set immediately after the failure of a
1521 verification in an ACL, and so is available in subsequent modifiers. In
1522 particular, the message can be preserved by coding like this:
1524 warn !verify = sender
1525 set acl_m0 = $acl_verify_message
1527 Previously, $acl_verify_message was set only while expanding "message"
1528 and "log_message" when a very denied access.
1530 PH/28 Modified OS/os.c-Linux with
1532 -#ifndef OS_LOAD_AVERAGE
1533 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1535 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1536 nomenclature these days.)
1538 PH/29 Installed patch from the Sieve maintainer that adds the options
1539 sieve_useraddress and sieve_subaddress to the redirect router.
1541 PH/30 In these circumstances:
1542 . Two addresses routed to the same list of hosts;
1543 . First host does not offer TLS;
1544 . First host accepts first address;
1545 . First host gives temporary error to second address;
1546 . Second host offers TLS and a TLS session is established;
1547 . Second host accepts second address.
1548 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1549 and peerdn, if requested) that were in fact used only for the second
1552 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1553 attention to any local part prefix or suffix that was matched by the
1554 router that accepted the address. It now behaves in the same way as it
1555 does for delivery: the affixes are removed from the local part unless
1556 rcpt_include_affixes is set on the transport.
1558 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1559 timeout during the DATA phase of an incoming message.
1561 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1562 applied a patch sanctioned by the Sieve maintainer.
1564 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1565 the uid or gid is negative. A case of a negative gid caused this to be
1566 noticed. The fix allows for either to be negative.
1568 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1569 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1572 PH/36 The change PH/12 above was broken. Fixed it.
1574 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1575 the grounds that routing the same address twice would always produce the
1576 same answer. This might have been true once, but it is certainly no
1577 longer true now. Routing a child address may depend on the previous
1578 routing that produced that child. Some complicated redirection strategies
1579 went wrong when messages had multiple recipients, and made Exim's
1580 behaviour dependent on the order in which the addresses were given.
1582 I have moved the duplicate checking until after the routing is complete.
1583 Exim scans the addresses that are assigned to local and remote
1584 transports, and removes any duplicates. This means that more work will be
1585 done, as duplicates will always all be routed, but duplicates are
1586 presumably rare, so I don't expect this is of any significance.
1588 For deliveries to pipes, files, and autoreplies, the duplicate checking
1589 still happens during the routing process, since they are not going to be
1592 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1593 It corrects a timeout issue with spamd. This is Ian's comment: "The
1594 background is that sometimes spamd either never reads data from a
1595 connection it has accepted, or it never writes response data. The exiscan
1596 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1597 blindly assumes that writes won't block so it may never time out."
1599 PH/39 Allow G after quota size as well as K and M.
1601 PH/40 The value set for $authenticated_id in an authenticator may not contain
1602 binary zeroes or newlines because the value is written to log lines and
1603 to spool files. There was no check on this. Now the value is run through
1604 the string_printing() function so that such characters are converted to
1605 printable escape sequences.
1607 PH/41 $message_linecount is a new variable that contains the total number of
1608 lines in the message. Compare $body_linecount, which is the count for the
1611 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1612 bounce and delay warning messages, except in certain special cases, which
1615 (a) An SMTP error message from a remote host;
1616 (b) A message specified in a :fail: redirection;
1617 (c) A message specified in a "fail" command in a system filter;
1618 (d) A message specified in a FAIL return from the queryprogram router;
1619 (e) A message specified by the cannot_route_message router option.
1621 In these cases only, Exim does include the error details in bounce and
1622 warning messages. There are also a few cases where bland messages such
1623 as "unrouteable address" or "local delivery error" are given.
1625 PH/43 $value is now also set for the "else" part of a ${run expansion.
1627 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1628 being worked on, but at least Exim now implements the latest version to
1631 PH/45 In a pipe transport, although a timeout while waiting for the pipe
1632 process to complete was treated as a delivery failure, a timeout while
1633 writing the message to the pipe was logged, but erroneously treated as a
1634 successful delivery. Such timeouts include transport filter timeouts. For
1635 consistency with the overall process timeout, these timeouts are now
1636 treated as errors, giving rise to delivery failures by default. However,
1637 there is now a new Boolean option for the pipe transport called
1638 timeout_defer, which, if set TRUE, converts the failures into defers for
1639 both kinds of timeout. A transport filter timeout is now identified in
1642 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1643 systems where "make" and "gmake" are different, calling "gmake" at top
1644 level broke things. I've arranged for the value of $(MAKE) to be passed
1645 from the Makefile to this script so that it can call the same version of
1649 A note about Exim versions 4.44 and 4.50
1650 ----------------------------------------
1652 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1653 changes of various kinds. As a consequence, a big documentation update was
1654 needed. This delayed the release for rather longer than seemed good, especially
1655 in the light of a couple of (minor) security issues. Therefore, the changes
1656 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1657 release. So 4.44 and 4.50 are in effect two different branches that both start
1660 I have left the 4.50 change log unchanged; it contains all the changes since
1661 4.43. The change log for 4.44 is below; many of its items are identical to
1662 those for 4.50. This seems to be the most sensible way to preserve the
1663 historical information.
1669 1. Minor wording change to the doc/README.SIEVE file.
1671 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1672 computation of the current number of files was incorrect.
1674 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1675 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1678 4. Give more explanation in the error message when the command for a transport
1679 filter fails to execute.
1681 5. There are several places where Exim runs a non-Exim command in a
1682 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1683 was being done only for the command run by the queryprogram router. It is
1684 now done for all such subprocesses. The other cases are: ${run, transport
1685 filters, and the commands run by the lmtp and pipe transports.
1687 6. Added CONFIGURE_GROUP build-time option.
1689 7. Some older OS have a limit of 256 on the maximum number of file
1690 descriptors. Exim was using setrlimit() to set 1000 as a large value
1691 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1692 systems. I've change it so that if it can't get 1000, it tries for 256.
1694 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1695 was an oversight, and furthermore, ever since the addition of extra
1696 controls (e.g. 4.43/32), the checks on when to allow different forms of
1697 "control" were broken. There should now be diagnostics for all cases when a
1698 control that does not make sense is encountered.
1700 9. Added the /retain_sender option to "control=submission".
1702 10. $recipients is now available in the predata ACL (oversight).
1704 11. Tidy the search cache before the fork to do a delivery from a message
1705 received from the command line. Otherwise the child will trigger a lookup
1706 failure and thereby defer the delivery if it tries to use (for example) a
1707 cached ldap connection that the parent has called unbind on.
1709 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1710 of $address_data from the recipient verification was clobbered by the
1711 sender verification.
1713 13. The value of address_data from a sender verification is now available in
1714 $sender_address_data in subsequent conditions in the ACL statement.
1716 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1718 15. Added a new option "connect=<time>" to callout options, to set a different
1721 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1722 was its contents. (It was OK if the option was not defined at all.)
1724 17. A "Completed" log line is now written for messages that are removed from
1725 the spool by the -Mrm option.
1727 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1728 information about exactly what failed.
1730 19. Added -dd to debug only the daemon process.
1732 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1733 handles timeouts, both on the server side and network timeouts. Renamed the
1734 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1736 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1737 It is now set to "smtps".
1739 22. $host_address is now set to the target address during the checking of
1740 ignore_target_hosts.
1742 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1743 being passed; this would have caused $sender_host_name to have been used if
1744 matching the list had actually called for a host name (not very likely,
1745 since this list is usually IP addresses). A host name is now passed as
1748 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1749 code that specifies a non-restarting handler (typically sigaction() in
1750 modern systems) in an attempt to fix a rare and obscure crash bug.
1752 25. Narrowed the window for a race in the daemon that could cause it to ignore
1753 SIGCHLD signals. This is not a major problem, because they are used only to
1754 wake it up if nothing else does.
1756 26. A malformed maildirsize file could cause Exim to calculate negative values
1757 for the mailbox size or file count. Odd effects could occur as a result.
1758 The maildirsize information is now recalculated if the size or filecount
1761 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1762 support for a long time. Removed HAVE_SYS_VFS_H.
1764 28. Installed the latest version of exipick from John Jetmore.
1766 29. In an address list, if the pattern was not a regular expression, an empty
1767 subject address (from a bounce message) matched only if the pattern was an
1768 empty string. Non-empty patterns were not even tested. This was the wrong
1769 because it is perfectly reasonable to use an empty address as part of a
1770 database query. An empty address is now tested by patterns that are
1771 lookups. However, all the other forms of pattern expect the subject to
1772 contain a local part and a domain, and therefore, for them, an empty
1773 address still always fails if the pattern is not itself empty.
1775 30. Exim went into a mad DNS loop when attempting to do a callout where the
1776 host was specified on an smtp transport, and looking it up yielded more
1777 than one IP address.
1779 31. Re-factored the code for checking spool and log partition space into a
1780 function that finds that data and another that does the check. The former
1781 is then used to implement four new variables: $spool_space, $log_space,
1782 $spool_inodes, and $log_inodes.
1784 32. The RFC2047 encoding function was originally intended for short strings
1785 such as real names; it was not keeping to the 75-character limit for
1786 encoded words that the RFC imposes. It now respects the limit, and
1787 generates multiple encoded words if necessary. To be on the safe side, I
1788 have increased the buffer size for the ${rfc2047: expansion operator from
1791 33. It is now permitted to omit both strings after an "if" condition; if the
1792 condition is true, the result is "true". As before, when the second string
1793 is omitted, a false condition yields an empty string. This makes it less
1794 cumbersome to write custom ACL and router conditions.
1796 34. Failure to deliver a bounce message always caused it to be frozen, even if
1797 there was an errors_to setting on the router. The errors_to setting is now
1800 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1801 canonical form (fully expanded) before being placed in
1802 $sender_host_address.
1804 36. The table in the code that translates DNS record types into text (T_A to
1805 "A" for instance) was missing entries for NS and CNAME. It is just possible
1806 that this could have caused confusion if both these types were looked up
1807 for the same domain, because the text type is used as part of Exim's
1808 per-process caching. But the chance of anyone hitting this buglet seems
1811 37. The dnsdb lookup has been extended in a number of ways.
1813 (1) There is a new type, "zns", which walks up the domain tree until it
1814 finds some nameserver records. It should be used with care.
1816 (2) There is a new type, "mxh", which is like "mx" except that it returns
1817 just the host names, not the priorities.
1819 (3) It is now possible to give a list of domains (or IP addresses) to be
1820 looked up. The behaviour when one of the lookups defers can be
1821 controlled by a keyword.
1823 (4) It is now possible to specify the separator character for use when
1824 multiple records are returned.
1826 38. The dnslists ACL condition has been extended: it is now possible to supply
1827 a list of IP addresses and/or domains to be looked up in a particular DNS
1830 39. Added log_selector=+queue_time_overall.
1832 40. When running the queue in the test harness, wait just a tad after forking a
1833 delivery process, to get repeatability of debugging output.
1835 41. Include certificate and key file names in error message when GnuTLS fails
1836 to set them up, because the GnuTLS error message doesn't include the name
1837 of the failing file when there is a problem reading it.
1839 42. Allow both -bf and -bF in the same test run.
1841 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1843 44. The "Exiscan patch" is now merged into the mainline Exim source.
1845 45. Sometimes the final signoff response after QUIT could fail to get
1846 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1847 < 0 before doing a fflush(). This bug looks as though it goes back to the
1848 introduction of TLS in release 3.20, but "sometimes" must have been rare
1849 because the tests only now provoked it.
1851 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1852 (this can affect the format of dates).
1854 47. exim_tidydb, when checking for the continued existence of a message for
1855 which it has found a message-specific retry record, was not finding
1856 messages that were in split spool directories. Consequently, it was
1857 deleting retry records that should have stayed in existence.
1859 48. Steve fixed some bugs in eximstats.
1861 49. The SPA authentication driver was not abandoning authentication and moving
1862 on to the next authenticator when an expansion was forced to fail,
1863 contradicting the general specification for all authenticators. Instead it
1864 was generating a temporary error. It now behaves as specified.
1866 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1867 (the order specifies the preference for clients). The order is now AES256,
1868 AES128, 3DES, ARCFOUR128.
1870 51. Small patch to Sieve code - explicitly set From: when generating an
1873 52. Exim crashed if a remote delivery caused a very long error message to be
1874 recorded - for instance if somebody sent an entire SpamAssassin report back
1875 as a large number of 550 error lines. This bug was coincidentally fixed by
1876 increasing the size of one of Exim's internal buffers (big_buffer) that
1877 happened as part of the Exiscan merge. However, to be on the safe side, I
1878 have made the code more robust (and fixed the comments that describe what
1881 53. Now that there can be additional text after "Completed" in log lines (if
1882 the queue_time_overall log selector is set), a one-byte patch to exigrep
1883 was needed to allow it to recognize "Completed" as not the last thing in
1886 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1887 patch that reportedly fixes this has been added. I am not expert enough to
1888 create a test for it. This is what the patch creator wrote:
1890 "I found a little strange behaviour of ldap code when working with
1891 Windows 2003 AD Domain, where users was placed in more than one
1892 Organization Units. When I tried to give exim partial DN, the exit code
1893 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1894 But simultaneously result of request was absolutely normal ldap result,
1895 so I produce this patch..."
1897 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1898 REFERENCE, so I have modified the code to exclude the patch when that macro
1901 55. Some experimental protocols are using DNS PTR records for new purposes. The
1902 keys for these records are domain names, not reversed IP addresses. The
1903 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1904 leaves it alone. Component reversal etc. now happens only for IP addresses.
1907 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1909 57. Double the size of the debug message buffer (to 2048) so that more of very
1910 long debug lines gets shown.
1912 58. The exicyclog utility now does better if the number of log files to keep
1913 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1915 59. Two changes related to the smtp_active_hostname option:
1917 (1) $smtp_active_hostname is now available as a variable.
1918 (2) The default for smtp_banner uses $smtp_active_hostname instead
1919 of $primary_hostname.
1921 60. The host_aton() function is supposed to be passed a string that is known
1922 to be a valid IP address. However, in the case of IPv6 addresses, it was
1923 not checking this. This is a hostage to fortune. Exim now panics and dies
1924 if the condition is not met. A case was found where this could be provoked
1925 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1926 components; fortuitously, this particular loophole had already been fixed
1927 by change 4.50/55 above.
1929 If there are any other similar loopholes, the new check in host_aton()
1930 itself should stop them being exploited. The report I received stated that
1931 data on the command line could provoke the exploit when Exim was running as
1932 exim, but did not say which command line option was involved. All I could
1933 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1934 running as the user.
1937 61. There was a buffer overflow vulnerability in the SPA authentication code
1938 (which came originally from the Samba project). I have added a test to the
1939 spa_base64_to_bits() function which I hope fixes it.
1942 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1943 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1945 63. The daemon start-up calls getloadavg() while still root for those OS that
1946 need the first call to be done as root, but it missed one case: when
1947 deliver_queue_load_max is set with deliver_drop_privilege. This is
1948 necessary for the benefit of the queue runner, because there is no re-exec
1949 when deliver_drop_privilege is set.
1951 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1952 This has been fixed.
1954 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1955 was in use, was not putting the data itself into the right store pool;
1956 consequently, it could be overwritten for a subsequent message in the same
1957 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1960 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1962 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1963 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1964 the function were treating the return as a boolean value, which happened to
1965 work because 0=false and not-0=true, but is not correct code.
1967 68. The host_aton() function was not handling scoped IPv6 addresses (those
1968 with, for example, "%eth0" on the end) correctly.
1970 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1971 negated items (that is, ~something) in unsigned ints. Some compilers
1972 apparently mutter when there is no cast.
1974 70. If an address verification called from an ACL failed, and did not produce a
1975 user-specific message (i.e. there was only a "system" message), nothing was
1976 put in $acl_verify_message. In this situation, it now puts the system
1979 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1980 session; change 4.31/43 added the unwanted input to the log line - except
1981 that it did not do this in the start of session case. It now does.
1983 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1984 This is wrong and can cause the other end to generate a synchronization
1985 error if it is another Exim or anything else that does the synchronization
1986 check. A QUIT command is no longer sent after a timeout.
1988 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1989 during host lookups.
1991 74. The defer_ok option of callout verification was not working if it was used
1992 when verifying addresses in header lines, that is, for this case:
1994 verify = header_sender/callout=defer_ok
1996 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1997 those file descriptors could be used for SMTP connections. If anything
1998 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1999 could be sent to the SMTP client, causing chaos. The daemon now opens
2000 stdin, stdout, and stderr to /dev/null when it puts itself into the
2003 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2004 log by default. The user can override this with suitable Perl magic.
2006 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2007 the log message when discard triggers, was not working for the DATA ACL or
2008 for the non-SMTP ACL.
2010 78. Error message wording change in sieve.c.
2012 79. If smtp_accept_max_per_host was set, the number of connections could be
2013 restricted to fewer than expected, because the daemon was trying to set up
2014 a new connection before checking whether the processes handling previous
2015 connections had finished. The check for completed processes is now done
2016 earlier. On busy systems, this bug wouldn't be noticed because something
2017 else would have woken the daemon, and it would have reaped the completed
2020 80. If a message was submitted locally by a user whose login name contained one
2021 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2022 It caused a spool format error. I have fixed the spool reading code. A
2023 related problem was that the "from" clause in the Received: line became
2024 illegal because of the space(s). It is now covered by ${quote_local_part.
2026 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2029 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2031 83. Patch from Sieve maintainer to fix unterminated string problem in
2032 "vacation" handling.
2034 84. Some minor changes to the Linux configuration files to help with other
2035 OS variants using glibc.
2037 85. One more patch for Sieve to update vacation handling to latest spec.
2040 ----------------------------------------------------
2041 See the note above about the 4.44 and 4.50 releases.
2042 ----------------------------------------------------
2048 1. Change 4.43/35 introduced a bug that caused file counts to be
2049 incorrectly computed when quota_filecount was set in an appendfile
2052 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2053 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2056 3. Give more explanation in the error message when the command for a transport
2057 filter fails to execute.
2059 4. There are several places where Exim runs a non-Exim command in a
2060 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2061 was being done only for the command run by the queryprogram router. It is
2062 now done for all such subprocesses. The other cases are: ${run, transport
2063 filters, and the commands run by the lmtp and pipe transports.
2065 5. Some older OS have a limit of 256 on the maximum number of file
2066 descriptors. Exim was using setrlimit() to set 1000 as a large value
2067 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2068 systems. I've change it so that if it can't get 1000, it tries for 256.
2070 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2071 was an oversight, and furthermore, ever since the addition of extra
2072 controls (e.g. 4.43/32), the checks on when to allow different forms of
2073 "control" were broken. There should now be diagnostics for all cases when a
2074 control that does not make sense is encountered.
2076 7. $recipients is now available in the predata ACL (oversight).
2078 8. Tidy the search cache before the fork to do a delivery from a message
2079 received from the command line. Otherwise the child will trigger a lookup
2080 failure and thereby defer the delivery if it tries to use (for example) a
2081 cached ldap connection that the parent has called unbind on.
2083 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2084 of $address_data from the recipient verification was clobbered by the
2085 sender verification.
2087 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2088 was its contents. (It was OK if the option was not defined at all.)
2090 11. A "Completed" log line is now written for messages that are removed from
2091 the spool by the -Mrm option.
2093 12. $host_address is now set to the target address during the checking of
2094 ignore_target_hosts.
2096 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2097 being passed; this would have caused $sender_host_name to have been used if
2098 matching the list had actually called for a host name (not very likely,
2099 since this list is usually IP addresses). A host name is now passed as
2102 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2103 code that specifies a non-restarting handler (typically sigaction() in
2104 modern systems) in an attempt to fix a rare and obscure crash bug.
2106 15. Narrowed the window for a race in the daemon that could cause it to ignore
2107 SIGCHLD signals. This is not a major problem, because they are used only to
2108 wake it up if nothing else does.
2110 16. A malformed maildirsize file could cause Exim to calculate negative values
2111 for the mailbox size or file count. Odd effects could occur as a result.
2112 The maildirsize information is now recalculated if the size or filecount
2115 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2116 support for a long time. Removed HAVE_SYS_VFS_H.
2118 18. Updated exipick to current release from John Jetmore.
2120 19. Allow an empty sender to be matched against a lookup in an address list.
2121 Previously the only cases considered were a regular expression, or an
2124 20. Exim went into a mad DNS lookup loop when doing a callout where the
2125 host was specified on the transport, if the DNS lookup yielded more than
2128 21. The RFC2047 encoding function was originally intended for short strings
2129 such as real names; it was not keeping to the 75-character limit for
2130 encoded words that the RFC imposes. It now respects the limit, and
2131 generates multiple encoded words if necessary. To be on the safe side, I
2132 have increased the buffer size for the ${rfc2047: expansion operator from
2135 22. Failure to deliver a bounce message always caused it to be frozen, even if
2136 there was an errors_to setting on the router. The errors_to setting is now
2139 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2140 canonical form (fully expanded) before being placed in
2141 $sender_host_address.
2143 24. Updated eximstats to version 1.33
2145 25. Include certificate and key file names in error message when GnuTLS fails
2146 to set them up, because the GnuTLS error message doesn't include the name
2147 of the failing file when there is a problem reading it.
2149 26. Expand error message when OpenSSL has problems setting up cert/key files.
2152 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2153 (this can affect the format of dates).
2155 28. exim_tidydb, when checking for the continued existence of a message for
2156 which it has found a message-specific retry record, was not finding
2157 messages that were in split spool directories. Consequently, it was
2158 deleting retry records that should have stayed in existence.
2160 29. eximstats updated to version 1.35
2161 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2162 1.35 - bugfix such that pie charts by volume are generated correctly
2164 30. The SPA authentication driver was not abandoning authentication and moving
2165 on to the next authenticator when an expansion was forced to fail,
2166 contradicting the general specification for all authenticators. Instead it
2167 was generating a temporary error. It now behaves as specified.
2169 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2170 (the order specifies the preference for clients). The order is now AES256,
2171 AES128, 3DES, ARCFOUR128.
2173 31. Small patch to Sieve code - explicitly set From: when generating an
2176 32. Exim crashed if a remote delivery caused a very long error message to be
2177 recorded - for instance if somebody sent an entire SpamAssassin report back
2178 as a large number of 550 error lines. This bug was coincidentally fixed by
2179 increasing the size of one of Exim's internal buffers (big_buffer) that
2180 happened as part of the Exiscan merge. However, to be on the safe side, I
2181 have made the code more robust (and fixed the comments that describe what
2184 33. Some experimental protocols are using DNS PTR records for new purposes. The
2185 keys for these records are domain names, not reversed IP addresses. The
2186 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2187 leaves it alone. Component reversal etc. now happens only for IP addresses.
2190 34. The host_aton() function is supposed to be passed a string that is known
2191 to be a valid IP address. However, in the case of IPv6 addresses, it was
2192 not checking this. This is a hostage to fortune. Exim now panics and dies
2193 if the condition is not met. A case was found where this could be provoked
2194 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2195 components; fortuitously, this particular loophole had already been fixed
2196 by change 4.50/55 or 4.44/33 above.
2198 If there are any other similar loopholes, the new check in host_aton()
2199 itself should stop them being exploited. The report I received stated that
2200 data on the command line could provoke the exploit when Exim was running as
2201 exim, but did not say which command line option was involved. All I could
2202 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2203 running as the user.
2206 35. There was a buffer overflow vulnerability in the SPA authentication code
2207 (which came originally from the Samba project). I have added a test to the
2208 spa_base64_to_bits() function which I hope fixes it.
2211 36. The daemon start-up calls getloadavg() while still root for those OS that
2212 need the first call to be done as root, but it missed one case: when
2213 deliver_queue_load_max is set with deliver_drop_privilege. This is
2214 necessary for the benefit of the queue runner, because there is no re-exec
2215 when deliver_drop_privilege is set.
2217 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2218 was in use, was not putting the data itself into the right store pool;
2219 consequently, it could be overwritten for a subsequent message in the same
2220 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2223 38. Sometimes the final signoff response after QUIT could fail to get
2224 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2225 < 0 before doing a fflush(). This bug looks as though it goes back to the
2226 introduction of TLS in release 3.20, but "sometimes" must have been rare
2227 because the tests only now provoked it.
2233 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2234 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2235 Now it returns an integer. A number of calls were still expecting a T/F
2236 return. Fortuitously, in all cases, the tests worked in OK situations,
2237 which is the norm. However, things would have gone wrong on any write
2238 failures on the smtp file descriptor. This function is used when sending
2239 messages over SMTP and also when doing verify callouts.
2241 2. When Exim is called to do synchronous delivery of a locally submitted
2242 message (the -odf or -odi options), it no longer closes stderr before doing
2245 3. Implemented the mua_wrapper option.
2247 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2249 5. Implemented the functions header_remove(), header_testname(),
2250 header_add_at_position(), and receive_remove_recipient(), and exported them
2253 6. If an ACL "warn" statement specified the addition of headers, Exim already
2254 inserted X-ACL-Warn: at the start if there was no header name. However, it
2255 was not making this test for the second and subsequent header lines if
2256 there were newlines in the string. This meant that an invalid header could
2257 be inserted if Exim was badly configured.
2259 7. Allow an ACL "warn" statement to add header lines at the start or after all
2260 the Received: headers, as well as at the end.
2262 8. Added the rcpt_4xx retry error code.
2264 9. Added postmaster_mailfrom=xxx to callout verification option.
2266 10. Added mailfrom=xxxx to the callout verification option, for verify=
2269 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2270 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2272 12. Inserted some casts to stop certain compilers warning when using pointer
2273 differences as field lengths or precisions in printf-type calls (mostly
2274 affecting debugging statements).
2276 13. Added optional readline() support for -be (dynamically loaded).
2278 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2279 same clock tick as a message's arrival, so that its received time was the
2280 same as the "first fail" time on the retry record, and that message
2281 remained on the queue past the ultimate address timeout, every queue runner
2282 would try a delivery (because it was past the ultimate address timeout) but
2283 after another failure, the ultimate address timeout, which should have then
2284 bounced the address, did not kick in. This was a "< instead of <=" error;
2285 in most cases the first failure would have been in the next clock tick
2286 after the received time, and all would be well.
2288 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2289 being recognized when the domain list was tested by the match_domain
2290 condition in an expansion string.
2292 16. Added the ${str2b64: operator.
2294 17. Exim was always calling setrlimit() to set a large limit for the number of
2295 processes, without checking whether the existing limit was already
2296 adequate. (It did check for the limit on file descriptors.) Furthermore,
2297 errors from getrlimit() and setrlimit() were being ignored. Now they are
2298 logged to the main and panic logs, but Exim does carry on, to try to do its
2299 job under whatever limits there are.
2301 18. Imported PCRE 5.0.
2303 19. Trivial typo in log message " temporarily refused connection" (the leading
2306 20. If the log selector return_path_on_delivery was set and an address was
2307 redirected to /dev/null, the delivery process crashed because it assumed
2308 that a return path would always be set for a "successful" delivery. In this
2309 case, the whole delivery is bypassed as an optimization, and therefore no
2312 21. Internal re-arrangement: the function for sending a challenge and reading
2313 a response while authentication was assuming a zero-terminated challenge
2314 string. It's now changed to take a pointer and a length, to allow for
2315 binary data in such strings.
2317 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2319 23. Exim was not respecting finduser_retries when seeking the login of the
2320 uid under which it was called; it was always trying 10 times. (The default
2321 setting of finduser_retries is zero.) Also, it was sleeping after the final
2322 failure, which is pointless.
2324 24. Implemented tls_on_connect_ports.
2326 25. Implemented acl_smtp_predata.
2328 26. If the domain in control=submission is set empty, Exim assumes that the
2329 authenticated id is a complete email address when it generates From: or
2330 Sender: header lines.
2332 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2333 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2334 chown and chgrp in /bin and hostname in /usr/bin.
2336 28. Exim was keeping the "process log" file open after each use, just as it
2337 does for the main log. This opens the possibility of it remaining open for
2338 long periods when the USR1 signal hits a daemon. Occasional processlog
2339 errors were reported, that could have been caused by this. Anyway, it seems
2340 much more sensible not to leave this file open at all, so that is what now
2343 29. The long-running daemon process does not normally write to the log once it
2344 has entered its main loop, and it closes the log before doing so. This is
2345 so that log files can straightforwardly be renamed and moved. However,
2346 there are a couple of unusual error situations where the daemon does write
2347 log entries, and I had neglected to close the log afterwards.
2349 30. The text of an SMTP error response that was received during a remote
2350 delivery was being truncated at 512 bytes. This is too short for some of
2351 the long messages that one sometimes sees. I've increased the limit to
2354 31. It is now possible to make retry rules that apply only when a message has a
2355 specific sender, in particular, an empty sender.
2357 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2358 it possible to be selective about when SMTP synchronization is enforced.
2360 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2362 32. Implemented hosts_connection_nolog.
2364 33. Added an ACL for QUIT.
2366 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2369 35. Added mailbox_size and mailbox_filecount to appendfile.
2371 36. Added control = no_multiline_responses to ACLs.
2373 37. There was a bug in the logic of the code that waits for the clock to tick
2374 in the case where the clock went backwards by a substantial amount such
2375 that the microsecond fraction of "now" was more than the microsecond
2376 fraction of "then" (but the whole seconds number was less).
2378 38. Added support for the libradius Radius client library this is found on
2379 FreeBSD (previously only the radiusclient library was supported).
2385 1. When certain lookups returned multiple values in the form name=value, the
2386 quoting of the values was not always being done properly. Specifically:
2387 (a) If the value started with a double quote, but contained no whitespace,
2389 (b) If the value contained whitespace other than a space character (i.e.
2390 tabs or newlines or carriage returns) it was not quoted.
2391 This fix has been applied to the mysql and pgsql lookups by writing a
2392 separate quoting function and calling it from the lookup code. The fix
2393 should probably also be applied to nisplus, ibase and oracle lookups, but
2394 since I cannot test any of those, I have not disturbed their existing code.
2396 2. A hit in the callout cache for a specific address caused a log line with no
2397 reason for rejecting RCPT. Now it says "Previous (cached) callout
2398 verification failure".
2400 3. There was an off-by-one bug in the queryprogram router. An over-long
2401 return line was truncated at 256 instead of 255 characters, thereby
2402 overflowing its buffer with the terminating zero. As well as fixing this, I
2403 have increased the buffer size to 1024 (and made a note to document this).
2405 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2406 when Exim is waiting for an SMTP response from a remote server, Exim
2407 restarts its select() call on the socket, thereby resetting its timeout.
2408 This is not a problem when such interrupts are rare. Somebody set up a cron
2409 job to run exiwhat every 2 minutes, which is less than the normal select()
2410 timeout (5 or 10 minutes). This meant that the select() timeout never
2411 kicked in because it was always reset. I have fixed this by comparing the
2412 time when an interrupt arrives with the time at the start of the first call
2413 to select(). If more time than the timeout has elapsed, the interrupt is
2414 treated as a timeout.
2416 5. Some internal re-factoring in preparation for the addition of Sieve
2417 extensions (by MH). In particular, the "personal" test is moved to a
2418 separate function, and given an option for scanning Cc: and Bcc: (which is
2419 not set for Exim filters).
2421 6. When Exim created an email address using the login of the caller as the
2422 local part (e.g. when creating a From: or Sender: header line), it was not
2423 quoting the local part when it contained special characters such as @.
2425 7. Installed new OpenBSD configuration files.
2427 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2428 try to make them clearer.
2430 9. Callout options, other than the timeout value, were being ignored when
2431 verifying sender addresses in header lines. For example, when using
2433 verify = header_sender/callout=no_cache
2435 the cache was (incorrectly) being used.
2437 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2438 only the Cygwin environment.
2440 11. When return_path_on_delivery was set as a log selector, if different remote
2441 addresses in the same message used different return paths and parallel
2442 remote delivery occurred, the wrong values would sometimes be logged.
2443 (Whenever a remote delivery process finished, the return path value from
2444 the most recently started remote delivery process was logged.)
2446 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2447 lines when AUTH and/or TLS are in use. This is the "received protocol"
2448 field. Exim used to use "asmtp" for authenticated SMTP, without any
2449 indication (in the protocol name) for TLS use. Now it follows the RFC and
2450 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2451 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2452 names appear in log lines as well as in Received: header lines.
2454 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2455 extensions, and comparison tests, and to fix some bugs.
2457 14. Changes to the "personal" filter test:
2459 (1) The test was buggy in that it was just doing the equivalent of
2460 "contains" tests on header lines. For example, if a user's address was
2461 anne@some.where, the "personal" test would incorrectly be true for
2463 To: susanne@some.where
2465 This test is now done by extracting each address from the header in turn,
2466 and checking the entire address. Other tests that are part of "personal"
2467 are now done using regular expressions (for example, to check local parts
2468 of addresses in From: header lines).
2470 (2) The list of non-personal local parts in From: addresses has been
2471 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2472 taken from the Sieve specification recommendations.
2474 (3) If the message contains any header line starting with "List-" it is
2475 treated as non-personal.
2477 (4) The test for "circular" in the Subject: header line has been removed
2478 because it now seems ill-conceived.
2480 15. Minor typos in src/EDITME comments corrected.
2482 16. Installed latest exipick from John Jetmore.
2484 17. If headers_add on a router specified a text string that was too long for
2485 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2486 of string_sprintf() is now avoided.
2488 18. $message_body_size was not set (it was always zero) when running the DATA
2489 ACL and the local_scan() function.
2491 19. For the "mail" command in an Exim filter, no default was being set for
2492 the once_repeat time, causing a random time value to be used if "once" was
2493 specified. (If the value happened to be <= 0, no repeat happened.) The
2494 default is now 0s, meaning "never repeat". The "vacation" command was OK
2495 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2496 (I found it when inspecting the code).
2498 20. There is now an overall timeout for performing a callout verification. It
2499 defaults to 4 times the callout timeout, which applies to individual SMTP
2500 commands during the callout. The overall timeout applies when there is more
2501 than one host that can be tried. The timeout is checked before trying the
2502 next host. This prevents very long delays if there are a large number of
2503 hosts and all are timing out (e.g. when the network connections are timing
2504 out). The value of the overall timeout can be changed by specifying an
2505 additional sub-option for "callout", called "maxwait". For example:
2507 verify = sender/callout=5s,maxwait=20s
2509 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2510 to the end before writing, but this should make it even safer).
2512 22. Exim was forgetting that it had advertised PIPELINING for the second and
2513 subsequent messages on an SMTP connection. It was also not resetting its
2514 memory on STARTTLS and an internal HELO.
2516 23. When Exim logs an SMTP synchronization error within a session, it now
2517 records whether PIPELINING has been advertised or not.
2519 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2520 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2521 rather than long int.
2523 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2525 26. Added the never_mail option to autoreply.
2531 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2532 crash if the getsockname() call failed; this can happen if a connection is
2533 closed very soon after it is established. The problem was simply in the
2534 order in which certain operations were done, causing Exim to try to write
2535 to the SMTP stream before it had set up the file descriptor. The bug has
2536 been fixed by making things happen in the correct order.
2542 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2543 before the connection was closed, thus losing the rejection response.
2545 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2546 some early Solaris releases, but causes trouble in current releases where
2547 socklen_t is defined.
2549 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2552 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2555 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2556 that releasing the top of it at the end releases what was used for sub-
2557 expansions (unless the block got too big). However, discard this block if
2558 the first thing is a variable or header, so that we can use its block when
2559 it is dynamic (useful for very large $message_headers, for example).
2561 6. Lookups now cache *every* query, not just the most recent. A new, separate
2562 store pool is used for this. It can be recovered when all lookup caches are
2563 flushed. Lookups now release memory at the end of their result strings.
2564 This has involved some general refactoring of the lookup sources.
2566 7. Some code has been added to the store_xxx() functions to reduce the amount
2567 of flapping under certain conditions.
2569 8. log_incoming_interface used to affect only the <= reception log lines. Now
2570 it causes the local interface and port to be added to several more SMTP log
2571 lines, for example "SMTP connection from", and rejection lines.
2573 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2575 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2577 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2578 could be overwritten at the end of the current message (or the start of a
2579 new message if it was set in a HELO ACL). The value is now preserved for
2580 the duration of the SMTP connection.
2582 12. If a transport had a headers_rewrite setting, and a matching header line
2583 contained an unqualified address, that address was qualified, even if it
2584 did not match any rewriting rules. The underlying bug was that the values
2585 of the flags that permit the existence of unqualified sender and recipient
2586 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2587 non-local messages, and by -bnq for local messages) were not being
2588 preserved with the message after it was received.
2590 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2591 "next input=" as part of the text comprising the host identity instead of
2592 the correct text. The code was using the same buffer for two different
2593 strings. However, depending on which order the printing function evaluated
2594 its arguments, the bug did not always show up. Under Linux, for example, my
2595 test suite worked just fine.
2597 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2598 doesn't work with some older versions of Perl. It has been changed to "my",
2599 which in any case is probably the better facility to use.
2601 15. A really picky compiler found some instances of statements for creating
2602 error messages that either had too many or two few arguments for the format
2605 16. The size of the buffer for calls to the DNS resolver has been increased
2606 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2607 for addresses that have a lot of PTR records. This alleviates a problem; it
2608 does not fully solve it.
2610 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2611 buffer now truncates the list and logs the incident, which is the same
2612 action as happens when Exim is looking up a host name and its aliases.
2613 Previously in this situation something unpredictable would happen;
2614 sometimes it was "internal error: store_reset failed".
2616 18. If a server dropped the connection unexpectedly when an Exim client was
2617 using GnuTLS and trying to read a response, the client delivery process
2618 crashed while trying to generate an error log message.
2620 19. If a "warn" verb in an ACL added multiple headers to a message in a single
2621 string, for example:
2623 warn message = H1: something\nH2: something
2625 the text was added as a single header line from Exim's point of view
2626 though it ended up OK in the delivered message. However, searching for the
2627 second and subsequent header lines using $h_h2: did not work. This has been
2628 fixed. Similarly, if a system filter added multiple headers in this way,
2629 the routers could not see them.
2631 20. Expanded the error message when iplsearch is called with an invalid key to
2632 suggest using net-iplsearch in a host list.
2634 21. When running tests using -bh, any delays imposed by "delay" modifiers in
2635 ACLs are no longer actually imposed (and a message to that effect is
2638 22. If a "gecos" field in a passwd entry contained escaped characters, in
2639 particular, if it contained a \" sequence, Exim got it wrong when building
2640 a From: or a Sender: header from that name. A second bug also caused
2641 incorrect handling when an unquoted " was present following a character
2642 that needed quoting.
2644 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2645 was not being matched caselessly.
2647 24. Arranged for all hyphens in the exim.8 source to be escaped with
2650 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2651 properly. Recipient callout cache records were still being keyed to include
2652 the sender, even when use_sender was set false. This led to far more
2653 callouts that were necessary. The sender is no longer included in the key
2654 when use_sender is false.
2656 26. Added "control = submission" modifier to ACLs.
2658 27. Added the ${base62d: operator to decode base 62 numbers.
2660 28. dnsdb lookups can now access SRV records.
2662 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2663 the configuration file.
2665 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2666 (-v) mode. This makes the output for a verbose queue run more intelligible.
2668 31. Added a use_postmaster feature to recipient callouts.
2670 32. Added the $body_zerocount variable, containing the number of binary zero
2671 bytes in the message body.
2673 33. The time of last modification of the "new" subdirectory is now used as the
2674 "mailbox time last read" when there is a quota error for a maildir
2677 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2679 35. Added +ignore_unknown as a special item in host lists.
2681 36. Code for decoding IPv6 addresses in host lists is now included, even if
2682 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2683 address was recognized as an IP address, but was then not correctly decoded
2684 into binary, causing unexpected and incorrect effects when compared with
2691 1. Very minor rewording of debugging text in manualroute to say "list of
2692 hosts" instead of "hostlist".
2694 2. If verify=header_syntax was set, and a header line with an unqualified
2695 address (no domain) and a large number of spaces between the end of the
2696 name and the colon was received, the reception process suffered a buffer
2697 overflow, and (when I tested it) crashed. This was caused by some obsolete
2698 code that should have been removed. The fix is to remove it!
2700 3. When running in the test harness, delay a bit after writing a bounce
2701 message to get a bit more predictability in the log output.
2703 4. Added a call to search_tidyup() just before forking a reception process. In
2704 theory, someone could use a lookup in the expansion of smtp_accept_max_
2705 per_host which, without the tidyup, could leave open a database connection.
2707 5. Added the variables $recipient_data and $sender_data which get set from a
2708 lookup success in an ACL "recipients" or "senders" condition, or a router
2709 "senders" option, similar to $domain_data and $local_part_data.
2711 6. Moved the writing of debug_print from before to after the "senders" test
2714 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2715 problems for message scanning, either using a data ACL, or using
2716 local_scan() because the Received: header was not generated till after they
2717 were called (in order to set the time as the time of reception completion).
2718 I have revised the way this works. The header is now generated after the
2719 body is received, but before the ACL or local_scan() are called. After they
2720 are run, the timestamp in the header is updated.
2726 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2727 before starting a queue runner without re-exec. This happened only when
2728 deliver_drop_privilege was set or when the Exim user was set to root. The
2729 effect of the bug was that timeouts during subsequent deliveries caused
2730 crashes instead of being properly handled. The handler is now left at its
2731 default (and expected) setting.
2733 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2734 message, again when deliver_drop_privilege is set or Exim is run as root.
2735 The bug described in (1) was not present in this case, but the tidying up
2736 of the other signals was missing. I have made the two cases consistent.
2738 3. The ignore_target_hosts setting on a manualroute router was being ignored
2739 for hosts that were looked up using the /MX notation.
2741 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2744 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2745 operated on the sender address. After changing the $sender_address to <>
2746 for the sender address verify, Exim was re-instated it as the original
2747 (before rewriting) address, but remembering that it had rewritten it, so it
2748 wasn't rewriting it again. This bug also had the effect of breaking the
2749 sender address verification caching when the sender address was rewritten.
2751 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2752 This has been changed so that if the ip literal address matches
2753 ignore_target_hosts, the router declines.
2755 7. Added expansion conditions match_domain, match_address, and match_local_
2756 part (NOT match_host).
2758 8. The placeholder for the Received: header didn't have a length field set.
2760 9. Added code to Exim itself and to exim_lock to test for a specific race
2761 condition that could lead to file corruption when using MBX delivery. The
2762 issue is with the lockfile that is created in /tmp. If this file is removed
2763 after a process has opened it but before that process has acquired a lock,
2764 there is the potential for a second process to recreate the file and also
2765 acquire a lock. This could lead to two Exim processes writing to the file
2766 at the same time. The added code performs the same test as UW imapd; it
2767 checks after acquiring the lock that its file descriptor still refers to
2768 the same named file.
2770 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2771 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2774 11. Added the smtp_active_hostname option. If used, this will typically be made
2775 to depend on the incoming interface address. Because $interface_address is
2776 not set up until the daemon has forked a reception process, error responses
2777 that can happen earlier (such as "too many connections") no longer contain
2780 12. If an expansion in a condition on a "warn" statement fails because a lookup
2781 defers, the "warn" statement is abandoned, and the next ACL statement is
2782 processed. Previously this caused the whole ACL to be aborted.
2784 13. Added the iplsearch lookup type.
2786 14. Added ident_timeout as a log selector.
2788 15. Added tls_certificate_verified as a log selector.
2790 16. Added a global option tls_require_ciphers (compare the smtp transport
2791 option of the same name). This controls incoming TLS connections.
2793 17. I finally figured out how to make tls_require_ciphers do a similar thing
2794 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2795 before starting the TLS session.
2797 18. Tabs are now shown as \t in -bP output.
2799 19. If the log selector return_path_on_delivery was set, Exim crashed when
2800 bouncing a message because it had too many Received: header lines.
2802 20. If two routers both had headers_remove settings, and the first one included
2803 a superfluous trailing colon, the final name in the first list and the
2804 first name in the second list were incorrectly joined into one item (with a
2805 colon in the middle).
2811 1. Added -C and -D options to the exinext utility, mainly to make it easier
2812 to include in the automated testing, but these could be helpful when
2813 multiple configurations are in use.
2815 2. The exinext utility was not formatting the output nicely when there was
2816 an alternate port involved in the retry record key, nor when there was a
2817 message id as well (for retries that were specific to a specific message
2818 and a specific host). It was also confused by IPv6 addresses, because of
2819 the additional colons they contain. I have fixed the IPv4 problem, and
2820 patched it up to do a reasonable job for IPv6.
2822 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2823 delivery, the log line now contains "pipelined" if PIPELINING was used.
2825 4. An SMTP transport process used to panic and die if the bind() call to set
2826 an explicit outgoing interface failed. This has been changed; it is now
2827 treated in the same way as a connect() failure.
2829 5. A reference to $sender_host_name in the part of a conditional expansion
2830 that was being skipped was still causing a DNS lookup. This no longer
2833 6. The def: expansion condition was not recognizing references to header lines
2834 that used bh_ and bheader_.
2836 7. Added the _cache feature to named lists.
2838 8. The code for checking quota_filecount in the appendfile transport was
2839 allowing one more file than it should have been.
2841 9. For compatibility with Sendmail, the command line option
2849 and sets the incoming protocol and host name (for trusted callers). The
2850 host name and its colon can be omitted when only the protocol is to be set.
2851 Note the Exim already has two private options, -pd and -ps, that refer to
2852 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2853 "s", but I don't think that's a major issue.
2855 10. A number of refactoring changes to the code, none of which should affect
2858 (a) The number of logging options was getting close to filling up the
2859 32-bit word that was used as a bit map. I have split them into two classes:
2860 those that are passed in the argument to log_write(), and those that are
2861 only ever tested independently outside of that function. These are now in
2862 separate 32-bit words, so there is plenty of room for expansion again.
2863 There is no change in the user interface or the logging behaviour.
2865 (b) When building, for example, log lines, the code previously used a
2866 macro that called string_cat() twice, in order to add two strings. This is
2867 not really sufficiently general. Furthermore, there was one instance where
2868 it was actually wrong because one of the argument was used twice, and in
2869 one call a function was used. (As it happened, calling the function twice
2870 did not affect the overall behaviour.) The macro has been replaced by a
2871 function that can join an arbitrary number of extra strings onto a growing
2874 (c) The code for expansion conditions now uses a table and a binary chop
2875 instead of a serial search (which was left over from when there were very
2876 few conditions). Also, it now recognizes conditions like "pam" even when
2877 the relevant support is not compiled in: a suitably worded error message is
2878 given if an attempt is made to use such a condition.
2880 11. Added ${time_interval:xxxxx}.
2882 12. A bug was causing one of the ddress fields not to be passed back correctly
2883 from remote delivery subprocesses. The field in question was not being
2884 subsequently used, so this caused to problems in practice.
2886 13. Added new log selectors queue_time and deliver_time.
2888 14. Might have fixed a bug in maildirsizefile handling that threw up
2889 "unexpected character" debug warnings, and recalculated the data
2890 unnecessarily. In any case, I expanded the warning message to give more
2893 15. Added the message "Restricted characters in address" to the statements in
2894 the default ACL that block characters like @ and % in local parts.
2896 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2897 Three changes have been made:
2899 (a) There was a serious bug; a negative response to MAIL caused the whole
2900 recipient domain to be cached as invalid, thereby blocking all messages
2901 to all local parts at the same domain, from all senders. This bug has
2902 been fixed. The domain is no longer cached after a negative response to
2903 MAIL if the sender used is not empty.
2905 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2908 (c) A new callout option, "use_sender" has been added for people who want
2909 the modified behaviour.
2915 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2918 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2919 indeed breaks things for older releases.
2921 3. Added additional logging to the case where there is a problem reading data
2922 from a filter that is running in a subprocess using a pipe, in order to
2923 try to track down a specific problem.
2925 4. Testing facility fudge: when running in the test harness and attempting
2926 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2927 getting "No route to host". Convert this to a timeout.
2929 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2932 6. Some OS don't have socklen_t but use size_t instead. This affects the
2933 fifth argument of getsockopt() amongst other things. This is now
2934 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2935 can be set for individual OS. I have set it for SunOS5, OSF1, and
2936 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2937 some earlier ones do not.
2939 7. Change 4.30/15 was not doing the test caselessly.
2941 8. The standard form for an IPv6 address literal was being rejected by address
2942 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2943 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2944 this, as well as the form without the "IPv6" on the front (but only when
2945 address literals are enabled, of course).
2947 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2949 10. Exim crashed if a message with an empty sender address specified by -f
2950 encountered a router with an errors_to setting. This could be provoked only
2951 by a command such as
2955 where an empty string was supplied; "<>" did not hit this bug.
2957 11. Installed PCRE release 4.5.
2959 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2960 remained set. It is now erased.
2962 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2963 times from message ids (which are base 36 rather than the normal 62).
2965 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2966 were being counted as actual protocol errors, and logged if the log
2967 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2968 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2969 and DATA following a set of rejected RCPTs do not count as protocol errors.
2970 In other words, Exim assumes they were pipelined, though this may not
2971 actually be the case. Of course, in all cases the client gets an
2972 appropriate error code.
2974 15. If a lookup fails in an ACL condition, a message about the failure may
2975 be available; it is used if testing the ACL cannot continue, because most
2976 such messages specify what the cause of the deferral is. However, some
2977 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2978 that caused an old message to be retained and used if a later statement
2979 caused a defer, replacing the real cause of the deferral.
2981 16. If an IP address had so many PTR records that the DNS lookup buffer
2982 was not large enough to hold them, Exim could crash while trying to process
2983 the truncated data. It now detects and logs this case.
2985 17. Further to 4.21/58, another change has been made: if (and only if) the
2986 first line of a message (the first header line) ends with CRLF, a bare LF
2987 in a subsequent header line has a space inserted after it, so as not to
2988 terminate the header.
2990 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2991 unnecessarily, used atoi() instead of strtol(), and didn't check the
2992 termination when getting file sizes from file names by regex.
2994 19. Completely re-implemented the support for maildirsize files, in the light
2995 of a number of problems with the previous contributed implementation
2996 (4.30/29). In particular:
2998 . If the quota is zero, the maildirsize file is maintained, but no quota is
3001 . If the maildir directory does not exist, it is created before any attempt
3002 to write a maildirsize file.
3004 . The quota value in the file is just a cache; if the quota is changed in
3005 the transport, the new value overrides.
3007 . A regular expression is available for excluding directories from the
3010 20. The autoreply transport checks the characters in options that define the
3011 message's headers; it allows continued headers, but it was checking with
3012 isspace() after an embedded newline instead of explicitly looking for a
3015 21. If all the "regular" hosts to which an address was routed had passed their
3016 expiry times, and had not reached their retry times, the address was
3017 bounced, even if fallback hosts were defined. Now Exim should go on to try
3020 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3021 equivalent code in the SMTP transport. Some hosts send humungous responses
3022 to HELO/EHLO, more than 1024 it seems.
3024 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3025 gives compiler warnings in some environments. I've now done it "properly",
3028 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3029 (because of problems with the built-in one) was declared to return uschar *
3030 instead of char *, causing compiler failure.
3032 25. Fixed a file descriptor leak when processing alias/forward files.
3034 26. Fixed a minor format string issue in dbfn.c.
3036 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3038 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3039 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3042 29. When a sender address is verified, it is cached, to save repeating the test
3043 when there is more than one recipient in a message. However, when the
3044 verification involves a callout, it is possible for different callout
3045 options to be set for different recipients. It is too complicated to keep
3046 track of this in the cache, so now Exim always runs a verification when a
3047 callout is required, relying on the callout cache for the optimization.
3048 The overhead is duplication of the address routing, but this should not be
3051 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3052 to be verified with callout=postmaster, and the main callout worked but the
3053 postmaster check failed, the verification correctly failed. However, if a
3054 subsequent RCPT command asked for sender verification *without* the
3055 postmaster check, incorrect caching caused this verification also to fail,
3058 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3059 it was not caching the DNS options (qualify_single, search_parents) that
3060 were used when the lookup failed. A subsequent lookup with different
3061 options therefore always gave the same answer, though there were cases
3062 where it should not have. (Example: a "domains = !$mx_any" option on a
3063 dnslookup router: the "domains" option is always processed without any
3064 widening, but the router might have qualify_single set.) Now Exim uses the
3065 cached value only when the same options are set.
3067 32. Added John Jetmore's "exipick" utility to the distribution.
3069 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3070 than a timeout (e.g. a certificate is required, and is not provided), an
3071 Exim server now closes the connection immediately. Previously it waited for
3072 the client to close - but if the client is SSL, it seems that they each
3073 wait for each other, leading to a delay before one of them times out.
3075 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3076 maintained 0.8.x compatibility because I don't think many are using it, and
3077 it is clearly obsolete.
3079 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3082 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3083 client certificate was expired. A simple patch fixes this, though I don't
3084 understand the full logic of why the verify callback is called multiple
3087 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3088 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3089 which causes problems with some clients (such as the Certicom SSL Plus
3090 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3091 disables the coutermeasure allowing Eudora to connect."
3093 38. Exim was not checking that a write() to a log file succeeded. This could
3094 lead to Bad Things if a log got too big, in particular if it hit a file
3095 size limit. Exim now panics and dies if it cannot write to a log file, just
3096 as it does if it cannot open a log file.
3098 39. Modified OS/Makefile-Linux so that it now contains
3100 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3102 The two -D definitions ensure that Exim is compiled with large file
3103 support, which makes it possible to handle log files that are bigger than
3106 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3107 instance) a domain was checked against a named list that involved a lookup,
3108 causing $domain_data to be set, then another domain was checked against the
3109 same list, then the first domain was re-checked, the value of $domain_data
3110 after the final check could be wrong. In particular, if the second check
3111 failed, it could be set empty. This bug probably also applied to
3114 41. The strip_trailing_dot option was not being applied to the address given
3115 with the -f command-line option.
3117 42. The code for reading a message's header from the spool was incrementing
3118 $received_count, but never initializing it. This meant that the value was
3119 incorrect (doubled) while delivering a message in the same process in which
3120 it was received. In the most common configuration of Exim, this never
3121 happens - a fresh exec is done - but it can happen when
3122 deliver_drop_privilege is set.
3124 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3125 it now includes up to 150 characters of the unexpected data in the log
3128 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3129 and building data strings. The size of both of these buffers was 10 000
3130 bytes - far larger than anybody would *ever* want, thought I. Needless to
3131 say, somebody hit the limit. I have increased the maximum line length to
3132 20 000 and the maximum data length of concatenated lines to 100 000. I have
3133 also fixed two bugs, because there was no checking on these buffers. Tsk,
3134 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3135 buffer is too small.
3137 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3138 lsearch lookups. Now it does.
3140 46. When parsing a route_list item in a manualroute router, a fixed-length
3141 buffer was used for the list of hosts. I made this 1024 bytes long,
3142 thinking that nobody would ever have a list of hosts that long. Wrong.
3143 Somebody had a whole pile of complicated expansion conditions, and the
3144 string was silently truncated, leading to an expansion error. It turns out
3145 that it is easier to change to an unlimited length (owing to other changes
3146 that have happened since this code was originally written) than to build
3147 structure for giving a limitation error. The length of the item that
3148 expands into the list of hosts is now unlimited.
3150 47. The lsearch lookup could not handle data where the length of text line was
3151 more than 4095 characters. Such lines were truncated, leading to shortened
3152 data being returned. It should now handle lines of any length.
3154 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3155 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3158 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3159 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3160 no longer happens in comment lines. A list of replacements is now placed
3161 at the head of all of the source files, except those whose only change is
3162 to replace PERL_COMMAND in the very first #! line.
3164 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3165 messages on the queue, with a bottom-up merge sort, using code contributed
3166 by Michael Haardt. This should make operations like -bp somewhat faster on
3167 large queues. It won't affect queue runners, except when queue_run_in_order
3170 51. Installed eximstats 1.31 in the distribution.
3172 52. Added support for SRV lookups to the dnslookup router.
3174 53. If an ACL referred to $message_body or $message_body_end, the value was not
3175 reset for any messages that followed in the same SMTP session.
3177 54. The store-handling optimization for building very long strings was not
3178 differentiating between the different store pools. I don't think this
3179 actually made any difference in practice, but I've tidied it.
3181 55. While running the routers to verify a sender address, $sender_address
3182 was still set to the sender address. This is wrong, because when routing to
3183 send a bounce to the sender, it would be empty. Therefore, I have changed
3184 it so that, while verifying a sender address, $sender_address is set to <>.
3185 (There is no change to what happens when verifying a recipient address.)
3187 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3188 target A or AAAA records (if not already returned) without resetting the
3189 qualify_single or search_parents options of the DNS resolver. These are
3190 inappropriate in this case because the targets of MX and SRV records must
3191 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3192 target that, when qualified, matched something in the local domain. These
3193 two options are now turned off when doing these lookups.
3195 57. It seems that at least some releases of Reiserfs (which does not have the
3196 concept of a fixed number of inodes) returns zero and not -1 for the
3197 number of available inodes. This interacted badly with check_spool_inodes,
3198 which assumed that -1 was the "no such thing" setting. What I have done is
3199 to check that the total number of inodes is greater than zero before doing
3200 the test of how many are available.
3202 58. When a "warn" ACL statement has a log_message modifier, the message is
3203 remembered, and not repeated. This is to avoid a lot of repetition when a
3204 message has many recipients that cause the same warning to be written.
3205 Howewer, Exim was preserving the list of already written lines for an
3206 entire SMTP session, which doesn't seem right. The memory is now reset if a
3207 new message is started.
3209 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3210 debugging output unless log_rewrite was also set.
3212 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3213 of (int)(handle) when we know that handle contains (void *)(-1).
3215 61. The Exim daemon panic-logs an error return when it closes the incoming
3216 connection. However "connection reset by peer" seems to be common, and
3217 isn't really an error worthy of noting specially, so that particular error
3220 62. When Exim is trying to find all the local interfaces, it used to panic and
3221 die if the ioctl to get the interface flags failed. However, it seems that
3222 on at least one OS (Solaris 9) it is possible to have an interface that is
3223 included in the list of interfaces, but for which you get a failure error
3224 for this call. This happens when the interface is not "plumbed" into a
3225 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3226 failure of the "get flags" call assumes that the interface is down.
3228 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3229 makes life easier for people who are doing arithmetic on fields extracted
3230 from dates, where you often get leading zeros that should not be
3231 interpreted as octal.
3233 64. Added qualify_domain to the redirect router, to override the global
3236 65. If a pathologically long header line contained very many addresses (the
3237 report of this problem mentioned 10 000) and each of them was rewritten,
3238 Exim could use up a very large amount of memory. (It kept on making new
3239 copies of the header line as it rewrote, and never released the old ones.)
3240 At the expense of a bit more processing, the header rewriting function has
3241 been changed so that it no longer eats memory in this way.
3243 66. The generation of the Received: header has been moved from the time that a
3244 message starts to be received, to the time that it finishes. The timestamp
3245 in the Received: header should now be very close to that of the <= log
3246 line. There are two side-effects of this change:
3248 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3249 logged header lines no longer include the local Received: line, because
3250 it has not yet been created. The same applies to a copy of the message
3251 that is returned to a non-SMTP sender when a message is rejected.
3253 (b) When a filter file is tested using -bf, no additional Received: header
3254 is added to the test message. After some thought, I decided that this
3257 This change does not affect the value of $received_for. It is still set
3258 after address rewriting, but before local_scan() is called.
3260 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3262 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3263 gave an unhelpful panic error message, and a defer error. I have managed to
3264 change this behaviour so that it now rejects any supplied certificate,
3265 which seems right, as the list of acceptable certificates is empty.
3267 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3268 gave an unhelpful defer error. I have not managed to make this reject any
3269 supplied certificates, but the error message it gives is "no certificate
3270 supplied", which is not helpful.
3272 70. exigrep's output now also includes lines that are not associated with any
3273 message, but which match the given pattern. Implemented by a patch from
3274 Martin Sluka, which also tidied up the Perl a bit.
3276 71. Recipient callout verification, like sender verification, was using <> in
3277 the MAIL FROM command. This isn't really the right thing, since the actual
3278 sender may affect whether the remote host accepts the recipient or not. I
3279 have changed it to use the actual sender in the callout; this means that
3280 the cache record is now keyed on a recipient/sender pair, not just the
3281 recipient address. There doesn't seem to be a real danger of callout loops,
3282 since a callout by the remote host to check the sender would use <>.
3283 [SEE ABOVE: changed after hitting problems.]
3285 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3286 temporary errors. However, in the case of such a code being given after
3287 the end of a data transmission (i.e. after ".") Exim was failing to write
3288 a retry record for the message. (Yes, there was some broken host that was
3289 actually sending 8xx at this point.)
3291 73. An unknown lookup type in a host list could cause Exim to panic-die when
3292 the list was checked. (An example that provoked this was putting <; in the
3293 middle of a list instead of at the start.) If this happened during a DATA
3294 ACL check, a -D file could be left lying around. This kind of configuration
3295 error no longer causes Exim to die; instead it causes a defer errror. The
3296 incident is still logged to the main and panic logs.
3298 74. Buglet left over from Exim 3 conversion. The message "too many messages
3299 in one connection" was written to the rejectlog but not the mainlog, except
3300 when address rewriting (yes!) was being logged.
3302 75. Added write_rejectlog option.
3304 76. When a system filter was run not as root (that is, when system_filter_user
3305 was set), the values of the $n variables were not being returned to the
3306 main process; thus, they were not subsequently available in the $sn
3309 77. Added +return_path_on_delivery log selector.
3311 78. A connection timeout was being treated differently from recipients deferred
3312 when testing hosts_max_try with a message that was older than the host's
3313 retry timeout. (The host should not be counted, thus allowing all hosts to
3314 be tried at least once before bouncing.) This may have been the cause of an
3315 occasionally reported bug whereby a message would remain on the queue
3316 longer than the retry timeout, but would be bounced if a delivery was
3317 forced. I say "may" because I never totally pinned down the problem;
3318 setting up timeout/retry tests is difficult. See also the next item.
3320 79. The ultimate address timeout was not being applied to errors that involved
3321 a combination of host plus message (for example, a timeout on a MAIL
3322 command). When an address resolved to a number of possible hosts, and they
3323 were not all tried for each delivery (e.g. because of hosts_max_try), a
3324 message could remain on the queue longer than the retry timeout.
3326 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3329 81. Fixed an obscure SMTP outgoing bug which required at least the following
3330 conditions: (a) there was another message waiting for the same server;
3331 (b) the server returned 5xx to all RCPT commands in the first message so
3332 that the message was not completed; (c) the server dropped the connection
3333 or gave a negative response to the RSET that Exim sends to abort the
3334 transaction. The observed case was a dropped connection after DATA that had
3335 been sent in pipelining mode. That is, the server had advertised PIPELINING
3336 but was not implementing it correctly. The effect of the bug was incorrect
3337 behaviour, such as trying another host, and this could lead to a crash.
3343 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3344 and daemon.c were passed as pointers to ints; they should have been
3345 pointers to socklen_t variables (which are typically unsigned ints).
3347 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3350 3. Fixed a really odd bug that affected only the testing scheme; patching a
3351 certain fixed string in the binary changed the value of another string that
3352 happened to be identical to the end of the original first string.
3354 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3355 name", it returns that address as the IP address. On some operating
3356 systems (e.g. Solaris), it also passes back the IP address string as the
3357 "host name". However, on others (e.g. Linux), it passes back an empty
3358 string. Exim wasn't checking for this, and was changing the host name to an
3359 empty string, assuming it had been canonicized.
3361 5. Although rare, it is permitted to have more than one PTR record for a given
3362 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3363 all the names associated with an address, because they do in Solaris.
3364 However, it seems that they do not in Linux for data that comes from the
3365 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3366 I found this out when I moved to a new Linux workstation and tried to run
3367 the Exim test suite.
3369 To get round this problem I have changed the code so that it now does its
3370 own call to the DNS to look up PTR records when searching for a host name.
3371 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3372 addresses that are only in /etc/hosts are still found.
3374 This behaviour is, however, controlled by an option called host_lookup_
3375 order, which defaults to "bydns:byaddr". If people want to use the other
3376 order, or indeed, just use one or the other means of lookup, they can
3377 specify it in this variable.
3379 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3380 some operating systems, this comes back from gethostbyaddr() as an empty
3381 string, and this is what Exim used to test for. However, it seems that in
3382 other systems, "." is yielded. Exim now tests for this case too.
3384 7. The values of check_spool_space and check_log_space are now held internally
3385 as a number of kilobytes instead of an absolute number of bytes. If a
3386 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3387 kilobyte. This means that much larger values can be stored.
3389 8. Exim monitor: an attempt to get the action menu when not actually pointing
3390 at a message produces an empty menu entitled "No message selected". This
3391 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3392 no entries in it ("Shell widget menu has zero width and/or height"). So I
3393 have added a single, blank menu entry in this case.
3395 9. Added ${quote_local_part.
3397 10. MIME decoding is now applied to the contents of Subject: header lines when
3400 11. Now that a reference to $sender_host_address automatically causes a reverse
3401 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3402 host lookup before query-style lookups in lists that might use this
3403 variable. This has therefore been abolished, and the "net-" prefix is no
3404 longer necessary for query-style lookups.
3406 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3407 have been a typo for LFLAGS, so it has been changed.
3409 13. The install script calls Exim with "-C /dev/null" in order to find the
3410 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3411 to be output. Howeve, since Exim outputs its version number before the
3412 error, it didn't break the script. It just looked ugly. I fixed this by
3413 always allowing "-C /dev/null" if the caller is root.
3415 14. Ignore overlarge ACL variable number when reading spool file - insurance
3416 against a later release with more variables having written the file.
3418 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3419 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3420 this, as well as the form without the "IPv6" on the front.
3422 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3423 OS/Makefile-Darwin file.
3425 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3426 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3428 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3429 variable. However, some OS use ints for inodes. I've added cast to long int
3430 to get rid of the compiler warning.
3432 19. I had forgotten to lock out "/../" in configuration file names when
3433 ALT_CONFIG_PREFIX was set.
3435 20. Routers used for verification do not need to specify transports. However,
3436 if such a router generated a host list, and callout was configured, Exim
3437 crashed, because it could not find a port number from the (non-existent)
3438 transport. It now assumes port 25 in this circumstance.
3440 21. Added the -t option to exigrep.
3442 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3443 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3444 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3445 them is set, LOOKUP_LSEARCH is forced.
3447 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3449 24. Added sender and host information to the "rejected by local_scan()" log
3450 line; previously there was no indication of these.
3452 25. Added .include_if_exists.
3454 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3455 Linux. It turns out that not all file systems support this. Apparently some
3456 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3457 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3458 not supported on the file descriptor, is now ignored when Exim is trying to
3459 sync a directory. This applies only to Linux.
3461 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3463 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3464 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3467 29. Added support for maildirsize files from supplied patch (modified a bit).
3469 30. The use of :fail: followed by an empty string could lead Exim to respond to
3470 sender verification failures with (e.g.):
3472 550 Verification failed for <xxx>
3473 550 Sender verify failed
3475 where the first response line was missing the '-' that indicates it is not
3476 the final line of the response.
3478 31. The loop for finding the name of the user that called Exim had a hardwired
3479 limit of 10; it now uses the value of finduser_retries, which is used for
3480 all other user lookups.
3482 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3485 33. Exim was neglecting to zero errno before one call of strtol() when
3486 expanding a string and expecting an integer value. On some systems this
3487 resulted in spurious "integer overflow" errors. Also, it was casting the
3488 result into an int without checking.
3490 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3491 did not work. The code looks as if it has *never* worked, though it appears
3492 to have been documented since at least releast 1.62. I have made it work.
3494 35. The "timeout_DNS" error in retry rules, also documented since at least
3495 1.62, also never worked. As it isn't clear exactly what this means, and
3496 clearly it isn't a major issue, I have abolished the feature by treating it
3497 as "timeout", and writing a warning to the main and panic logs.
3499 36. The display of retry rules for -brt wasn't always showing the error code
3502 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3503 timeout_connect_A, timeout_connect_MX.
3505 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3506 to the empty sender.
3508 39. The daemon was not analysing the content of -oX till after it had closed
3509 stderr and disconnected from the controlling terminal. This meant that any
3510 syntax errors were only noted on the panic log, and the return code from
3511 the command was 0. By re-arranging the code a little, I've made the
3512 decoding happen first, so such errors now appear on stderr, and the return
3513 code is 1. However, the actual setting up of the sockets still happens in
3514 the disconnected process, so errors there are still only recorded on the
3517 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3518 connections (as happens on some IP stacks) was logged at start up time as
3519 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3520 it from "IPv6 and IPv4", which means that two separate sockets are being
3523 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3524 says whether AF_INET or AF_INET6 was passed as an argument.
3526 42. Exiwhat output was messed up when time zones were included in log
3529 43. Exiwhat now gives more information about the daemon's listening ports,
3530 and whether -tls-on-connect was used.
3532 44. The "port" option of the smtp transport is now expanded.
3534 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3535 silently ignored. Now an error message is written to the main and panic
3538 46. There's a new ACL modifier called "logwrite" which writes to a log file
3539 as soon as it is encountered.
3541 47. Added $local_user_uid and $local_user_gid at routing time.
3543 48. Exim crashed when trying to verify a sender address that was being
3546 49. Exim was recognizing only a space character after ".include". It now also
3547 recognizes a tab character.
3549 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3550 extracting the relevant information from the specification. The man page no
3551 longer contains scrambled data for the -d option, and I've added a section
3552 at the front about calling Exim under different names.
3554 51. Added "extra_headers" argument to the "mail" command in filter files.
3556 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3559 53. Installed eximstats 1.29.
3561 54. Added transport_filter_timeout as a generic transport option.
3563 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3564 Cc: header lines. This was required by RFC 822, but it not required by RFC
3567 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3568 incoming messages that did not have them. Now it does so only if the
3569 message originates locally, that is, if there is no associated remote host
3570 address. When Resent- header lines are present, this applies to the Resent-
3571 lines rather than the non-Resent- lines.
3573 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3574 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3576 58. Messages for configuration errors now include the name of the main
3577 configuration file - useful now that there may be more than one file in a
3578 list (.included file names were always shown).
3580 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3581 for those rare installations that do not start the daemon as root or run it
3582 setuid root. I've cut out the call to initgroups() if the daemon is not
3585 60. The Exim user and group can now be bound into the binary as text strings
3586 that are looked up at the start of Exim's processing.
3588 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3590 62. Added $mailstore_basename variable.
3592 63. Installed patch to sieve.c from Michael Haardt.
3594 64. When Exim failed to open the panic log after failing to open the main log,
3595 the original message it was trying to log was written to stderr and debug
3596 output, but if they were not available (the usual case in production), it
3597 was lost. Now it is written to syslog before the two lines that record the
3598 failures to open the logs.
3600 65. Users' Exim filters run in subprocesses under the user's uid. It is
3601 possible for a "deliver" command or an alias in a "personal" command to
3602 provoke an address rewrite. If logging of address rewriting is configured,
3603 this fails because the process is not running as root or exim. There may be
3604 a better way of dealing with this, but for the moment (because 4.30 needs
3605 to be released), I have disabled address rewrite logging when running a
3606 filter in a non-root, non-exim process.
3612 1. The buildconfig auxiliary program wasn't quoting the value set for
3613 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3614 not defined. This bug was masked in 4.22 by the effect that was fixed in
3617 2. Some messages that were rejected after a message id was allocated were
3618 shown as "incomplete" by exigrep. It no longer does this for messages that
3619 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3621 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3622 have allow_domain_literals set, the ID did not get logged in the <= line.
3623 Domain literals are now always recognized in Message-ID: header lines.
3625 4. The first argument for a ${extract expansion item is the key name or field
3626 number. Leading and trailing spaces in this item were not being ignored,
3627 causing some misleading effects.
3629 5. When deliver_drop_privilege was set, single queue runner processes started
3630 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3631 same command in the process it spins off) were not dropping privilege.
3633 6. When the daemon running as "exim" started a queue runner, it always
3634 re-executed Exim in the spun-off process. This is a waste of effort when
3635 deliver_drop_privilege is set. The new process now just calls the
3636 queue-runner function directly.
3642 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3645 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3646 ignored. Though the use of -oP was forcing the writing of a pid file, it
3647 was always written to the default place.
3649 3. If the message "no IP address found for host xxxx" is generated during
3650 incoming verification, it is now followed by identification of the incoming
3651 connection (so you can more easily find what provoked it).
3653 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3655 5. Added some features to "harden" Exim a bit more against certain attacks:
3657 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3658 be put in Local/Makefile. This is like the never_users runtime option,
3659 but it cannot be overridden. The default setting is "root".
3661 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3662 prefix string with which any file named in a -C command line option
3665 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3666 is retained for -C and -D only if the caller of Exim is root. Without
3667 it, the exim user may also use -C and -D and retain privilege.
3669 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3670 command line option is disabled.
3672 6. Macro names set by the -D option must start with an upper case letter, just
3673 like macro names defined in the configuration file.
3675 7. Added "dereference=" facility to LDAP.
3677 8. Two instances of the typo "uknown" in the source files are fixed.
3679 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3680 the Configure-Makefile script screwed up while processing it.
3682 10. Incorporated PCRE 4.4.
3684 11. The SMTP synchronization check was not operating right at the start of an
3685 SMTP session. For example, it could not catch a HELO sent before the client
3686 waited for the greeting. There is now a check for outstanding input at the
3687 point when the greeting is written. Because of the duplex, asynchronous
3688 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3689 way, but not yet received, when the check is performed.
3691 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3692 on TCP/IP sockets, because this apparently causes some broken clients to
3695 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3696 unchanged) from the Cygwin maintainer.
3698 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3699 was supported instead of testing for mbx. Effectively a typo.
3701 15. The spa authenticator server code was not checking that the input it
3702 received was valid base64.
3704 16. The debug output line for the "set" modifier in ACLs was not showing the
3705 name of the variable that was being set.
3707 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3709 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3710 lookup on its own. Something else was needed to trigger the lookup. For
3711 example, a match in host_lookup or the need for a host name in a host list.
3712 Now, if $sender_host_name is referenced and the host name has not yet been
3713 looked up, a lookup is performed. If the lookup fails, the variable remains
3714 empty, and $host_lookup_failed is set to "1".
3716 19. Added "eqi" as a case-independent comparison operator.
3718 20. The saslauthd authentication condition could segfault if neither service
3719 nor realm was specified.
3721 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3722 error message that was logged was misleading, and incoming SMTP
3723 connections were dropped. The message is now more accurate, and temporary
3724 errors are given to SMTP connections.
3726 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3727 (except RSET) with a 421 error, until QUIT is received. However, it was
3728 failing to send a response to QUIT.
3730 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3731 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3732 ACL is now run after the helo_try_verify_hosts code.
3734 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3735 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3736 case-independent, so other case variants are also recognized). Apparently
3737 some systems use these upper case variants.
3739 25. If more than two messages were waiting for the same host, and a transport
3740 filter was specified for the transport, Exim sent two messages over the
3741 same TCP/IP connection, and then failed with "socket operation on non-
3742 socket" when it tried to send the third.
3744 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3746 27. The extern definition of crypt16() in expand.c was not being excluded when
3747 the OS had its own crypt16() function.
3749 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3750 as a preferred synonym for return_size_limit, both as an option and as an
3753 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3755 30. Changed the default configuration ACL to relax the local part checking rule
3756 for addresses that are not in any local domains. For these addresses,
3757 slashes and pipe symbols are allowed within local parts, but the sequence
3758 /../ is explicitly forbidden.
3760 31. SPA server authentication was not clearing the challenge buffer before
3763 32. log_message in a "warn" ACL statement was writing to the reject log as
3764 well as to the main log, which contradicts the documentation and doesn't
3765 seem right (because no rejection is happening). So I have stopped it.
3767 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3768 However, I am unable to do any testing of this.
3770 34. Fixed an infelicity in the appendfile transport. When checking directories
3771 for a mailbox, to see if any needed to be created, it was accidentally
3772 using path names with one or more superfluous leading slashes; tracing
3773 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3775 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3776 contents are added to the log line that is written for every discarded
3777 recipient. (Previously a log_message setting was ignored.)
3779 36. The ${quote: operator now quotes the string if it is empty.
3781 37. The install script runs exim in order to find its version number. If for
3782 some reason other than non-existence or emptiness, which it checks, it
3783 could not run './exim', it was installing it with an empty version number,
3784 i.e. as "exim-". This error state is now caught, and the installation is
3787 38. An argument was missing from the function that creates an error message
3788 when Exim fails to connect to the socket for saslauthd authentication.
3789 This could cause Exim to crash, or give a corrupted message.
3791 39. Added isip, isip4, and isip6 to ${if conditions.
3793 40. The ACL variables $acl_xx are now saved with the message, and can be
3794 accessed later in routers, transports, and filters.
3796 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3797 strings in the file are not string-expanded.
3799 42. If a MAIL command specified a SIZE value that was too large to fit into an
3800 int variable, the check against message_size_limit failed. Such values are
3801 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3802 day this will have to be increased, but I don't think I want to be around
3803 when emails are that large.
3810 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3811 iconv() is not standard in FreeBSD.
3813 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3814 IPv6 enabled. The observed symptom was a segmentation fault on return from
3815 the function os_common_find_running_interfaces() in src/os.c.
3817 3. In the check_special_case() function in daemon.c I had used "errno" as an
3818 argument name, which causes warnings on some systems. This was basically a
3819 typo, since it was named "eno" in the comments!
3821 4. The code that waits for the clock to tick (at a resolution of some fraction
3822 of a second) so as to ensure message-id uniqueness was always waiting for
3823 at least one whole tick, when it could have waited for less. [This is
3824 almost certainly not relevant at current processor speeds, where it is
3825 unlikely to ever wait at all. But we try to future-proof.]
3827 5. The function that sleeps for a time interval that includes fractions of a
3828 second contained a race. It did not block SIGALRM between setting the
3829 timer, and suspending (a couple of lines later). If the interval was short
3830 and the sigsuspend() was delayed until after it had expired, the suspension
3831 never ended. On busy systems this could lead to processes getting stuck for
3834 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3835 process, before it forks any delivery processes. The open lookup caching
3836 mechanism meant that the open file or database connection was passed into
3837 the delivery process. The problem was that delivery processes always tidy
3838 up cached lookup data. This could cause a problem for the next delivery
3839 process started by the queue runner, because the external queue runner
3840 process does not know about the closure. So the next delivery process
3841 still has data in the lookup cache. In the case of a file lookup, there was
3842 no problem because closing a file descriptor in a subprocess doesn't affect
3843 the parent. However, if the lookup was caching a connection to a database,
3844 the connection was closed, and the second delivery process was likely to
3845 see errors such as "PGSQL: query failed: server closed the connection
3846 unexpectedly". The problem has been fixed by closing all cached lookups
3847 in a queue runner before running a delivery process.
3849 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3850 seem to have the "const" qualifier which it has on other OS. I've
3853 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3854 *of the same type* (client or server) with the same public name that an
3855 error should be diagnosed.
3857 9. When Exim looked up a host name for an IP address, but failed to find the
3858 original IP address when looking up the host name (a safety check), it
3859 output the message "<ip address> does not match any IP for NULL", which was
3860 confusing, to say the least. The bug was that the host name should have
3861 appeared instead of "NULL".
3863 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3864 user that is built into the binary, and the -C or -D options is used, root
3865 privilege is dropped before the configuration file is read. In addition,
3866 logging is switched to stderr instead of the normal log files. If the
3867 configuration then re-defines the Exim user, the unprivileged environment
3868 is probably not what is expected, so Exim logs a panic warning message (but
3871 However, if deliver_drop_privilege is set, the unprivileged state may well
3872 be exactly what is intended, so the warning has been cut out in that case,
3873 and Exim is allowed to try to write to its normal log files.
3879 1. smtp_return_error_details was not giving details for temporary sender
3880 or receiver verification errors.
3882 2. Diagnose a configuration error if two authenticators have the same public
3885 3. Exim used not to create the message log file for a message until the first
3886 delivery attempt. This could be confusing when incoming messages were held
3887 for policy or load reasons. The message log file is now created at the time
3888 the message is received, and an initial "Received" line is written to it.
3890 4. The automatically generated man page for command line options had a minor
3891 bug that caused no ill effects; however, a more serious problem was that
3892 the procedure for building the man page automatically didn't always
3893 operate. Consequently, release 4.20 contains an out-of-date version. This
3894 shouldn't happen again.
3896 5. When building Exim with embedded Perl support, the script that builds the
3897 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3898 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3900 6. The freeze_tell option was not being used for messages that were frozen on
3901 arrival, either by an ACL or by local_scan().
3903 7. Added the smtp_incomplete_transaction log selector.
3905 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3906 was accepting AUTH without a new EHLO.
3908 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3909 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3912 10. Logging of TCP/IP connections (when configured) now happens in the main
3913 daemon process instead of the child process, so that the TCP/IP connection
3914 count is more accurate (but it can never be perfect).
3916 11. The use of "drop" in a nested ACL was not being handled correctly in the
3917 outer ACL. Now, if condition failure induced by the nested "drop" causes
3918 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3919 or "require"), the connection is dropped.
3921 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3922 that yield "discard" can now be used with an "accept" or a "discard" verb,
3923 but an error is generated for any others (because I can't see a useful way
3924 to define what should happen).
3926 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3927 are now processed in the same way as lines in the Exim configuration file.
3928 In particular, continuation lines are supported.
3930 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3932 15. Added -ti meaning -t -i.
3934 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3935 domains, and warn by logging if others are found.
3937 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3938 returned from GIFCONF when Exim is trying to find the list of interfaces on
3939 a host. The code in os.c has been modified to copy each ifreq to an aligned
3940 structure in all cases.
3942 Also, in some cases, the returned ifreq's were being copied to a 'struct
3943 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3944 means the last couple of bytes of an IPv6 address could be chopped if the
3945 ifreq contained only a normal sockaddr (14 bytes storage).
3947 18. Named domain lists were not supported in the hosts_treat_as_local option.
3948 An entry such as +xxxx was not recognized, and was treated as a literal
3951 19. Ensure that header lines added by a DATA ACL are included in the reject log
3952 if the ACL subsequently rejects the message.
3954 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3955 MD5 (which is deprecated).
3957 21. When testing a filter file using -bf, Exim was writing a message when it
3958 took the sender from a "From " line in the message, but it was not doing so
3959 when it took $return_path from a Return-Path: header line. It now does.
3961 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3962 with a valid header line field name (a series of printing characters
3963 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3965 23. Changed "disc" in the source to "disk" to conform to the documentation and
3966 the book and for uniformity.
3968 24. Ignore Sendmail's -Ooption=value command line item.
3970 25. When execve() failed while trying to run a command in a pipe transport,
3971 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3972 could be confused with a return value of 69 from the command itself. This
3973 has been changed to 127, the value the shell returns if it is asked to run
3974 a non-existent command. The wording for the related log line suggests a
3975 non-existent command as the problem.
3977 26. If received_header_text expands to an empty string, do not add a Received:
3978 header line to the message. (Well, it adds a token one on the spool, but
3979 marks it "old" so that it doesn't get used or transmitted.)
3981 27. Installed eximstats 1.28 (addition of -nt option).
3983 28. There was no check for failure on the call to getsockname() in the daemon
3984 code. This can fail if there is a shortage of resources on the system, with
3985 ENOMEM, for example. A temporary error is now given on failure.
3987 29. Contrary to the C standard, it seems that in some environments, the
3988 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3989 program. Exim now does this explicitly; it affects the formatting of
3990 timestamps using strftime().
3992 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3993 complaints. I've now initialized all the variables, to avoid this.
3995 32. Header lines added by a system filter were not being "seen" during
3996 transport-time rewrites.
3998 33. The info_callback() function passed to OpenSSL is set up with type void
3999 (*)(SSL *, int, int), as described somewhere. However, when calling the
4000 function (actually a macro) that sets it up, the type void(*)() is
4001 expected. I've put in a cast to prevent warnings from picky compilers.
4003 34. If a DNS black list lookup found a CNAME record, but there were no A
4004 records associated with the domain it pointed at, Exim crashed.
4006 35. If a DNS black list lookup returned more than one A record, Exim ignored
4007 all but the first. It now scans all returned addresses if a particular IP
4008 value is being sought. In this situation, the contents of the
4009 $dnslist_value variable are a list of all the addresses, separated by a
4012 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4013 to accept a host name and all its aliases if the forward lookup for any of
4014 them yielded the IP address of the incoming connection. Now it accepts only
4015 those names whose forward lookup yields the correct IP address. Any other
4016 names are discarded. This closes a loophole whereby a rogue DNS
4017 administrator could create reverse DNS records to break through a
4018 wildcarded host restriction in an ACL.
4020 37. If a user filter or a system filter that ran in a subprocess used any of
4021 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4022 the wrong values were passed to the pipe command ($thisaddress had the
4023 value of $0, $0 had the value of $1, etc). This bug was introduced by
4024 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4026 38. Improved the line breaking for long SMTP error messages from ACLs.
4027 Previously, if there was no break point between 40 and 75 characters, Exim
4028 left the rest of the message alone. Two changes have been made: (a) I've
4029 reduced the minimum length to 35 characters; (b) if it can't find a break
4030 point between 35 and 75 characters, it looks ahead and uses the first one
4031 that it finds. This may give the occasional overlong line, but at least the
4032 remaining text gets split now.
4034 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4035 file descriptors might be low, and that setting 1000 would always raise it.
4036 It turns out that in some environments, the limit is already over 1000 and
4037 that lowering it causes trouble. So now Exim takes care not to decrease it.
4039 40. When delivering a message, the value of $return_path is set to $sender_
4040 address at the start of routing (routers may change the value). By an
4041 oversight, this default was not being set up when an address was tested by
4042 -bt or -bv, which affected the outcome if any router or filter referred to
4045 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4046 log only when "message" or "log_message" are set. However, if one of the
4047 conditions was an address verification, or a call to a nested ACL, the
4048 messages generated by the underlying test were being passed through. This
4049 no longer happens. The underlying message is available in $acl_verify_
4050 message for both "message" and "log_message" expansions, so it can be
4051 passed through if needed.
4053 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4054 new expansion $bh_ to give the encoded byte string without charset
4055 translation. Translation happens only if iconv() is available; HAVE_ICONV
4056 indicates this at build time. HEADERS_CHARSET gives the charset to
4057 translate to; headers_charset can change it in the configuration, and
4058 "headers charset" can change it in an individual filter file.
4060 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4061 that creates RFC 2047 encoded "words" labels them as that charset instead
4062 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4063 expansion operator; (ii) when Exim creates a From: line for a local
4064 message; (iii) when a header line is rewritten to include a "phrase" part.
4066 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4067 buggy, causing it to skip the first lines of messages whose message ID
4068 ended in 'D'. This would not have bitten before Exim release 4.14, because
4069 message IDs were unlikely to end in 'D' before then. The effect was to have
4070 incorrect size information for certain domains.
4072 45. #include "config.h" was missing at the start of the crypt16.c module. This
4073 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4076 46. If there was a timeout during a "random" callout check, Exim treated it as
4077 a failure of the random address, and carried on sending RSET and the real
4078 address. If the delay was just some slowness somewhere, the response to the
4079 original RCPT would be taken as a response to RSET and so on, causing
4080 mayhem of various kinds.
4082 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4083 when I implemented it. It didn't allow for the fact that some option values
4084 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4085 the right test for positive values.
4087 48. Domain names in DNS records are case-independent. Exim always looks them up
4088 in lower case. Some resolvers return domain names in exactly the case they
4089 appear in the zone file, that is, they may contain uppercase letters. Not
4090 all resolvers do this - some return always lower case. Exim was treating a
4091 change of case by a resolver as a change of domain, similar to a widening
4092 of a domain abbreviation. This triggered its re-routing code and so it was
4093 trying to route what was effectively the same domain again. This normally
4094 caused routing to fail (because the router wouldn't handle the domain
4095 twice). Now Exim checks for this case specially, and just changes the
4096 casing of the domain that it ultimately uses when it transmits the message
4099 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4102 50. If a filter generated a file delivery with a non-absolute name (possible if
4103 no home directory exists for the router), the forbid_file option was not
4106 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4107 the existing equality matching.
4109 52. Exim was using ints instead of ino_t variables in some places where it was
4110 dealing with inode numbers.
4112 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4113 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4114 called TMPDIR, and if it finds it is different, it changes its value.
4116 54. The smtp_printf() function is now made available to local_scan() so
4117 additional output lines can be written before returning. There is also an
4118 smtp_fflush() function to enable the detection of a dropped connection.
4119 The variables smtp_input and smtp_batched_input are exported to
4122 55. Changed the default runtime configuration: the message "Unknown user"
4123 has been removed from the ACL, and instead placed on the localuser router,
4124 using the cannot_route_message feature. This means that any verification
4125 failures that generate their own messages won't get overridden. Similarly,
4126 the "Unrouteable address" message that was in the ACL for unverifiable
4127 relay addresses has also been removed.
4129 56. Added hosts_avoid_esmtp to the smtp transport.
4131 57. The exicyclog script was not checking for the esoteric option
4132 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4133 will work only if exicyclog is run under the appropriate euid.
4135 58. Following a discussion on the list, the rules by which Exim recognises line
4136 endings on incoming messages have been changed. The -dropcr and drop_cr
4137 options are now no-ops, retained only for backwards compatibility. The
4138 following line terminators are recognized: LF CRLF CR. However, special
4139 processing applies to CR:
4141 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4142 nor a local message in the state where . is a terminator.
4144 (ii) If a bare CR is encountered in a header line, an extra space is added
4145 after the line terminator so as not to end the header. The reasoning
4146 behind this is that bare CRs in header lines are most likely either
4147 to be mistakes, or people trying to play silly games.
4149 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4150 was being incorrectly given as 18 bytes larger than it should have been.
4151 This is a VOB (very old bug).
4153 60. This may never have affected anything current, but just in case it has:
4154 When the local host is found other than at the start of a list of hosts,
4155 the local host, those with the same MX, and any that follow, are discarded.
4156 When the list in question was part of a longer list of hosts, the following
4157 hosts (not currently being processed) were also being discarded. This no
4158 longer happens. I'm not sure if this situation could ever has previously
4161 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4164 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4165 header. This is something that is recommended in a new Internet Draft, and
4166 is something that is documented as being done by Sendmail. There are two
4167 possible values. For messages generated by the autoreply transport, Exim
4170 Auto-Submitted: auto-replied
4172 whereas for all other generated messages (e.g. bounces) it adds
4174 Auto-Submitted: auto-generated
4176 63. The "personal" condition in filters now includes a test for the
4177 Auto-Submitted: header. If it contains the string "auto-" the message it
4178 not considered personal.
4180 64. Added rcpt_include_affixes as a generic transport option.
4182 65. Added queue_only_override (default true).
4184 66. Added the syslog_duplication option.
4186 67. If what should have been the first header line of a message consisted of
4187 a space followed by a colon, Exim was mis-interpreting it as a header line.
4188 It isn't of course - it is syntactically invalid and should therefore be
4189 treated as the start of the message body. The misbehaviour could have
4190 caused a number of strange effects, including loss of data in subsequent
4191 header lines, and spool format errors.
4193 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4194 client host had authenticated. This control can now be exercised by an ACL
4195 for more flexibility.
4197 69. By default, callouts do not happen when testing with -bh. There is now a
4198 variant, -bhc, which does actually run the callout code, including
4199 consulting and updating the callout cache.
4201 70. Added support for saslauthd authentication, courtesy of Alexander
4204 71. If statvfs() failed on the spool or log directories while checking their
4205 size for availability, Exim confusingly gave the error "space shortage".
4206 Furthermore, in debugging mode it crashed with a floating point exception.
4207 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4208 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4209 serious problem, Exim now writes to the main and panic logs when this
4210 happens, with details of the failure. It then refuses to accept the
4211 incoming message, giving the message "spool directory problem" or "log
4212 directory problem" with a 421 code for SMTP messages.
4214 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4215 0, 1, and 2 exist, because some OS complain for execs without them (see
4216 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4217 descriptors. However, the code omitted to check that the open succeeded,
4218 causing mysterious errors if for some reason the permissions on /dev/null
4219 got screwed. Now Exim writes a message to the main and panic logs, and
4220 bombs out if it can't open /dev/null.
4222 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4223 interact so that it is all more flexible. It is supposed to remain
4224 backwards compatible. Also added extra_local_interfaces.
4226 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4227 to bomb out with an assertion failure - to the client this appears as a
4228 connection drop. This problem occurs in the part of the code that was taken
4229 from the Samba project. Fortunately, the assertion is in a very simple
4230 function, so I have fixed this by reproducing the function inline in the
4231 one place where it is called, and arranging for authentication to fail
4232 instead of killing the process with assert().
4234 75. The SPA client code was not working when the server requested OEM rather
4235 than Unicode encoding.
4237 76. Added code to make require_files with a specific uid setting more usable in
4238 the case where statting the file as root fails - usually a non-root-mounted
4239 NFS file system. When this happens and the failure is EACCES, Exim now
4240 forks a subprocess and does the per-uid checking as the relevant uid.
4242 77. Added process_log_path.
4244 78. If log_file_path was not explicitly set, a setting of check_log_space or
4245 check_log_inodes was ignored.
4247 79. If a space check for the spool or log partitions fails, the incident is now
4248 logged. Of course, in the latter case the data may get lost...
4250 80. Added the %p formatting code to string_format() so that it can be used to
4251 print addresses in debug_print(). Adjusted all the address printing in the
4252 debugging in store.c to use %p rather than %d.
4254 81. There was a concern that a line of code in smtp_in.c could overflow a
4255 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4256 initially expressed, the concern was not well-founded, because trailing
4257 spaces are removed early. However, if the trailing spaces were followed by
4258 a NULL, they did not get removed, so the overflow was possible. Two fixes
4261 (a) I re-wrote the offending code in a cleaner fashion.
4262 (b) If an incoming SMTP command contains a NULL character, it is rejected
4265 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4266 runs initgroups(), so that if the Exim user is in any additional groups,
4267 they will be used during message reception.
4273 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob