1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.440 2006/12/04 15:00:19 ph10 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
10 leftover -K file (the existence of which was triggered by #402).
11 While we were at it, introduced process PID as part of the -K
12 filename. This should rule out race conditions when creating
15 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
16 processing considerably. Previous code took too long for large mails,
17 triggering a timeout which in turn triggers #401.
19 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
20 in the DK code in transports.c. sendfile() is not really portable,
21 hence the _LINUX specificness.
23 TF/01 In the add_headers option to the mail command in an Exim filter,
24 there was a bug that Exim would claim a syntax error in any
25 header after the first one which had an odd number of characters
28 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
29 callout verification, Exim cached a "reject" for the entire domain. This
30 is correct for most verifications, but it is not correct for a recipient
31 verification with use_sender or use_postmaster set, because in that case
32 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
33 case of MAIL FROM:<> rejection from other early rejections (e.g.
34 rejection of HELO). When verifying a recipient using a non-null MAIL
35 address, the cache is ignored if it shows MAIL FROM:<> rejection.
36 Whatever the result of the callout, the value of the domain cache is
37 left unchanged (for any other kind of callout, getting as far as trying
38 RCPT means that the domain itself is ok).
40 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
43 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
44 manifest itself as EPIPE rather than ECONNECT. When tidying away a
45 session, the daemon ignores ECONNECT errors and logs others; it now
46 ignores EPIPE as well.
48 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
49 (quoted-printable decoding).
51 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
52 later the small subsequent patch to fix an introduced bug.
54 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
56 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
58 PH/08 An error is now given if message_size_limit is specified negative.
60 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
61 to be given (somewhat) arbitrary names.
63 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
66 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
67 miscellaneous code fixes
69 PH/10 Added the log_reject_target ACL modifier to specify where to log
72 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
73 hostname. This is wrong, because it relates to the incoming message (and
74 probably the interface on which it is arriving) and not to the outgoing
75 callout (which could be using a different interface). This has been
76 changed to use the value of the helo_data option from the smtp transport
77 instead - this is what is used when a message is actually being sent. If
78 there is no remote transport (possible with a router that sets up host
79 addresses), $smtp_active_hostname is used.
81 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
82 tweaks were necessary in order to get it to work (see also 21 below):
83 (a) The code assumed that strncpy() returns a negative number on buffer
84 overflow, which isn't the case. Replaced with Exim's string_format()
86 (b) There were several signed/unsigned issues. I just did the minimum
87 hacking in of casts. There is scope for a larger refactoring.
88 (c) The code used strcasecmp() which is not a standard C function.
89 Replaced with Exim's strcmpic() function.
90 (d) The code set only $1; it now sets $auth1 as well.
91 (e) A simple test gave the error "authentication client didn't specify
92 service in request". It would seem that Dovecot has changed its
93 interface. Fortunately there's a specification; I followed it and
94 changed what the client sends and it appears to be working now.
96 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
99 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
100 address A is aliased to B and C, where B exists and C does not. Without
101 -v the output is "A verified" because verification stops after a
102 successful redirection if more than one address is generated. However,
103 with -v the child addresses are also verified. Exim was outputting "A
104 failed to verify" and then showing the successful verification for C,
105 with its parentage. It now outputs "B failed to verify", showing B's
106 parentage before showing the successful verification of C.
108 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
109 look up a TXT record in a specific list after matching in a combined
112 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
113 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
114 they consult the DNS. I had assumed they would set it the way they
115 wanted; and indeed my experiments on Linux seem to show that in some
116 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
117 To be on the safe side, however, I have now made the interface to
118 host_find_byname() similar to host_find_bydns(), with an argument
119 containing the DNS resolver options. The host_find_byname() function now
120 sets these options at its start, just as host_find_bydns() does. The smtp
121 transport options dns_qualify_single and dns_search_parents are passed to
122 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
123 of host_find_byname() use the default settings of RES_DEFNAMES
124 (qualify_single) but not RES_DNSRCH (search_parents).
126 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
127 spool_read_header() do less string testing, by means of a preliminary
128 switch on the second character of optional "-foo" lines. (This is
129 overdue, caused by the large number of possibilities that now exist.
130 Originally there were few.) While I was there, I also converted the
131 str(n)cmp tests so they don't re-test the leading "-" and the first
132 character, in the hope this might squeeze out yet more improvement.
134 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
135 flag allowing group syntax was set by the header_syntax check but not
136 turned off, possible causing trouble later; (2) The flag was not being
137 set at all for the header_verify test, causing "group"-style headers to
138 be rejected. I have now set it in this case, and also caused header_
139 verify to ignore an empty address taken from a group. While doing this, I
140 came across some other cases where the code for allowing group syntax
141 while scanning a header line wasn't quite right (mostly, not resetting
142 the flag correctly in the right place). These bugs could have caused
143 trouble for malformed header lines. I hope it is now all correct.
145 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
146 with the "reply" argument non-NULL. The code, however (which originally
147 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
148 but it didn't always do it. This confused somebody who was copying the
149 code for some other use. I have removed all the tests.
151 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
152 feature that was used to support insecure browsers during the U.S. crypto
153 embargo. It requires special client support, and Exim is probably the
154 only MTA that supported it -- and would never use it because real RSA is
155 always available. This code has been removed, because it had the bad
156 effect of slowing Exim down by computing (never used) parameters for the
157 RSA_EXPORT functionality.
159 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
160 authenticator to fail if there's a tab character in the incoming data
161 (there should never be unless someone is messing about, as it's supposed
162 to be base64-encoded). Also added, on Timo's advice, the "secured" option
163 if the connection is using TLS or if the remote IP is the same as the
164 local IP, and the "valid-client-cert option" if a client certificate has
167 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
168 authenticators. This can be used for authorization after authentication
169 succeeds. (In the case of plaintext, it servers for both authentication
172 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
173 if any retry times were supplied.
175 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
176 connection, where there is no client IP address to check. In this
177 situation, the verify now always succeeds.
179 PH/25 Applied John Jetmore's -Mset patch.
181 PH/26 Added -bem to be like -Mset, but loading a message from a file.
183 PH/27 In a string expansion for a processed (not raw) header when multiple
184 headers of the same name were present, leading whitespace was being
185 removed from all of them, but trailing whitespace was being removed only
186 from the last one. Now trailing whitespace is removed from each header
187 before concatenation. Completely empty headers in a concatenation (as
190 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
191 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
193 PH/29 [Removed. This was a change that I later backed out, and forgot to
194 correct the ChangeLog entry (that I had efficiently created) before
195 committing the later change.]
197 PH/30 Exim was sometimes attempting to deliver messages that had suffered
198 address errors (4xx response to RCPT) over the same connection as other
199 messages routed to the same hosts. Such deliveries are always "forced",
200 so retry times are not inspected. This resulted in far too many retries
201 for the affected addresses. The effect occurred only when there were more
202 hosts than the hosts_max_try setting in the smtp transport when it had
203 the 4xx errors. Those hosts that it had tried were not added to the list
204 of hosts for which the message was waiting, so if all were tried, there
205 was no problem. Two fixes have been applied:
207 (i) If there are any address or message errors in an SMTP delivery, none
208 of the hosts (tried or untried) are now added to the list of hosts
209 for which the message is waiting, so the message should not be a
210 candidate for sending over the same connection that was used for a
211 successful delivery of some other message. This seems entirely
212 reasonable: after all the message is NOT "waiting for some host".
213 This is so "obvious" that I'm not sure why it wasn't done
214 previously. Hope I haven't missed anything, but it can't do any
215 harm, as the worst effect is to miss an optimization.
217 (ii) If, despite (i), such a delivery is accidentally attempted, the
218 routing retry time is respected, so at least it doesn't keep
219 hammering the server.
221 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
222 in ${readsocket because some servers need this prod.
224 PH/32 Added some extra debug output when updating a wait-xxx database.
226 PH/33 The hint "could be header name not terminated by colon", which has been
227 given for certain expansion errors for a long time, was not being given
228 for the ${if def:h_colon_omitted{... case.
230 PH/34 The spec says: "With one important exception, whenever a domain list is
231 being scanned, $domain contains the subject domain." There was at least
232 one case where this was not true.
234 PH/35 The error "getsockname() failed: connection reset by peer" was being
235 written to the panic log as well as the main log, but it isn't really
236 panic-worthy as it just means the connection died rather early on. I have
237 removed the panic log writing for the ECONNRESET error when getsockname()
240 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
241 runs only) independently of the message's sender address. This meant
242 that, if the 4xx error was in fact related to the sender, a different
243 message to the same recipient with a different sender could confuse
244 things. In particualar, this can happen when sending to a greylisting
245 server, but other circumstances could also provoke similar problems.
246 I have changed the default so that the retry time for these errors is now
247 based a combination of the sender and recipient addresses. This change
248 can be overridden by setting address_retry_include_sender=false in the
251 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
252 remote server are returned as part of bounce messages. This was not
253 happening for LMTP over a pipe (the lmtp transport), but now it is the
254 same for both kinds of LMTP.
256 PH/38 Despite being documented as not happening, Exim was rewriting addresses
257 in header lines that were in fact CNAMEs. This is no longer the case.
259 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
260 and queue runs started by the daemon processed all messages. This has
261 been fixed so that -R and -S can now usefully be given with -q<time>.
263 PH/40 Import PCRE release 6.7 (fixes some bugs).
265 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
267 PH/42 Give an error if -q is specified more than once.
269 PH/43 Renamed the variables $interface_address and $interface_port as
270 $received_ip_address and $received_port, to make it clear that these
271 values apply to message reception, and not to the outgoing interface when
272 a message is delivered. (The old names remain recognized, of course.)
274 PH/44 There was no timeout on the connect() call when using a Unix domain
275 socket in the ${readsocket expansion. There now is.
277 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
278 be meaningful with "accept".
280 SC/01 Eximstats V1.43
281 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
283 SC/02 Eximstats V1.44
284 Use a glob alias rather than an array ref in the generated
285 parser. This improves both readability and performance.
287 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
288 Collect SpamAssassin and rejection statistics.
289 Don't display local sender or destination tables unless
290 there is data to show.
291 Added average volumes into the top table text output.
293 SC/04 Eximstats V1.46
294 Collect data on the number of addresses (recipients)
295 as well as the number of messages.
297 SC/05 Eximstats V1.47
298 Added 'Message too big' to the list of mail rejection
299 reasons (thanks to Marco Gaiarin).
301 SC/06 Eximstats V1.48
302 Mainlog lines which have GMT offsets and are too short to
303 have a flag are now skipped.
305 SC/07 Eximstats V1.49 (Alain Williams)
306 Added the -emptyok flag.
308 SC/08 Eximstats V1.50
309 Fixes for obtaining the IP address from reject messages.
311 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
312 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
313 whitesspace changes from 4.64-PH/27
315 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
318 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
319 are found, allow negative numbers in numeric criteria)
321 JJ/06 exipick.20061117.2, added new $message_body_missing variable
323 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
324 to match changes made in 4.64-PH/43
326 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
328 PH/47 Put in an explicit test for a DNS lookup of an address record where the
329 "domain" is actually an IP address, and force a failure. This locks out
330 those revolvers/nameservers that support "A-for-A" lookups, in
331 contravention of the specifications.
333 PH/48 When a host name was looked up from an IP address, and the subsequent
334 forward lookup of the name timed out, the host name was left in
335 $sender_host_name, contrary to the specification.
337 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
338 restricted to single-key lookups, Exim was not diagnosing an error if
339 * or *@ was used with a query-style lookup.
345 SC/01 Use a glob alias rather than an array ref in eximstats generated
346 parser. This improves both readability and performance.
348 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
349 Don't display local sender or destination tables in eximstats unless
350 there is data to show.
351 Added average volumes into the eximstats top table text output.
353 SC/03 Collect data on the number of addresses (recipients) as well
354 as the number of messages in eximstats.
356 TF/01 Correct an error in the documentation for the redirect router. Exim
357 does (usually) call initgroups() when daemonizing.
359 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
360 with consistent privilege compared to when running as a daemon.
362 TF/03 Note in the spec that $authenticated_id is not set for local
363 submissions from trusted users.
365 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
366 Thanks to Dean Brooks <dean@iglou.com> for the patch.
368 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
369 by adding some example configuration directives to the default
370 configuration file. A little bit of work is required to uncomment the
371 directives and define how usernames and passwords are checked, but
372 there is now a framework to start from.
374 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
375 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
376 without this. I don't know how relevant this is to other LDAP libraries.
378 PH/02 Add the verb name to the "unknown ACL verb" error.
380 PH/03 Magnus Holmgren's patch for filter_prepend_home.
382 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
384 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
385 directory not expanded when it should be if an expanded home directory
386 was set for the address (which is overridden by the transport).
388 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
391 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
392 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
393 because it is too late at that time, and has no effect.
395 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
396 security issue with \' (bugzilla #107). I could not use the
397 PQescapeStringConn() function, because it needs a PGconn value as one of
400 PH/08 When testing addresses using -bt, indicate those final addresses that
401 are duplicates that would not cause an additional delivery. At least one
402 person was confused, thinking that -bt output corresponded to deliveries.
403 (Suppressing duplicates isn't a good idea as you lose the information
404 about possibly different redirections that led to the duplicates.)
406 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
407 systems where poll() doesn't work, in particular OS X.
409 PH/10 Added more information to debugging output for retry time not reached.
411 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
412 operations in malware.c.
414 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
417 PH/13 If write_rejectlog was set false when logging was sent to syslog with
418 syslog_duplication set false, log lines that would normally be written
419 both the the main log and to the reject log were not written to syslog at
422 PH/14 In the default configuration, change the use of "message" in ACL warn
423 statements to "add_header".
425 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
426 not followed by a command (e.g. "seen endif").
428 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
429 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
432 PH/17 Added extra conditions to the default value of delay_warning_condition
436 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
437 { match{$h_precedence:}{(?i)bulk|list|junk} } \
438 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
441 The Auto-Submitted: and various List- headers are standardised, whereas I
442 don't think Precedence: ever was.
444 PH/18 Refactored debugging code in route_finduser() to show more information,
445 in particular, the error code if getpwnam() issues one.
447 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
448 This is apparently needed in addition to the PH/07 change above to avoid
449 any possible encoding problems.
451 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
452 but not after initializing Perl.
454 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
455 output them only if debugging. By default they are written stderr,
456 apparently, which is not desirable.
458 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
461 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
464 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
466 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
467 authenticated or an ident call has been made. Suppress the default
468 values for $authenticated_id and $authenticated_sender (but permit -oMai
469 and -oMas) when testing with -bh.
471 PH/24 Re-jigged the order of the tests in the default configuration so that the
472 tests for valid domains and recipients precede the DNS black list and CSA
473 tests, on the grounds that those ones are more expensive.
475 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
476 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
477 This bug exists in every version of Exim that I still have, right back to
480 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
481 However, an attempt to turn on case-sensitivity in a regex key by
482 including (?-i) didn't work because the subject string was already
483 lowercased, and the effects were non-intuitive. It turns out that a
484 one-line patch can be used to allow (?-i) to work as expected.
490 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
491 other effects) broke the use of negated acl sub-conditions.
493 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
496 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
497 "Deny" causes Exim to reject the incoming connection with a 554 error.
498 Unfortunately, if there is a major crisis, such as a disk failure,
499 tcp-wrappers gives "deny", whereas what one would like would be some
500 kind of temporary error. A kludge has been added to help with this.
501 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
502 554 error is used if errno is still zero or contains ENOENT (which occurs
503 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
506 PH/03 Add -lutil to the default FreeBSD LIBS setting.
508 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
509 errors. Otherwise a message that provokes a temporary error (when other
510 messages do not) can cause a whole host to time out.
512 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
513 addresses were routed directly to files or pipes from a redirect router.
514 File deliveries just didn't batch; pipe deliveries might have suffered
517 PH/06 A failure to get a lock for a hints database would erroneously always say
518 "Failed to get write lock", even when it was really a read lock.
520 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
521 of 0600. This has been changed to use the value of the lockfile_mode
522 option (which defaults to 0600).
524 PH/08 Applied small patch from the Sieve maintainer.
526 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
527 folder from quota calculations, a direct delivery into this folder messed
528 up the contents of the maildirsize file. This was because the regex was
529 used only to exclude .Trash (or whatever) when the size of the mailbox
530 was calculated. There was no check that a delivery was happening into an
531 excluded directory. This bug has been fixed by ignoring all quota
532 processing for deliveries into excluded directories.
534 PH/10 Added the maildirfolder_create_regex option to appendfile.
540 PH/01 The code for finding all the local interface addresses on a FreeBSD
541 system running IPv6 was broken. This may well have applied to all BSD
542 systems, as well as to others that have similar system calls. The broken
543 code found IPv4 interfaces correctly, but gave incorrect values for the
544 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
545 that it would not match correctly against @[] and not recognize the IPv6
548 PH/02 The ipliteral router was not recognizing addresses of the form user@
549 [ipv6:....] because it didn't know about the "ipv6:" prefix.
551 PH/03 Added disable_ipv6.
553 PH/04 Changed $reply_address to use the raw form of the headers instead of the
554 decoded form, because it is most often used to construct To: headers
555 lines in autoreplies, and the decoded form may well be syntactically
556 invalid. However, $reply_address has leading white space removed, and all
557 newlines turned into spaces so that the autoreply transport does not
560 PH/05 If group was specified without a user on a router, and no group or user
561 was specified on a transport, the group from the router was ignored.
563 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
564 for visible compile-time settings that can be used to change these
565 numbers, for those that want even more. Backwards compatibility with old
566 spool files has been maintained. However, going back to a previous Exim
567 release will lost any variables that are in spool files.
569 PH/07 Two small changes when running in the test harness: increase delay when
570 passing a TCP/IP connection to a new process, in case the original
571 process has to generate a bounce, and remove special handling of
572 127.0.0.2 (sic), which is no longer necessary.
574 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
575 be the same on different OS.
577 PH/09 Moved a debug statement in filter processing to avoid a race problem when
580 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
581 whether --show-vars was specified or not
583 JJ/02 exipick: Added support for new ACL variable spool format introduced
586 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
587 syntactically invalid From: or Reply-to: line, and a filter used this to
588 generate an autoreply, and therefore failed to obtain an address for the
589 autoreply, Exim could try to deliver to a non-existent relative file
590 name, causing unrelated and misleading errors. What now happens is that
591 it logs this as a hard delivery error, but does not attempt to create a
594 PH/11 The exinext utility has a -C option for testing purposes, but although
595 the given file was scanned by exinext itself; it wasn't being passed on
596 when Exim was called.
598 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
599 an end-of-file indication when reading a command response.
601 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
602 compiled. In many other places in Exim, IPv6 addresses are always
603 recognized, so I have changed this. It also means that IPv4 domain
604 literals of the form [IPV4:n.n.n.n] are now always recognized.
606 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
607 used if the router is not running as root, for example, when verifying at
608 ACL time, or when using -bh. The debugging output from this situation was
609 non-existent - all you got was a failure to exec. I have made two
612 (a) Failures to set uid/gid, the current directory, or a process leader
613 in a subprocess such as that created by queryprogram now generate
614 suitable debugging ouput when -d is set.
616 (b) The queryprogram router detects when it is not running as root,
617 outputs suitable debugging information if -d is set, and then runs
618 the subprocess without attempting to change uid/gid.
620 PH/15 Minor change to Makefile for building test_host (undocumented testing
623 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
624 additional section of a DNS packet that returns MX or SRV records.
625 Instead, it always explicitly searches for A/AAAA records. This avoids
626 major problems that occur when a DNS server includes only records of one
627 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
628 fixed another bug: if SRV records were looked up and the corresponding
629 address records were *not* found in the additional section, the port
630 values from the SRV records were lost.
632 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
633 using the correct key (the original address) when searching the retry
634 rules in order to find which one to use for generating the retry hint.
636 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
637 adding the default one. Similarly, if it contains a Reply-To: header, the
638 errors_reply_to option, if set, is not used.
640 PH/19 When calculating a retry time, Exim used to measure the "time since
641 failure" by looking at the "first failed" field in the retry record. Now
642 it does not use this if it is later than than the arrival time of the
643 message. Instead it uses the arrival time. This makes for better
644 behaviour in cases where some deliveries succeed, thus re-setting the
645 "first failed" field. An example is a quota failure for a huge message
646 when small messages continue to be delivered. Without this change, the
647 "time since failure" will always be short, possible causing more frequent
648 delivery attempts for the huge message than are intended.
649 [Note: This change was subsequently modified - see PH/04 for 4.62.]
651 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
652 $1, $2, $3) because the numerical variables can be reset during some
653 expansion items (e.g. "match"), thereby losing the authentication data.
655 PH/21 Make -bV show the size of off_t variables so that the test suite can
656 decide whether to run tests for quotas > 2G.
658 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
659 mailbox_size, and mailbox_filecount in the appendfile transport. If a
660 filecount value is greater than 2G or if a quota value is greater than 2G
661 on a system where the size of off_t is not greater than 4, a panic error
664 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
665 never match. The debug and -bh output now contains an explicit error
666 message indicating a malformed IPv4 address or mask.
668 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
669 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
672 PH/25 Do not write to syslog when running in the test harness. The only
673 occasion when this arises is a failure to open the main or panic logs
674 (for which there is an explicit test).
676 PH/26 Added the /no_tell option to "control=freeze".
678 PH/27 If a host name lookup failed very early in a connection, for example, if
679 the IP address matched host_lookup and the reverse lookup yielded a name
680 that did not have a forward lookup, an error message of the form "no IP
681 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
682 could be logged. Now it outputs the IP address instead of "NULL".
684 PH/28 An enabling patch from MH: add new function child_open_exim2() which
685 allows the sender and the authenticated sender to be set when
686 submitting a message from within Exim. Since child_open_exim() is
687 documented for local_scan(), the new function should be too.
689 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
690 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
691 results in an empty string is now treated as unset.
693 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
695 PH/31 Added sender_verify_fail logging option.
697 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
698 needed by RFC 822 but not by RFC 2822 was commented out. I have now
699 tidied the source and removed it altogether.
701 PH/33 When a queue run was abandoned because the load average was too high, a
702 log line was always written; now it is written only if the queue_run log
703 selector is set. In addition, the log line for abandonment now contains
704 information about the queue run such as the pid. This is always present
705 in "start" and "stop" lines but was omitted from the "abandon" line.
707 PH/34 Omit spaces between a header name and the colon in the error message that
708 is given when verify = headers_syntax fails (if there are lots of them,
709 the message gets confusing).
711 PH/35 Change the default for dns_check_names_pattern to allow slashes within
712 names, as there are now some PTR records that contain slashes. This check
713 is only to protect against broken name servers that fall over on strange
714 characters, so the fact that it applies to all lookups doesn't matter.
716 PH/36 Now that the new test suite is complete, we can remove some of the
717 special code in Exim that was needed for the old test suite. For example,
718 sorting DNS records because real resolvers return them in an arbitrary
719 order. The new test suite's fake resolver always returns records in the
722 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
723 bounces) except when queue_only is set, to avoid logging races between
724 the different processes.
726 PH/38 Panic-die if .include specifies a non-absolute path.
728 PH/39 A tweak to the "H" retry rule from its user.
730 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
731 a label. They prevented compilation on older perls.
733 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
734 a warning to be raised on newish perls.
736 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
737 on queue. Changes to match documented behaviour of showing count of
738 messages matching specified criteria.
740 PH/40 Changed the default ident timeout from 30s to 5s.
742 PH/41 Added support for the use of login_cap features, on those BSD systems
743 that have them, for controlling the resources used by pipe deliveries.
745 PH/42 The content-scanning code uses fopen() to create files in which to put
746 message data. Previously it was not paying any attention to the mode of
747 the files. Exim runs with umask(0) because the rest of the code creates
748 files with open(), and sets the required mode explicitly. Thus, these
749 files were ending up world-writeable. This was not a big issue, because,
750 being within the spool directory, they were not world-accessible. I have
751 created a function called modefopen, which takes an additional mode
752 argument. It sets umask(777), creates the file, chmods it to the required
753 mode, then resets the umask. All the relevant calls to fopen() in the
754 content scanning code have been changed to use this function.
756 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
757 to 24 hours. This avoids potential overflow problems when processing G
758 and H retry rules. I suspect nobody ever tinkers with this value.
760 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
762 PH/45 When the plaintext authenticator is running as a client, the server's
763 challenges are checked to ensure they are valid base64 strings. By
764 default, the authentication attempt is cancelled if an invalid string is
765 received. Setting client_ignore_invalid_base64 true ignores these errors.
766 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
767 they are received. Thus, the responses can be made to depend on the
768 challenges. If an invalid string is ignored, an empty string is placed in
771 PH/46 Messages that are created by the autoreply transport now contains a
772 References: header, in accordance with RFCs 2822 and 3834.
774 PH/47 Added authenticated_sender_force to the smtp transport.
776 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
778 PH/49 Installed latest patch from the Sieve maintainer.
780 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
781 was also set, the appendfile transport was unnecessarily scanning a
782 directory of message files (e.g. for maildir delivery) to find the count
783 of files (along with the size), even though it did not need this
784 information. It now does the scan only if it needs to find either the
785 size of the count of files.
787 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
789 PH/52 Two bugs concerned with error handling when the smtp transport is
792 (i) Exim was not creating retry information for temporary errors given
793 for individual recipients after the DATA command when the smtp transport
794 was used in LMTP mode. This meant that they could be retried too
795 frequently, and not timed out correctly.
797 (ii) Exim was setting the flag that allows error details to be returned
798 for LMTP errors on RCPT commands, but not for LMTP errors for individual
799 recipients that were returned after the DATA command.
801 PH/53 This is related to PH/52, but is more general: for any failing address,
802 when detailed error information was permitted to be returned to the
803 sender, but the error was temporary, then after the final timeout, only
804 "retry timeout exceeded" was returned. Now it returns the full error as
805 well as "retry timeout exceeded".
807 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
808 do this, and (what is worse) MTAs that accept it.
810 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
811 will now be deprecated.
813 PH/56 New os.c-cygwin from the Cygwin maintainer.
815 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
816 formats (previously only available in exim formats via -bpr, -bpru,
817 and -bpra. Now also available in native and exiqgrep formats)
819 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
820 with very large, slow to parse queues
822 JJ/08 exipick: added ! as generic prefix to negate any criteria format
824 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
826 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
827 responses to authentication challenges, though it was showing the
828 challenges; (ii) I've removed the CR characters from the debug output for
831 PH/58 Allow for the insertion of a newline as well as a space when a string
832 is turned into more than one encoded-word during RFC 2047 encoding. The
833 Sieve code now uses this.
835 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
836 data_4xx, lost_connection, tls_required.
838 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
839 message was being sent as an SMTP response.
841 PH/61 Add -l and -k options to exicyclog.
843 PH/62 When verifying, if an address was redirected to one new address, so that
844 verification continued, and the new address failed or deferred after
845 having set something in $address_data, the value of $address_data was not
846 passed back to the ACL. This was different to the case when no
847 redirection occurred. The value is now passed back in both cases.
849 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
850 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
851 use login_cap.h, so on its own it isn't the distinguishing feature. The
852 new name refers directly to the setclassresources() function.
854 PH/65 Added configuration files for NetBSD3.
856 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
858 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
859 is preferred over IPv4.
861 PH/68 The bounce_return_message and bounce_return_body options were not being
862 honoured for bounces generated during the reception of non-SMTP messages.
863 In particular, this applied to messages rejected by the ACL. This bug has
864 been fixed. However, if bounce_return_message is true and bounce_return_
865 body is false, the headers that are returned for a non-SMTP message
866 include only those that have been read before the error was detected.
867 (In the case of an ACL rejection, they have all been read.)
869 PH/69 The HTML version of the specification is now built in a directory called
870 spec_html instead of spec.html, because the latter looks like a path with
871 a MIME-type, and this confuses some software.
873 PH/70 Catch two compiler warnings in sieve.c.
875 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
876 function verify_get_ident() calls ip_connect() to connect a socket, but
877 if the "connect()" function timed out, ip_connect() used to close the
878 socket. However, verify_get_ident() also closes the socket later, and in
879 between Exim writes to the log, which may get opened at this point. When
880 the socket was closed in ip_connect(), the log could get the same file
881 descriptor number as the socket. This naturally causes chaos. The fix is
882 not to close the socket in ip_connect(); the socket should be closed by
883 the function that creates it. There was only one place in the code where
884 this was missing, in the iplookup router, which I don't think anybody now
885 uses, but I've fixed it anyway.
887 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
888 well as to direct DNS lookups. Otherwise the handling of names in host
889 lists is inconsistent and therefore confusing.
895 PH/01 Two changes to the default runtime configuration:
897 (1) Move the checks for relay_from_hosts and authenticated clients from
898 after to before the (commented out) DNS black list checks.
900 (2) Add control=submission to the relay_from_hosts and authenticated
901 clients checks, on the grounds that messages accepted by these
902 statements are most likely to be submissions.
904 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
906 (1) Generate an error if the third argument for the ${prvs expansion is
909 (2) Treat a missing third argument of ${prvscheck as if it were an empty
912 (3) Reset the variables that are obtained from the first argument of
913 ${prvscheck and used in the second argument before leaving the code,
914 because their memory is reclaimed, so using them afterwards may do
917 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
918 one (it's much easier than Tom thought :-).
920 (5) Because of (4), we can now allow for the use of $prvscheck_result
921 inside the third argument.
923 PH/03 For some reason, the default setting of PATH when running a command from
924 a pipe transport was just "/usr/bin". I have changed it to
927 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
928 anything to be listed in the output from -bV.
930 PH/05 When a filter generated an autoreply, the entire To: header line was
931 quoted in the delivery log line, like this:
933 => >A.N.Other <ano@some.domain> <original@ddress> ...
935 This has been changed so that it extracts the operative address. There
936 may be more than one such address. If so, they are comma-separated, like
939 => >ano@some.domain,ona@other.domain <original@ddress> ...
941 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
942 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
943 not being looked up in the rDNS to get a host name, Exim was showing the
944 IP address twice in Received: lines, even though the IP addresses were
945 identical. For example:
947 Received: from [1.2.3.4] (helo=[1.2.3.4])
949 However, if the real host name was known, it was omitting the HELO data
950 if it matched the actual IP address. This has been tidied up so that it
951 doesn't show the same IP address twice.
953 PH/07 When both +timestamp and +memory debugging was on, the value given by
954 $tod_xxx expansions could be wrong, because the tod_stamp() function was
955 called by the debug printing, thereby overwriting the timestamp buffer.
956 Debugging no longer uses the tod_stamp() function when +timestamp is set.
958 PH/08 When the original message was included in an autoreply transport, it
959 always said "this is a copy of the message, including all the headers",
960 even if body_only or headers_only was set. It now gives an appropriate
963 PH/09 Applied a patch from the Sieve maintainer which:
965 o fixes some comments
966 o adds the (disabled) notify extension core
967 o adds some debug output for the result of if/elsif tests
968 o points to the current vacation draft in the documentation
969 and documents the missing references header update
973 o fixes a bug in processing the envelope test (when testing
974 multiple envelope elements, the last element determinted the
977 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
978 Electronic Mail") by including:
980 Auto-submitted: auto-generated
982 in the messages that it generates (bounce messages and others, such as
983 warnings). In the case of bounce messages for non-SMTP mesages, there was
984 also a typo: it was using "Auto_submitted" (underscore instead of
985 hyphen). Since every message generated by Exim is necessarily in response
986 to another message, thes have all been changed to:
988 Auto-Submitted: auto-replied
990 in accordance with these statements in the RFC:
992 The auto-replied keyword:
994 - SHOULD be used on messages sent in direct response to another
995 message by an automatic process,
997 - MUST NOT be used on manually-generated messages,
999 - MAY be used on Delivery Status Notifications (DSNs) and Message
1000 Disposition Notifications (MDNs),
1002 - MUST NOT be used on messages generated by automatic or periodic
1003 processes, except for messages which are automatic responses to
1006 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1007 to the default Received: header definition.
1009 PH/12 Added log selector acl_warn_skipped (default on).
1011 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1012 variables because (a) they are in the wrong storage pool and (b) even if
1013 they were copied, it wouldn't work properly because of the caching.
1015 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1016 checking when decoding. Apparently there are clients that generate
1017 overlong encoded strings. Why am I not surprised?
1019 PH/15 If the first argument of "${if match_address" was not empty, but did not
1020 contain an "@" character, Exim crashed. Now it writes a panic log message
1021 and treats the condition as false.
1023 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1025 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1026 extension "envelope-auth". The code is finished and in agreement with
1027 other implementations, but there is no documentation so far and in fact,
1028 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1029 not changing the active code.
1031 Print executed "if" and "elsif" statements when debugging is used. This
1032 helps a great deal to understand what a filter does.
1034 Document more things not specified clearly in RFC3028. I had all this
1035 sorted out, when out of a sudden new issues came to my mind. Oops."
1037 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1040 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1041 canonical form (as documented). However, after a host name lookup from
1042 the IP address, check_host() was doing a simple string comparison with
1043 addresses acquired from the DNS when checking that the found name did
1044 have the original IP as one of its addresses. Since any found IPv6
1045 addresses are likely to be in abbreviated form, the comparison could
1046 fail. Luckily, there already exists a function for doing the comparison
1047 by converting both addresses to binary, so now that is used instead of
1048 the text comparison.
1050 PH/20 There was another similar case to PH/19, when a complete host name was
1051 given in a host list; looking up its IP address could give an abbreviated
1052 form, whereas the current host's name might or might not be abbreviated.
1053 The same fix has been applied.
1059 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1060 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1063 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1064 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1066 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1068 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1069 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1070 Also, for "Auto-Submitted", treat anything other than "no" as
1071 non-personal, in accordance with RFC 3834. (Previously it treated
1072 anything starting "auto-" as non-personal.)
1074 TF/01 The control=submission/name=... option had a problem with syntax
1075 errors if the name included a slash character. The /name= option
1076 now slurps the rest of the string, so it can include any characters
1077 but it must come last in the list of options (after /sender_retain
1080 PH/05 Some modifications to the interface to the fake nameserver for the new
1088 TK/01 Added the "success_on_redirect" address verification option. See
1089 NewStuff for rationale and an example.
1091 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1093 PH/02 Patch to exigrep to allow it to work on syslog lines.
1095 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1096 fread() to skip over the body file's header line, because in Cygwin the
1097 header line is locked and is inaccessible.
1099 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1100 co-exist for some time) to make it clear that it is the Exim ID that is
1101 referenced, not the Message-ID: header line.
1103 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1104 string_format() function, because snprintf() does not exist on all
1107 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1108 ${stat: expansion item.
1110 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1111 protocol synchronization error", to keep the pedants happy.
1113 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1114 well as for IRIX systems, when gcc is being used. See the host.c source
1117 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1119 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1122 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1123 transport and to the smtp transport in LMTP mode.
1125 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1127 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1128 run a filter in a subprocess. This could lead to confusion in subsequent
1129 lookups in the parent process. There should also be a search_tidyup() at
1130 the end of the subprocess.
1132 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1133 only if the host matched helo_try_verify_hosts, which caused the
1134 verification to occur when the EHLO/HELO command was issued. The ACL just
1135 tested the remembered result. Now, if a previous verification attempt has
1136 not happened, "verify = helo" does it there and then.
1138 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1140 TK/03 Fix log output including CR from clamd.
1142 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1143 exist provoked a memory error which could cause a segfault.
1145 PH/15 Installed PCRE 6.2
1147 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1149 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1150 of the problem. Specifically, suggested +O2 rather than +O1 for the
1153 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1155 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1156 record, the port was not correctly logged when the outgoing_port log
1157 selector was set (it logged the transort's default port).
1159 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1160 fallback_hosts, and "hosts" in the smtp transport.
1162 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1163 host errors such as "Connection refused".
1165 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1166 authentication with radiusclient 0.4.9:
1168 - Error returned from rc_read_config was caught wrongly
1169 - Username/password not passed on to radius server due to wrong length.
1171 The presumption is that some radiusclient API changes for 4.51/PH/17
1172 were not taken care of correctly. The code is still untested by me (my
1173 Linux distribution still has 0.3.2 of radiusclient), but it was
1174 contributed by a Radius user.
1176 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1177 expanding the "port" option of the smtp transport.
1179 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1180 while reading a MIME header. Thanks to Tom Hughes for a patch.
1182 PH/24 Include config.h inside local_scan.h so that configuration settings are
1185 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1186 that in an ACL for RCPT (for example), you can examine exactly what was
1189 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1190 commands, but it was not correctly comparing the address with the actual
1191 client host address. Thus, it would show the EHLO address in Received:
1192 header lines when this was not necessary.
1194 PH/27 Added the % operator to ${eval:}.
1196 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1197 it should be ignoring failures (because with -C, for example, it has lost
1198 privilege). It wasn't ignoring creation failures other than "already
1201 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1204 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1205 because an input line was too long, either on its own, or by virtue of
1206 too many continuations, the temporary file was not being removed, and the
1207 return code was incorrect.
1209 PH/31 Missing "BOOL" in function definition in filtertest.c.
1211 PH/32 Applied Sieve patches from the maintainer.
1213 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1215 PH/33 Added "verify = not_blind".
1217 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1218 Local/Makefile (with some defaults set). These are used in built scripts
1219 such as exicyclog, but they have never been used in the exim_install
1220 script (though there are many overriding facilities there). I have
1221 arranged that the exim_install script now takes note of these two
1224 PH/35 Installed configuration files for Dragonfly.
1226 PH/36 When a locally submitted message by a trusted user did not contain a
1227 From: header, and the sender address was obtained from -f or from an SMTP
1228 MAIL command, and the trusted user did not use -F to supply a sender
1229 name, $originator_name was incorrectly used when constructing a From:
1230 header. Furthermore, $originator_name was used for submission mode
1231 messages from external hosts without From: headers in a similar way,
1232 which is clearly wrong.
1234 PH/37 Added control=suppress_local_fixups.
1236 PH/38 When log_selector = +received_sender was set, and the addition of the
1237 sender made the log line's construction buffer exactly full, or one byte
1238 less than full, an overflow happened when the terminating "\n" was
1241 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1242 when the result of a list match is failure because a DNS lookup failed.
1244 PH/40 RM_COMMAND is now used in the building process.
1246 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1247 the "build-* directories that it finds.
1249 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1250 address in a domain literal was a prefix of an interface address.
1252 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1253 when verifying a sender address, unless rewrite_headers is false.
1255 PH/44 Wrote a long comment about why errors_to addresses are verified as
1256 recipients, not senders.
1258 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1259 the ratelimit ACL was added.
1261 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1263 PH/46 Added extra information about PostgreSQL errors to the error string.
1265 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1266 avoiding the need to install special zones in a real server. This is
1267 backwards compatible; if it can't find the fake resolver, it drops back.
1268 Thus, both old and new test suites can be run.
1270 TF/02 Added util/ratelimit.pl
1272 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1273 clock is set back in time.
1275 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1276 Candler <B.Candler@pobox.com>.
1278 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1279 OK for addresses that are the result of redirections.
1281 PH/48 A number of further additions for the benefit of the new test suite,
1282 including a fake gethostbyname() that interfaces to the fake DNS resolver
1285 TF/06 The fix for widen_domains has also been applied to qualify_single and
1286 search_parents which are the other dnslookup options that can cause
1289 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1292 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1294 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1295 service" if it encountered an MX record with an empty target hostname.
1296 The message is now "an MX or SRV record indicated no SMTP service".
1298 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1299 if the DNS of the sending site is misconfigured. This is quite a
1300 common situation. This change restores the behaviour of treating a
1301 helo verification defer as a failure.
1303 PH/51 If self=fail was set on a router, the bounce message did not include the
1304 actual error message.
1310 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1312 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1313 command itself ended in error, the underlying message about the transport
1314 filter timeout was being overwritten with the pipe command error. Now the
1315 underlying error message should be appended to the second error message.
1317 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1319 PH/02 Reduce the amount of output that "make" produces by default. Full output
1320 can still be requested.
1322 PH/03 The warning log line about a condition test deferring for a "warn" verb
1323 was being output only once per connection, rather than after each
1324 occurrence (because it was using the same function as for successful
1325 "warn" verbs). This seems wrong, so I have changed it.
1327 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1328 it should not have, which might have caused a crash in the right
1329 circumstances, but probably never did.
1331 PH/04 Installed a modified version of Tony Finch's patch to make submission
1332 mode fix the return path as well as the Sender: header line, and to
1333 add a /name= option so that you can make the user's friendly name appear
1336 TF/03 Added the control = fakedefer ACL modifier.
1338 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1339 Mark Lowes for thorough testing.
1341 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1343 TK/03 Merged latest SRS patch from Miles Wilton.
1345 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1346 with the definition in sysexits.h (which is #included earlier).
1347 Fortunately, Exim does not actually use EX_OK. The code used to try to
1348 preserve the sysexits.h value, by assumimg that macro definitions were
1349 scanned for macro replacements. I have been disabused of this notion,
1350 so now the code just undefines EX_OK before #including unistd.h.
1352 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1353 in the smtp transport. When a block could not be written in a single
1354 write() function, the timeout was being re-applied to each part-write.
1355 This seems wrong - if the receiver was accepting one byte at a time it
1356 would take for ever. The timeout is now adjusted when this happens. It
1357 doesn't have to be particularly precise.
1359 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1360 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1362 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1363 without a domain if the check to <postmaster@domain> fails.
1365 SC/01 Eximstats: added -xls and the ability to specify output files
1366 (patch written by Frank Heydlauf).
1368 SC/02 Eximstats: use FileHandles for outputing results.
1370 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1372 SC/04 Eximstats: fixed display of large numbers with -nvr option
1374 SC/05 Eximstats: fixed merging of reports with empty tables.
1376 SC/06 Eximstats: added the -include_original_destination flag
1378 SC/07 Eximstats: removed tabs and trailing whitespace.
1380 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1382 TK/06 MBOX spool code: Add real "From " MBOX separator line
1383 so the .eml file is really in mbox format (even though
1384 most programs do not really care). Patch from Alex Miller.
1386 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1387 The latter is generated from $received_to and is only set if the
1388 message has one envelope recipient. SA can use these headers,
1389 obviously out-of-the-box. Patch from Alex Miller.
1391 PH/08 The ${def test on a variable was returning false if the variable's
1392 value was "0", contrary to what the specification has always said!
1393 The result should be true unless the variable is empty.
1395 PH/09 The syntax error of a character other than { following "${if
1396 def:variable_name" (after optional whitespace) was not being diagnosed.
1397 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1398 accidental colon was present, for example, could give incorrect results.
1400 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1401 result is used (not including appendfile, where other changes are about
1404 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1405 This involved changing a lot of size variables from int to off_t. It
1406 should work with maildirs and everything.
1408 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1409 spamd dying while we are connected to it.
1411 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1414 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1415 function for BDB 4.3.
1417 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1419 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1420 the sources for PCRE have been re-arranged and more files are now
1423 PH/15 The code I had for printing potentially long long variables in PH/11
1424 above was not the best (it lost precision). The length of off_t variables
1425 is now inspected at build time, and an appropriate printing format (%ld
1426 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1427 to be "long long int" or "long int". This is needed for the internal
1428 formatting function string_vformat().
1430 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1431 the configuration file to be ":syslog", then the script "guesses" where
1432 the logs files are, rather than using the compiled in default. In our
1433 case the guess is not the same as the compiled default, so the script
1434 suddenly stopped working when I started to use syslog. The patch checks
1435 to see if log_file_path is "". If so, it attempts to read it from exim
1436 with no configuration file to get the compiled in version, before it
1437 falls back to the previous guessing code."
1439 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1440 implementing BATV in an Exim configuration. See NewStuff for the gory
1443 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1444 Makefile that are specific to HP-UX.
1446 PH/18 If the "use_postmaster" option was set for a recipient callout together
1447 with the "random" option, the postmaster address was used as the MAIL
1448 FROM address for the random test, but not for the subsequent recipient
1449 test. It is now used for both.
1451 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1452 patch removes a few documentation additions to RFC 3028, because the
1453 latest draft now contains them. It adds the new en;ascii-case comparator
1454 and a new error check for 8bit text in MIME parts. Comparator and
1455 require names are now matched exactly. I enabled the subaddress
1456 extension, but it is not well tested yet (read: it works for me)."
1458 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1459 rework some of the code of TK/09 above to avoid the hardwired use of
1460 "%lld" and "long long". Replaced the call to snprintf() with a call to
1463 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1464 records point to non-existent hosts", "retry timeout exceeded", and
1465 "retry time not reached for any host after a long failure period".
1467 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1468 experimental DomainKeys support:
1470 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1471 (2) On an error such as an illegally used "control", the wrong name for
1472 the control was given.
1474 These problems did NOT occur unless DomainKeys support was compiled.
1476 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1478 PH/24 Added ${if match_ip condition.
1480 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1481 databases so that it will be absolutely obvious if a crash occurs in the
1482 DB library. This is a regular occurrence (often caused by mis-matched
1485 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1486 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1487 picked up on a user's system that detects such things. There doesn't seem
1488 to be a gcc warning option for this - only an attribute that has to be
1489 put on the function's prototype. It seems that in Fedora Core 4 they have
1490 set this on a number of new functions. No doubt there will be more in due
1493 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1494 specify a transport. However, if an address that was verified by such a
1495 router was the subject of a callout, Exim crashed because it tried to
1496 read the rcpt_include_affixes from the non-existent transport. Now it
1497 just assumes that the setting of that option is false. This bug was
1498 introduced by 4.51/PH/31.
1500 PH/28 Changed -d+all to exclude +memory, because that information is very
1501 rarely of interest, but it makes the output a lot bigger. People tend to
1502 do -d+all out of habit.
1504 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1505 code in os-type was giving problems when libc.so lives in lib64, like on
1508 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1509 aren't the modern standard, and it seems that some systems' include files
1510 don't always have them. Exim was already checking for some of the newer
1511 ones like T_AAAA, and defining it itself. I've added checks for all the
1512 record types that Exim uses.
1514 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1515 not automatically generating a new one, as it is supposed to. This
1516 prevented TLS from working. If the file did exist, but contained invalid
1517 data, a new version was generated, as expected. It was only the case of a
1518 non-existent file that was broken.
1520 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1521 with a change in libdomainkeys > 0.64.
1523 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1524 from DNS. If the selector record carries the flag, it now has
1525 precedence over the domain-wide flag.
1527 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1529 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1530 the use of a port name that isn't defined in /etc/services) occurred, the
1531 message was deferred as in a normal delivery, and thus remained on the
1532 spool, instead of being failed because of the mua_wrapper setting. This
1533 is now fixed, and I tidied up some of the mua_wrapper messages at the
1536 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1537 the messages in a hash of arrays rather than using individual hashes.
1538 This is a bit cleaner and results in dramatic memory savings, albeit
1539 at a slight CPU cost.
1541 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1542 as requested by Marc Sherman.
1544 SC/10 Eximstats: added histograms for user specified patterns as requested
1547 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1549 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1550 fopen() in the content-scanning modules that did not already have it.
1556 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1557 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1559 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1561 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1563 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1564 to test Sieve filters that use "vacation".
1566 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1567 that changes the way the GnuTLS parameters are stored in the cache file.
1568 The new format can be generated externally. For backward compatibility,
1569 if the data in the cache doesn't make sense, Exim assumes it has read an
1570 old-format file, and it generates new data and writes a new file. This
1571 means that you can't go back to an older release without removing the
1574 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1575 work if there are any delivery delays because "one_time" forces the
1576 parent to be marked "delivered", so its unseen clone is never tried
1577 again. For this reason, Exim now forbids the simultaneous setting of
1580 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1581 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1582 entry if you want to know the details. The fix, however, neglected to
1583 consider the case when local delivery batching is involved. The test for
1584 "previously delivered" was not happening when checking to see if an
1585 address could be batched with a previous (undelivered) one; under
1586 certain circumstances this could lead to multiple deliveries to the same
1589 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1590 in its include files, and this causes problems building Exim.
1592 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1593 header_syntax) but Exim was just ignoring anything given after a slash.
1594 In particular, this caused confusion with an attempt to use "verify =
1595 reverse_host_lookup/defer_ok". An error is now given when options are
1596 supplied for verify items that do not have them. (Maybe reverse_host_
1597 lookup should have a defer_ok option, but that's a different point.)
1599 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1600 defined by RFC 821) to 2048, because there were problems with some AUTH
1601 commands, and RFC 1869 says the size should be increased for extended
1602 SMTP commands that take arguments.
1604 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1607 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1608 "unknown" error; now it says that the functionality isn't in the binary.
1610 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1611 an address' error message when a string expansion fails (syntax or
1612 whatever). Otherwise the password may appear in the log. Following change
1613 PH/42 below, there is no longer a chance of it appearing in a bounce
1616 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1618 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1619 of its addresses was ever tried. (Bugzilla bug #2.)
1621 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1622 the result incorrectly in the debug output. (It correctly added a newline
1623 to what was transported.)
1625 TF/01 Added $received_time.
1627 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1628 commented out examples of how to interface to a virus scanner and to
1629 SpamAssassin. Also added commented examples of av_scanner and
1630 spamd_address settings.
1632 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1633 and controls are allowed in which ACLs. There were a couple of minor
1634 errors. Some of the entries in the conditions table (which is a table of
1635 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1636 negation of where the condition IS allowed.
1638 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1640 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1641 header file does not have a version number, so I've had to invent a new
1642 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1643 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1644 radiusclient), but it was contributed by a Radius user.
1646 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1647 files or directories, for OpenSSL.
1649 PH/19 When an Exim process that is running as root has to create an Exim log
1650 file, it does so in a subprocess that runs as exim:exim so as to get the
1651 ownership right at creation (otherwise, other Exim processes might see
1652 the file with the wrong ownership). There was no test for failure of this
1653 fork() call, which would lead to the process getting stuck as it waited
1654 for a non-existent subprocess. Forks do occasionally fail when resources
1655 run out. I reviewed all the other calls to fork(); they all seem to check
1658 PH/20 When checking for unexpected SMTP input at connect time (before writing
1659 the banner), Exim was not dealing correctly with a non-positive return
1660 from the read() function. If the client had disconnected by this time,
1661 the result was a log entry for a synchronization error with an empty
1662 string after "input=" when read() returned zero. If read() returned -1
1663 (an event I could not check), uninitialized data bytes were printed.
1664 There were reports of junk text (parts of files, etc) appearing after
1667 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1669 PH/22 Added support for macro redefinition, and (re)definition in between
1670 driver and ACL definitions.
1672 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1673 forgetting to use the resulting value; it was using the unexpanded value.
1675 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1676 hadn't been configured. The fix is from Juergen Kreileder, who
1677 understands it better than I do:
1679 "Here's what I see happening with three configured cyrus_sasl
1680 authenticators configured (plain, login, cram-md5):
1682 On startup auth_cyrus_sasl_init() gets called for each of these.
1683 This means three calls to sasl_listmech() without a specified mech_list.
1684 => SASL tests which mechs of all available mechs actually work
1685 => three warnings about OTP not working
1686 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1688 With the patch, sasl_listmech() also gets called three times. But now
1689 SASL's mech_list option is set to the server_mech specified in the the
1690 authenticator. Or in other words, the answer from sasl_listmech()
1691 gets limited to just the mech you're testing for (which is different
1693 => the return list contains just 'plain' or 'login', 'cram-md5' or
1694 nothing depending on the value of ob->server_mech.
1696 I've just tested the patch: Authentication still works fine,
1697 unavailable mechs specified in the exim configuration are still
1698 caught, and the auth.log warnings about OTP are gone."
1700 PH/25 When debugging is enabled, the contents of the command line are added
1701 to the debugging output, even when log_selector=+arguments is not
1704 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1705 answer is "GNU", and only if the return is "GNU/something" is the answer
1708 PH/27 $acl_verify_message is now set immediately after the failure of a
1709 verification in an ACL, and so is available in subsequent modifiers. In
1710 particular, the message can be preserved by coding like this:
1712 warn !verify = sender
1713 set acl_m0 = $acl_verify_message
1715 Previously, $acl_verify_message was set only while expanding "message"
1716 and "log_message" when a very denied access.
1718 PH/28 Modified OS/os.c-Linux with
1720 -#ifndef OS_LOAD_AVERAGE
1721 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1723 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1724 nomenclature these days.)
1726 PH/29 Installed patch from the Sieve maintainer that adds the options
1727 sieve_useraddress and sieve_subaddress to the redirect router.
1729 PH/30 In these circumstances:
1730 . Two addresses routed to the same list of hosts;
1731 . First host does not offer TLS;
1732 . First host accepts first address;
1733 . First host gives temporary error to second address;
1734 . Second host offers TLS and a TLS session is established;
1735 . Second host accepts second address.
1736 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1737 and peerdn, if requested) that were in fact used only for the second
1740 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1741 attention to any local part prefix or suffix that was matched by the
1742 router that accepted the address. It now behaves in the same way as it
1743 does for delivery: the affixes are removed from the local part unless
1744 rcpt_include_affixes is set on the transport.
1746 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1747 timeout during the DATA phase of an incoming message.
1749 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1750 applied a patch sanctioned by the Sieve maintainer.
1752 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1753 the uid or gid is negative. A case of a negative gid caused this to be
1754 noticed. The fix allows for either to be negative.
1756 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1757 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1760 PH/36 The change PH/12 above was broken. Fixed it.
1762 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1763 the grounds that routing the same address twice would always produce the
1764 same answer. This might have been true once, but it is certainly no
1765 longer true now. Routing a child address may depend on the previous
1766 routing that produced that child. Some complicated redirection strategies
1767 went wrong when messages had multiple recipients, and made Exim's
1768 behaviour dependent on the order in which the addresses were given.
1770 I have moved the duplicate checking until after the routing is complete.
1771 Exim scans the addresses that are assigned to local and remote
1772 transports, and removes any duplicates. This means that more work will be
1773 done, as duplicates will always all be routed, but duplicates are
1774 presumably rare, so I don't expect this is of any significance.
1776 For deliveries to pipes, files, and autoreplies, the duplicate checking
1777 still happens during the routing process, since they are not going to be
1780 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1781 It corrects a timeout issue with spamd. This is Ian's comment: "The
1782 background is that sometimes spamd either never reads data from a
1783 connection it has accepted, or it never writes response data. The exiscan
1784 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1785 blindly assumes that writes won't block so it may never time out."
1787 PH/39 Allow G after quota size as well as K and M.
1789 PH/40 The value set for $authenticated_id in an authenticator may not contain
1790 binary zeroes or newlines because the value is written to log lines and
1791 to spool files. There was no check on this. Now the value is run through
1792 the string_printing() function so that such characters are converted to
1793 printable escape sequences.
1795 PH/41 $message_linecount is a new variable that contains the total number of
1796 lines in the message. Compare $body_linecount, which is the count for the
1799 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1800 bounce and delay warning messages, except in certain special cases, which
1803 (a) An SMTP error message from a remote host;
1804 (b) A message specified in a :fail: redirection;
1805 (c) A message specified in a "fail" command in a system filter;
1806 (d) A message specified in a FAIL return from the queryprogram router;
1807 (e) A message specified by the cannot_route_message router option.
1809 In these cases only, Exim does include the error details in bounce and
1810 warning messages. There are also a few cases where bland messages such
1811 as "unrouteable address" or "local delivery error" are given.
1813 PH/43 $value is now also set for the "else" part of a ${run expansion.
1815 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1816 being worked on, but at least Exim now implements the latest version to
1819 PH/45 In a pipe transport, although a timeout while waiting for the pipe
1820 process to complete was treated as a delivery failure, a timeout while
1821 writing the message to the pipe was logged, but erroneously treated as a
1822 successful delivery. Such timeouts include transport filter timeouts. For
1823 consistency with the overall process timeout, these timeouts are now
1824 treated as errors, giving rise to delivery failures by default. However,
1825 there is now a new Boolean option for the pipe transport called
1826 timeout_defer, which, if set TRUE, converts the failures into defers for
1827 both kinds of timeout. A transport filter timeout is now identified in
1830 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1831 systems where "make" and "gmake" are different, calling "gmake" at top
1832 level broke things. I've arranged for the value of $(MAKE) to be passed
1833 from the Makefile to this script so that it can call the same version of
1837 A note about Exim versions 4.44 and 4.50
1838 ----------------------------------------
1840 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1841 changes of various kinds. As a consequence, a big documentation update was
1842 needed. This delayed the release for rather longer than seemed good, especially
1843 in the light of a couple of (minor) security issues. Therefore, the changes
1844 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1845 release. So 4.44 and 4.50 are in effect two different branches that both start
1848 I have left the 4.50 change log unchanged; it contains all the changes since
1849 4.43. The change log for 4.44 is below; many of its items are identical to
1850 those for 4.50. This seems to be the most sensible way to preserve the
1851 historical information.
1857 1. Minor wording change to the doc/README.SIEVE file.
1859 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1860 computation of the current number of files was incorrect.
1862 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1863 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1866 4. Give more explanation in the error message when the command for a transport
1867 filter fails to execute.
1869 5. There are several places where Exim runs a non-Exim command in a
1870 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1871 was being done only for the command run by the queryprogram router. It is
1872 now done for all such subprocesses. The other cases are: ${run, transport
1873 filters, and the commands run by the lmtp and pipe transports.
1875 6. Added CONFIGURE_GROUP build-time option.
1877 7. Some older OS have a limit of 256 on the maximum number of file
1878 descriptors. Exim was using setrlimit() to set 1000 as a large value
1879 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1880 systems. I've change it so that if it can't get 1000, it tries for 256.
1882 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1883 was an oversight, and furthermore, ever since the addition of extra
1884 controls (e.g. 4.43/32), the checks on when to allow different forms of
1885 "control" were broken. There should now be diagnostics for all cases when a
1886 control that does not make sense is encountered.
1888 9. Added the /retain_sender option to "control=submission".
1890 10. $recipients is now available in the predata ACL (oversight).
1892 11. Tidy the search cache before the fork to do a delivery from a message
1893 received from the command line. Otherwise the child will trigger a lookup
1894 failure and thereby defer the delivery if it tries to use (for example) a
1895 cached ldap connection that the parent has called unbind on.
1897 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1898 of $address_data from the recipient verification was clobbered by the
1899 sender verification.
1901 13. The value of address_data from a sender verification is now available in
1902 $sender_address_data in subsequent conditions in the ACL statement.
1904 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1906 15. Added a new option "connect=<time>" to callout options, to set a different
1909 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1910 was its contents. (It was OK if the option was not defined at all.)
1912 17. A "Completed" log line is now written for messages that are removed from
1913 the spool by the -Mrm option.
1915 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1916 information about exactly what failed.
1918 19. Added -dd to debug only the daemon process.
1920 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1921 handles timeouts, both on the server side and network timeouts. Renamed the
1922 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1924 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1925 It is now set to "smtps".
1927 22. $host_address is now set to the target address during the checking of
1928 ignore_target_hosts.
1930 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1931 being passed; this would have caused $sender_host_name to have been used if
1932 matching the list had actually called for a host name (not very likely,
1933 since this list is usually IP addresses). A host name is now passed as
1936 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1937 code that specifies a non-restarting handler (typically sigaction() in
1938 modern systems) in an attempt to fix a rare and obscure crash bug.
1940 25. Narrowed the window for a race in the daemon that could cause it to ignore
1941 SIGCHLD signals. This is not a major problem, because they are used only to
1942 wake it up if nothing else does.
1944 26. A malformed maildirsize file could cause Exim to calculate negative values
1945 for the mailbox size or file count. Odd effects could occur as a result.
1946 The maildirsize information is now recalculated if the size or filecount
1949 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1950 support for a long time. Removed HAVE_SYS_VFS_H.
1952 28. Installed the latest version of exipick from John Jetmore.
1954 29. In an address list, if the pattern was not a regular expression, an empty
1955 subject address (from a bounce message) matched only if the pattern was an
1956 empty string. Non-empty patterns were not even tested. This was the wrong
1957 because it is perfectly reasonable to use an empty address as part of a
1958 database query. An empty address is now tested by patterns that are
1959 lookups. However, all the other forms of pattern expect the subject to
1960 contain a local part and a domain, and therefore, for them, an empty
1961 address still always fails if the pattern is not itself empty.
1963 30. Exim went into a mad DNS loop when attempting to do a callout where the
1964 host was specified on an smtp transport, and looking it up yielded more
1965 than one IP address.
1967 31. Re-factored the code for checking spool and log partition space into a
1968 function that finds that data and another that does the check. The former
1969 is then used to implement four new variables: $spool_space, $log_space,
1970 $spool_inodes, and $log_inodes.
1972 32. The RFC2047 encoding function was originally intended for short strings
1973 such as real names; it was not keeping to the 75-character limit for
1974 encoded words that the RFC imposes. It now respects the limit, and
1975 generates multiple encoded words if necessary. To be on the safe side, I
1976 have increased the buffer size for the ${rfc2047: expansion operator from
1979 33. It is now permitted to omit both strings after an "if" condition; if the
1980 condition is true, the result is "true". As before, when the second string
1981 is omitted, a false condition yields an empty string. This makes it less
1982 cumbersome to write custom ACL and router conditions.
1984 34. Failure to deliver a bounce message always caused it to be frozen, even if
1985 there was an errors_to setting on the router. The errors_to setting is now
1988 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1989 canonical form (fully expanded) before being placed in
1990 $sender_host_address.
1992 36. The table in the code that translates DNS record types into text (T_A to
1993 "A" for instance) was missing entries for NS and CNAME. It is just possible
1994 that this could have caused confusion if both these types were looked up
1995 for the same domain, because the text type is used as part of Exim's
1996 per-process caching. But the chance of anyone hitting this buglet seems
1999 37. The dnsdb lookup has been extended in a number of ways.
2001 (1) There is a new type, "zns", which walks up the domain tree until it
2002 finds some nameserver records. It should be used with care.
2004 (2) There is a new type, "mxh", which is like "mx" except that it returns
2005 just the host names, not the priorities.
2007 (3) It is now possible to give a list of domains (or IP addresses) to be
2008 looked up. The behaviour when one of the lookups defers can be
2009 controlled by a keyword.
2011 (4) It is now possible to specify the separator character for use when
2012 multiple records are returned.
2014 38. The dnslists ACL condition has been extended: it is now possible to supply
2015 a list of IP addresses and/or domains to be looked up in a particular DNS
2018 39. Added log_selector=+queue_time_overall.
2020 40. When running the queue in the test harness, wait just a tad after forking a
2021 delivery process, to get repeatability of debugging output.
2023 41. Include certificate and key file names in error message when GnuTLS fails
2024 to set them up, because the GnuTLS error message doesn't include the name
2025 of the failing file when there is a problem reading it.
2027 42. Allow both -bf and -bF in the same test run.
2029 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2031 44. The "Exiscan patch" is now merged into the mainline Exim source.
2033 45. Sometimes the final signoff response after QUIT could fail to get
2034 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2035 < 0 before doing a fflush(). This bug looks as though it goes back to the
2036 introduction of TLS in release 3.20, but "sometimes" must have been rare
2037 because the tests only now provoked it.
2039 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2040 (this can affect the format of dates).
2042 47. exim_tidydb, when checking for the continued existence of a message for
2043 which it has found a message-specific retry record, was not finding
2044 messages that were in split spool directories. Consequently, it was
2045 deleting retry records that should have stayed in existence.
2047 48. Steve fixed some bugs in eximstats.
2049 49. The SPA authentication driver was not abandoning authentication and moving
2050 on to the next authenticator when an expansion was forced to fail,
2051 contradicting the general specification for all authenticators. Instead it
2052 was generating a temporary error. It now behaves as specified.
2054 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2055 (the order specifies the preference for clients). The order is now AES256,
2056 AES128, 3DES, ARCFOUR128.
2058 51. Small patch to Sieve code - explicitly set From: when generating an
2061 52. Exim crashed if a remote delivery caused a very long error message to be
2062 recorded - for instance if somebody sent an entire SpamAssassin report back
2063 as a large number of 550 error lines. This bug was coincidentally fixed by
2064 increasing the size of one of Exim's internal buffers (big_buffer) that
2065 happened as part of the Exiscan merge. However, to be on the safe side, I
2066 have made the code more robust (and fixed the comments that describe what
2069 53. Now that there can be additional text after "Completed" in log lines (if
2070 the queue_time_overall log selector is set), a one-byte patch to exigrep
2071 was needed to allow it to recognize "Completed" as not the last thing in
2074 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2075 patch that reportedly fixes this has been added. I am not expert enough to
2076 create a test for it. This is what the patch creator wrote:
2078 "I found a little strange behaviour of ldap code when working with
2079 Windows 2003 AD Domain, where users was placed in more than one
2080 Organization Units. When I tried to give exim partial DN, the exit code
2081 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2082 But simultaneously result of request was absolutely normal ldap result,
2083 so I produce this patch..."
2085 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2086 REFERENCE, so I have modified the code to exclude the patch when that macro
2089 55. Some experimental protocols are using DNS PTR records for new purposes. The
2090 keys for these records are domain names, not reversed IP addresses. The
2091 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2092 leaves it alone. Component reversal etc. now happens only for IP addresses.
2095 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2097 57. Double the size of the debug message buffer (to 2048) so that more of very
2098 long debug lines gets shown.
2100 58. The exicyclog utility now does better if the number of log files to keep
2101 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2103 59. Two changes related to the smtp_active_hostname option:
2105 (1) $smtp_active_hostname is now available as a variable.
2106 (2) The default for smtp_banner uses $smtp_active_hostname instead
2107 of $primary_hostname.
2109 60. The host_aton() function is supposed to be passed a string that is known
2110 to be a valid IP address. However, in the case of IPv6 addresses, it was
2111 not checking this. This is a hostage to fortune. Exim now panics and dies
2112 if the condition is not met. A case was found where this could be provoked
2113 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2114 components; fortuitously, this particular loophole had already been fixed
2115 by change 4.50/55 above.
2117 If there are any other similar loopholes, the new check in host_aton()
2118 itself should stop them being exploited. The report I received stated that
2119 data on the command line could provoke the exploit when Exim was running as
2120 exim, but did not say which command line option was involved. All I could
2121 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2122 running as the user.
2125 61. There was a buffer overflow vulnerability in the SPA authentication code
2126 (which came originally from the Samba project). I have added a test to the
2127 spa_base64_to_bits() function which I hope fixes it.
2130 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2131 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2133 63. The daemon start-up calls getloadavg() while still root for those OS that
2134 need the first call to be done as root, but it missed one case: when
2135 deliver_queue_load_max is set with deliver_drop_privilege. This is
2136 necessary for the benefit of the queue runner, because there is no re-exec
2137 when deliver_drop_privilege is set.
2139 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2140 This has been fixed.
2142 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2143 was in use, was not putting the data itself into the right store pool;
2144 consequently, it could be overwritten for a subsequent message in the same
2145 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2148 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2150 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2151 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2152 the function were treating the return as a boolean value, which happened to
2153 work because 0=false and not-0=true, but is not correct code.
2155 68. The host_aton() function was not handling scoped IPv6 addresses (those
2156 with, for example, "%eth0" on the end) correctly.
2158 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2159 negated items (that is, ~something) in unsigned ints. Some compilers
2160 apparently mutter when there is no cast.
2162 70. If an address verification called from an ACL failed, and did not produce a
2163 user-specific message (i.e. there was only a "system" message), nothing was
2164 put in $acl_verify_message. In this situation, it now puts the system
2167 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2168 session; change 4.31/43 added the unwanted input to the log line - except
2169 that it did not do this in the start of session case. It now does.
2171 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2172 This is wrong and can cause the other end to generate a synchronization
2173 error if it is another Exim or anything else that does the synchronization
2174 check. A QUIT command is no longer sent after a timeout.
2176 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2177 during host lookups.
2179 74. The defer_ok option of callout verification was not working if it was used
2180 when verifying addresses in header lines, that is, for this case:
2182 verify = header_sender/callout=defer_ok
2184 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2185 those file descriptors could be used for SMTP connections. If anything
2186 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2187 could be sent to the SMTP client, causing chaos. The daemon now opens
2188 stdin, stdout, and stderr to /dev/null when it puts itself into the
2191 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2192 log by default. The user can override this with suitable Perl magic.
2194 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2195 the log message when discard triggers, was not working for the DATA ACL or
2196 for the non-SMTP ACL.
2198 78. Error message wording change in sieve.c.
2200 79. If smtp_accept_max_per_host was set, the number of connections could be
2201 restricted to fewer than expected, because the daemon was trying to set up
2202 a new connection before checking whether the processes handling previous
2203 connections had finished. The check for completed processes is now done
2204 earlier. On busy systems, this bug wouldn't be noticed because something
2205 else would have woken the daemon, and it would have reaped the completed
2208 80. If a message was submitted locally by a user whose login name contained one
2209 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2210 It caused a spool format error. I have fixed the spool reading code. A
2211 related problem was that the "from" clause in the Received: line became
2212 illegal because of the space(s). It is now covered by ${quote_local_part.
2214 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2217 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2219 83. Patch from Sieve maintainer to fix unterminated string problem in
2220 "vacation" handling.
2222 84. Some minor changes to the Linux configuration files to help with other
2223 OS variants using glibc.
2225 85. One more patch for Sieve to update vacation handling to latest spec.
2228 ----------------------------------------------------
2229 See the note above about the 4.44 and 4.50 releases.
2230 ----------------------------------------------------
2236 1. Change 4.43/35 introduced a bug that caused file counts to be
2237 incorrectly computed when quota_filecount was set in an appendfile
2240 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2241 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2244 3. Give more explanation in the error message when the command for a transport
2245 filter fails to execute.
2247 4. There are several places where Exim runs a non-Exim command in a
2248 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2249 was being done only for the command run by the queryprogram router. It is
2250 now done for all such subprocesses. The other cases are: ${run, transport
2251 filters, and the commands run by the lmtp and pipe transports.
2253 5. Some older OS have a limit of 256 on the maximum number of file
2254 descriptors. Exim was using setrlimit() to set 1000 as a large value
2255 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2256 systems. I've change it so that if it can't get 1000, it tries for 256.
2258 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2259 was an oversight, and furthermore, ever since the addition of extra
2260 controls (e.g. 4.43/32), the checks on when to allow different forms of
2261 "control" were broken. There should now be diagnostics for all cases when a
2262 control that does not make sense is encountered.
2264 7. $recipients is now available in the predata ACL (oversight).
2266 8. Tidy the search cache before the fork to do a delivery from a message
2267 received from the command line. Otherwise the child will trigger a lookup
2268 failure and thereby defer the delivery if it tries to use (for example) a
2269 cached ldap connection that the parent has called unbind on.
2271 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2272 of $address_data from the recipient verification was clobbered by the
2273 sender verification.
2275 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2276 was its contents. (It was OK if the option was not defined at all.)
2278 11. A "Completed" log line is now written for messages that are removed from
2279 the spool by the -Mrm option.
2281 12. $host_address is now set to the target address during the checking of
2282 ignore_target_hosts.
2284 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2285 being passed; this would have caused $sender_host_name to have been used if
2286 matching the list had actually called for a host name (not very likely,
2287 since this list is usually IP addresses). A host name is now passed as
2290 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2291 code that specifies a non-restarting handler (typically sigaction() in
2292 modern systems) in an attempt to fix a rare and obscure crash bug.
2294 15. Narrowed the window for a race in the daemon that could cause it to ignore
2295 SIGCHLD signals. This is not a major problem, because they are used only to
2296 wake it up if nothing else does.
2298 16. A malformed maildirsize file could cause Exim to calculate negative values
2299 for the mailbox size or file count. Odd effects could occur as a result.
2300 The maildirsize information is now recalculated if the size or filecount
2303 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2304 support for a long time. Removed HAVE_SYS_VFS_H.
2306 18. Updated exipick to current release from John Jetmore.
2308 19. Allow an empty sender to be matched against a lookup in an address list.
2309 Previously the only cases considered were a regular expression, or an
2312 20. Exim went into a mad DNS lookup loop when doing a callout where the
2313 host was specified on the transport, if the DNS lookup yielded more than
2316 21. The RFC2047 encoding function was originally intended for short strings
2317 such as real names; it was not keeping to the 75-character limit for
2318 encoded words that the RFC imposes. It now respects the limit, and
2319 generates multiple encoded words if necessary. To be on the safe side, I
2320 have increased the buffer size for the ${rfc2047: expansion operator from
2323 22. Failure to deliver a bounce message always caused it to be frozen, even if
2324 there was an errors_to setting on the router. The errors_to setting is now
2327 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2328 canonical form (fully expanded) before being placed in
2329 $sender_host_address.
2331 24. Updated eximstats to version 1.33
2333 25. Include certificate and key file names in error message when GnuTLS fails
2334 to set them up, because the GnuTLS error message doesn't include the name
2335 of the failing file when there is a problem reading it.
2337 26. Expand error message when OpenSSL has problems setting up cert/key files.
2340 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2341 (this can affect the format of dates).
2343 28. exim_tidydb, when checking for the continued existence of a message for
2344 which it has found a message-specific retry record, was not finding
2345 messages that were in split spool directories. Consequently, it was
2346 deleting retry records that should have stayed in existence.
2348 29. eximstats updated to version 1.35
2349 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2350 1.35 - bugfix such that pie charts by volume are generated correctly
2352 30. The SPA authentication driver was not abandoning authentication and moving
2353 on to the next authenticator when an expansion was forced to fail,
2354 contradicting the general specification for all authenticators. Instead it
2355 was generating a temporary error. It now behaves as specified.
2357 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2358 (the order specifies the preference for clients). The order is now AES256,
2359 AES128, 3DES, ARCFOUR128.
2361 31. Small patch to Sieve code - explicitly set From: when generating an
2364 32. Exim crashed if a remote delivery caused a very long error message to be
2365 recorded - for instance if somebody sent an entire SpamAssassin report back
2366 as a large number of 550 error lines. This bug was coincidentally fixed by
2367 increasing the size of one of Exim's internal buffers (big_buffer) that
2368 happened as part of the Exiscan merge. However, to be on the safe side, I
2369 have made the code more robust (and fixed the comments that describe what
2372 33. Some experimental protocols are using DNS PTR records for new purposes. The
2373 keys for these records are domain names, not reversed IP addresses. The
2374 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2375 leaves it alone. Component reversal etc. now happens only for IP addresses.
2378 34. The host_aton() function is supposed to be passed a string that is known
2379 to be a valid IP address. However, in the case of IPv6 addresses, it was
2380 not checking this. This is a hostage to fortune. Exim now panics and dies
2381 if the condition is not met. A case was found where this could be provoked
2382 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2383 components; fortuitously, this particular loophole had already been fixed
2384 by change 4.50/55 or 4.44/33 above.
2386 If there are any other similar loopholes, the new check in host_aton()
2387 itself should stop them being exploited. The report I received stated that
2388 data on the command line could provoke the exploit when Exim was running as
2389 exim, but did not say which command line option was involved. All I could
2390 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2391 running as the user.
2394 35. There was a buffer overflow vulnerability in the SPA authentication code
2395 (which came originally from the Samba project). I have added a test to the
2396 spa_base64_to_bits() function which I hope fixes it.
2399 36. The daemon start-up calls getloadavg() while still root for those OS that
2400 need the first call to be done as root, but it missed one case: when
2401 deliver_queue_load_max is set with deliver_drop_privilege. This is
2402 necessary for the benefit of the queue runner, because there is no re-exec
2403 when deliver_drop_privilege is set.
2405 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2406 was in use, was not putting the data itself into the right store pool;
2407 consequently, it could be overwritten for a subsequent message in the same
2408 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2411 38. Sometimes the final signoff response after QUIT could fail to get
2412 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2413 < 0 before doing a fflush(). This bug looks as though it goes back to the
2414 introduction of TLS in release 3.20, but "sometimes" must have been rare
2415 because the tests only now provoked it.
2421 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2422 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2423 Now it returns an integer. A number of calls were still expecting a T/F
2424 return. Fortuitously, in all cases, the tests worked in OK situations,
2425 which is the norm. However, things would have gone wrong on any write
2426 failures on the smtp file descriptor. This function is used when sending
2427 messages over SMTP and also when doing verify callouts.
2429 2. When Exim is called to do synchronous delivery of a locally submitted
2430 message (the -odf or -odi options), it no longer closes stderr before doing
2433 3. Implemented the mua_wrapper option.
2435 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2437 5. Implemented the functions header_remove(), header_testname(),
2438 header_add_at_position(), and receive_remove_recipient(), and exported them
2441 6. If an ACL "warn" statement specified the addition of headers, Exim already
2442 inserted X-ACL-Warn: at the start if there was no header name. However, it
2443 was not making this test for the second and subsequent header lines if
2444 there were newlines in the string. This meant that an invalid header could
2445 be inserted if Exim was badly configured.
2447 7. Allow an ACL "warn" statement to add header lines at the start or after all
2448 the Received: headers, as well as at the end.
2450 8. Added the rcpt_4xx retry error code.
2452 9. Added postmaster_mailfrom=xxx to callout verification option.
2454 10. Added mailfrom=xxxx to the callout verification option, for verify=
2457 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2458 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2460 12. Inserted some casts to stop certain compilers warning when using pointer
2461 differences as field lengths or precisions in printf-type calls (mostly
2462 affecting debugging statements).
2464 13. Added optional readline() support for -be (dynamically loaded).
2466 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2467 same clock tick as a message's arrival, so that its received time was the
2468 same as the "first fail" time on the retry record, and that message
2469 remained on the queue past the ultimate address timeout, every queue runner
2470 would try a delivery (because it was past the ultimate address timeout) but
2471 after another failure, the ultimate address timeout, which should have then
2472 bounced the address, did not kick in. This was a "< instead of <=" error;
2473 in most cases the first failure would have been in the next clock tick
2474 after the received time, and all would be well.
2476 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2477 being recognized when the domain list was tested by the match_domain
2478 condition in an expansion string.
2480 16. Added the ${str2b64: operator.
2482 17. Exim was always calling setrlimit() to set a large limit for the number of
2483 processes, without checking whether the existing limit was already
2484 adequate. (It did check for the limit on file descriptors.) Furthermore,
2485 errors from getrlimit() and setrlimit() were being ignored. Now they are
2486 logged to the main and panic logs, but Exim does carry on, to try to do its
2487 job under whatever limits there are.
2489 18. Imported PCRE 5.0.
2491 19. Trivial typo in log message " temporarily refused connection" (the leading
2494 20. If the log selector return_path_on_delivery was set and an address was
2495 redirected to /dev/null, the delivery process crashed because it assumed
2496 that a return path would always be set for a "successful" delivery. In this
2497 case, the whole delivery is bypassed as an optimization, and therefore no
2500 21. Internal re-arrangement: the function for sending a challenge and reading
2501 a response while authentication was assuming a zero-terminated challenge
2502 string. It's now changed to take a pointer and a length, to allow for
2503 binary data in such strings.
2505 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2507 23. Exim was not respecting finduser_retries when seeking the login of the
2508 uid under which it was called; it was always trying 10 times. (The default
2509 setting of finduser_retries is zero.) Also, it was sleeping after the final
2510 failure, which is pointless.
2512 24. Implemented tls_on_connect_ports.
2514 25. Implemented acl_smtp_predata.
2516 26. If the domain in control=submission is set empty, Exim assumes that the
2517 authenticated id is a complete email address when it generates From: or
2518 Sender: header lines.
2520 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2521 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2522 chown and chgrp in /bin and hostname in /usr/bin.
2524 28. Exim was keeping the "process log" file open after each use, just as it
2525 does for the main log. This opens the possibility of it remaining open for
2526 long periods when the USR1 signal hits a daemon. Occasional processlog
2527 errors were reported, that could have been caused by this. Anyway, it seems
2528 much more sensible not to leave this file open at all, so that is what now
2531 29. The long-running daemon process does not normally write to the log once it
2532 has entered its main loop, and it closes the log before doing so. This is
2533 so that log files can straightforwardly be renamed and moved. However,
2534 there are a couple of unusual error situations where the daemon does write
2535 log entries, and I had neglected to close the log afterwards.
2537 30. The text of an SMTP error response that was received during a remote
2538 delivery was being truncated at 512 bytes. This is too short for some of
2539 the long messages that one sometimes sees. I've increased the limit to
2542 31. It is now possible to make retry rules that apply only when a message has a
2543 specific sender, in particular, an empty sender.
2545 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2546 it possible to be selective about when SMTP synchronization is enforced.
2548 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2550 32. Implemented hosts_connection_nolog.
2552 33. Added an ACL for QUIT.
2554 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2557 35. Added mailbox_size and mailbox_filecount to appendfile.
2559 36. Added control = no_multiline_responses to ACLs.
2561 37. There was a bug in the logic of the code that waits for the clock to tick
2562 in the case where the clock went backwards by a substantial amount such
2563 that the microsecond fraction of "now" was more than the microsecond
2564 fraction of "then" (but the whole seconds number was less).
2566 38. Added support for the libradius Radius client library this is found on
2567 FreeBSD (previously only the radiusclient library was supported).
2573 1. When certain lookups returned multiple values in the form name=value, the
2574 quoting of the values was not always being done properly. Specifically:
2575 (a) If the value started with a double quote, but contained no whitespace,
2577 (b) If the value contained whitespace other than a space character (i.e.
2578 tabs or newlines or carriage returns) it was not quoted.
2579 This fix has been applied to the mysql and pgsql lookups by writing a
2580 separate quoting function and calling it from the lookup code. The fix
2581 should probably also be applied to nisplus, ibase and oracle lookups, but
2582 since I cannot test any of those, I have not disturbed their existing code.
2584 2. A hit in the callout cache for a specific address caused a log line with no
2585 reason for rejecting RCPT. Now it says "Previous (cached) callout
2586 verification failure".
2588 3. There was an off-by-one bug in the queryprogram router. An over-long
2589 return line was truncated at 256 instead of 255 characters, thereby
2590 overflowing its buffer with the terminating zero. As well as fixing this, I
2591 have increased the buffer size to 1024 (and made a note to document this).
2593 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2594 when Exim is waiting for an SMTP response from a remote server, Exim
2595 restarts its select() call on the socket, thereby resetting its timeout.
2596 This is not a problem when such interrupts are rare. Somebody set up a cron
2597 job to run exiwhat every 2 minutes, which is less than the normal select()
2598 timeout (5 or 10 minutes). This meant that the select() timeout never
2599 kicked in because it was always reset. I have fixed this by comparing the
2600 time when an interrupt arrives with the time at the start of the first call
2601 to select(). If more time than the timeout has elapsed, the interrupt is
2602 treated as a timeout.
2604 5. Some internal re-factoring in preparation for the addition of Sieve
2605 extensions (by MH). In particular, the "personal" test is moved to a
2606 separate function, and given an option for scanning Cc: and Bcc: (which is
2607 not set for Exim filters).
2609 6. When Exim created an email address using the login of the caller as the
2610 local part (e.g. when creating a From: or Sender: header line), it was not
2611 quoting the local part when it contained special characters such as @.
2613 7. Installed new OpenBSD configuration files.
2615 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2616 try to make them clearer.
2618 9. Callout options, other than the timeout value, were being ignored when
2619 verifying sender addresses in header lines. For example, when using
2621 verify = header_sender/callout=no_cache
2623 the cache was (incorrectly) being used.
2625 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2626 only the Cygwin environment.
2628 11. When return_path_on_delivery was set as a log selector, if different remote
2629 addresses in the same message used different return paths and parallel
2630 remote delivery occurred, the wrong values would sometimes be logged.
2631 (Whenever a remote delivery process finished, the return path value from
2632 the most recently started remote delivery process was logged.)
2634 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2635 lines when AUTH and/or TLS are in use. This is the "received protocol"
2636 field. Exim used to use "asmtp" for authenticated SMTP, without any
2637 indication (in the protocol name) for TLS use. Now it follows the RFC and
2638 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2639 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2640 names appear in log lines as well as in Received: header lines.
2642 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2643 extensions, and comparison tests, and to fix some bugs.
2645 14. Changes to the "personal" filter test:
2647 (1) The test was buggy in that it was just doing the equivalent of
2648 "contains" tests on header lines. For example, if a user's address was
2649 anne@some.where, the "personal" test would incorrectly be true for
2651 To: susanne@some.where
2653 This test is now done by extracting each address from the header in turn,
2654 and checking the entire address. Other tests that are part of "personal"
2655 are now done using regular expressions (for example, to check local parts
2656 of addresses in From: header lines).
2658 (2) The list of non-personal local parts in From: addresses has been
2659 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2660 taken from the Sieve specification recommendations.
2662 (3) If the message contains any header line starting with "List-" it is
2663 treated as non-personal.
2665 (4) The test for "circular" in the Subject: header line has been removed
2666 because it now seems ill-conceived.
2668 15. Minor typos in src/EDITME comments corrected.
2670 16. Installed latest exipick from John Jetmore.
2672 17. If headers_add on a router specified a text string that was too long for
2673 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2674 of string_sprintf() is now avoided.
2676 18. $message_body_size was not set (it was always zero) when running the DATA
2677 ACL and the local_scan() function.
2679 19. For the "mail" command in an Exim filter, no default was being set for
2680 the once_repeat time, causing a random time value to be used if "once" was
2681 specified. (If the value happened to be <= 0, no repeat happened.) The
2682 default is now 0s, meaning "never repeat". The "vacation" command was OK
2683 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2684 (I found it when inspecting the code).
2686 20. There is now an overall timeout for performing a callout verification. It
2687 defaults to 4 times the callout timeout, which applies to individual SMTP
2688 commands during the callout. The overall timeout applies when there is more
2689 than one host that can be tried. The timeout is checked before trying the
2690 next host. This prevents very long delays if there are a large number of
2691 hosts and all are timing out (e.g. when the network connections are timing
2692 out). The value of the overall timeout can be changed by specifying an
2693 additional sub-option for "callout", called "maxwait". For example:
2695 verify = sender/callout=5s,maxwait=20s
2697 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2698 to the end before writing, but this should make it even safer).
2700 22. Exim was forgetting that it had advertised PIPELINING for the second and
2701 subsequent messages on an SMTP connection. It was also not resetting its
2702 memory on STARTTLS and an internal HELO.
2704 23. When Exim logs an SMTP synchronization error within a session, it now
2705 records whether PIPELINING has been advertised or not.
2707 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2708 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2709 rather than long int.
2711 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2713 26. Added the never_mail option to autoreply.
2719 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2720 crash if the getsockname() call failed; this can happen if a connection is
2721 closed very soon after it is established. The problem was simply in the
2722 order in which certain operations were done, causing Exim to try to write
2723 to the SMTP stream before it had set up the file descriptor. The bug has
2724 been fixed by making things happen in the correct order.
2730 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2731 before the connection was closed, thus losing the rejection response.
2733 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2734 some early Solaris releases, but causes trouble in current releases where
2735 socklen_t is defined.
2737 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2740 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2743 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2744 that releasing the top of it at the end releases what was used for sub-
2745 expansions (unless the block got too big). However, discard this block if
2746 the first thing is a variable or header, so that we can use its block when
2747 it is dynamic (useful for very large $message_headers, for example).
2749 6. Lookups now cache *every* query, not just the most recent. A new, separate
2750 store pool is used for this. It can be recovered when all lookup caches are
2751 flushed. Lookups now release memory at the end of their result strings.
2752 This has involved some general refactoring of the lookup sources.
2754 7. Some code has been added to the store_xxx() functions to reduce the amount
2755 of flapping under certain conditions.
2757 8. log_incoming_interface used to affect only the <= reception log lines. Now
2758 it causes the local interface and port to be added to several more SMTP log
2759 lines, for example "SMTP connection from", and rejection lines.
2761 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2763 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2765 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2766 could be overwritten at the end of the current message (or the start of a
2767 new message if it was set in a HELO ACL). The value is now preserved for
2768 the duration of the SMTP connection.
2770 12. If a transport had a headers_rewrite setting, and a matching header line
2771 contained an unqualified address, that address was qualified, even if it
2772 did not match any rewriting rules. The underlying bug was that the values
2773 of the flags that permit the existence of unqualified sender and recipient
2774 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2775 non-local messages, and by -bnq for local messages) were not being
2776 preserved with the message after it was received.
2778 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2779 "next input=" as part of the text comprising the host identity instead of
2780 the correct text. The code was using the same buffer for two different
2781 strings. However, depending on which order the printing function evaluated
2782 its arguments, the bug did not always show up. Under Linux, for example, my
2783 test suite worked just fine.
2785 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2786 doesn't work with some older versions of Perl. It has been changed to "my",
2787 which in any case is probably the better facility to use.
2789 15. A really picky compiler found some instances of statements for creating
2790 error messages that either had too many or two few arguments for the format
2793 16. The size of the buffer for calls to the DNS resolver has been increased
2794 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2795 for addresses that have a lot of PTR records. This alleviates a problem; it
2796 does not fully solve it.
2798 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2799 buffer now truncates the list and logs the incident, which is the same
2800 action as happens when Exim is looking up a host name and its aliases.
2801 Previously in this situation something unpredictable would happen;
2802 sometimes it was "internal error: store_reset failed".
2804 18. If a server dropped the connection unexpectedly when an Exim client was
2805 using GnuTLS and trying to read a response, the client delivery process
2806 crashed while trying to generate an error log message.
2808 19. If a "warn" verb in an ACL added multiple headers to a message in a single
2809 string, for example:
2811 warn message = H1: something\nH2: something
2813 the text was added as a single header line from Exim's point of view
2814 though it ended up OK in the delivered message. However, searching for the
2815 second and subsequent header lines using $h_h2: did not work. This has been
2816 fixed. Similarly, if a system filter added multiple headers in this way,
2817 the routers could not see them.
2819 20. Expanded the error message when iplsearch is called with an invalid key to
2820 suggest using net-iplsearch in a host list.
2822 21. When running tests using -bh, any delays imposed by "delay" modifiers in
2823 ACLs are no longer actually imposed (and a message to that effect is
2826 22. If a "gecos" field in a passwd entry contained escaped characters, in
2827 particular, if it contained a \" sequence, Exim got it wrong when building
2828 a From: or a Sender: header from that name. A second bug also caused
2829 incorrect handling when an unquoted " was present following a character
2830 that needed quoting.
2832 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2833 was not being matched caselessly.
2835 24. Arranged for all hyphens in the exim.8 source to be escaped with
2838 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2839 properly. Recipient callout cache records were still being keyed to include
2840 the sender, even when use_sender was set false. This led to far more
2841 callouts that were necessary. The sender is no longer included in the key
2842 when use_sender is false.
2844 26. Added "control = submission" modifier to ACLs.
2846 27. Added the ${base62d: operator to decode base 62 numbers.
2848 28. dnsdb lookups can now access SRV records.
2850 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2851 the configuration file.
2853 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2854 (-v) mode. This makes the output for a verbose queue run more intelligible.
2856 31. Added a use_postmaster feature to recipient callouts.
2858 32. Added the $body_zerocount variable, containing the number of binary zero
2859 bytes in the message body.
2861 33. The time of last modification of the "new" subdirectory is now used as the
2862 "mailbox time last read" when there is a quota error for a maildir
2865 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2867 35. Added +ignore_unknown as a special item in host lists.
2869 36. Code for decoding IPv6 addresses in host lists is now included, even if
2870 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2871 address was recognized as an IP address, but was then not correctly decoded
2872 into binary, causing unexpected and incorrect effects when compared with
2879 1. Very minor rewording of debugging text in manualroute to say "list of
2880 hosts" instead of "hostlist".
2882 2. If verify=header_syntax was set, and a header line with an unqualified
2883 address (no domain) and a large number of spaces between the end of the
2884 name and the colon was received, the reception process suffered a buffer
2885 overflow, and (when I tested it) crashed. This was caused by some obsolete
2886 code that should have been removed. The fix is to remove it!
2888 3. When running in the test harness, delay a bit after writing a bounce
2889 message to get a bit more predictability in the log output.
2891 4. Added a call to search_tidyup() just before forking a reception process. In
2892 theory, someone could use a lookup in the expansion of smtp_accept_max_
2893 per_host which, without the tidyup, could leave open a database connection.
2895 5. Added the variables $recipient_data and $sender_data which get set from a
2896 lookup success in an ACL "recipients" or "senders" condition, or a router
2897 "senders" option, similar to $domain_data and $local_part_data.
2899 6. Moved the writing of debug_print from before to after the "senders" test
2902 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2903 problems for message scanning, either using a data ACL, or using
2904 local_scan() because the Received: header was not generated till after they
2905 were called (in order to set the time as the time of reception completion).
2906 I have revised the way this works. The header is now generated after the
2907 body is received, but before the ACL or local_scan() are called. After they
2908 are run, the timestamp in the header is updated.
2914 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2915 before starting a queue runner without re-exec. This happened only when
2916 deliver_drop_privilege was set or when the Exim user was set to root. The
2917 effect of the bug was that timeouts during subsequent deliveries caused
2918 crashes instead of being properly handled. The handler is now left at its
2919 default (and expected) setting.
2921 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2922 message, again when deliver_drop_privilege is set or Exim is run as root.
2923 The bug described in (1) was not present in this case, but the tidying up
2924 of the other signals was missing. I have made the two cases consistent.
2926 3. The ignore_target_hosts setting on a manualroute router was being ignored
2927 for hosts that were looked up using the /MX notation.
2929 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2932 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2933 operated on the sender address. After changing the $sender_address to <>
2934 for the sender address verify, Exim was re-instated it as the original
2935 (before rewriting) address, but remembering that it had rewritten it, so it
2936 wasn't rewriting it again. This bug also had the effect of breaking the
2937 sender address verification caching when the sender address was rewritten.
2939 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2940 This has been changed so that if the ip literal address matches
2941 ignore_target_hosts, the router declines.
2943 7. Added expansion conditions match_domain, match_address, and match_local_
2944 part (NOT match_host).
2946 8. The placeholder for the Received: header didn't have a length field set.
2948 9. Added code to Exim itself and to exim_lock to test for a specific race
2949 condition that could lead to file corruption when using MBX delivery. The
2950 issue is with the lockfile that is created in /tmp. If this file is removed
2951 after a process has opened it but before that process has acquired a lock,
2952 there is the potential for a second process to recreate the file and also
2953 acquire a lock. This could lead to two Exim processes writing to the file
2954 at the same time. The added code performs the same test as UW imapd; it
2955 checks after acquiring the lock that its file descriptor still refers to
2956 the same named file.
2958 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2959 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2962 11. Added the smtp_active_hostname option. If used, this will typically be made
2963 to depend on the incoming interface address. Because $interface_address is
2964 not set up until the daemon has forked a reception process, error responses
2965 that can happen earlier (such as "too many connections") no longer contain
2968 12. If an expansion in a condition on a "warn" statement fails because a lookup
2969 defers, the "warn" statement is abandoned, and the next ACL statement is
2970 processed. Previously this caused the whole ACL to be aborted.
2972 13. Added the iplsearch lookup type.
2974 14. Added ident_timeout as a log selector.
2976 15. Added tls_certificate_verified as a log selector.
2978 16. Added a global option tls_require_ciphers (compare the smtp transport
2979 option of the same name). This controls incoming TLS connections.
2981 17. I finally figured out how to make tls_require_ciphers do a similar thing
2982 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2983 before starting the TLS session.
2985 18. Tabs are now shown as \t in -bP output.
2987 19. If the log selector return_path_on_delivery was set, Exim crashed when
2988 bouncing a message because it had too many Received: header lines.
2990 20. If two routers both had headers_remove settings, and the first one included
2991 a superfluous trailing colon, the final name in the first list and the
2992 first name in the second list were incorrectly joined into one item (with a
2993 colon in the middle).
2999 1. Added -C and -D options to the exinext utility, mainly to make it easier
3000 to include in the automated testing, but these could be helpful when
3001 multiple configurations are in use.
3003 2. The exinext utility was not formatting the output nicely when there was
3004 an alternate port involved in the retry record key, nor when there was a
3005 message id as well (for retries that were specific to a specific message
3006 and a specific host). It was also confused by IPv6 addresses, because of
3007 the additional colons they contain. I have fixed the IPv4 problem, and
3008 patched it up to do a reasonable job for IPv6.
3010 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3011 delivery, the log line now contains "pipelined" if PIPELINING was used.
3013 4. An SMTP transport process used to panic and die if the bind() call to set
3014 an explicit outgoing interface failed. This has been changed; it is now
3015 treated in the same way as a connect() failure.
3017 5. A reference to $sender_host_name in the part of a conditional expansion
3018 that was being skipped was still causing a DNS lookup. This no longer
3021 6. The def: expansion condition was not recognizing references to header lines
3022 that used bh_ and bheader_.
3024 7. Added the _cache feature to named lists.
3026 8. The code for checking quota_filecount in the appendfile transport was
3027 allowing one more file than it should have been.
3029 9. For compatibility with Sendmail, the command line option
3037 and sets the incoming protocol and host name (for trusted callers). The
3038 host name and its colon can be omitted when only the protocol is to be set.
3039 Note the Exim already has two private options, -pd and -ps, that refer to
3040 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3041 "s", but I don't think that's a major issue.
3043 10. A number of refactoring changes to the code, none of which should affect
3046 (a) The number of logging options was getting close to filling up the
3047 32-bit word that was used as a bit map. I have split them into two classes:
3048 those that are passed in the argument to log_write(), and those that are
3049 only ever tested independently outside of that function. These are now in
3050 separate 32-bit words, so there is plenty of room for expansion again.
3051 There is no change in the user interface or the logging behaviour.
3053 (b) When building, for example, log lines, the code previously used a
3054 macro that called string_cat() twice, in order to add two strings. This is
3055 not really sufficiently general. Furthermore, there was one instance where
3056 it was actually wrong because one of the argument was used twice, and in
3057 one call a function was used. (As it happened, calling the function twice
3058 did not affect the overall behaviour.) The macro has been replaced by a
3059 function that can join an arbitrary number of extra strings onto a growing
3062 (c) The code for expansion conditions now uses a table and a binary chop
3063 instead of a serial search (which was left over from when there were very
3064 few conditions). Also, it now recognizes conditions like "pam" even when
3065 the relevant support is not compiled in: a suitably worded error message is
3066 given if an attempt is made to use such a condition.
3068 11. Added ${time_interval:xxxxx}.
3070 12. A bug was causing one of the ddress fields not to be passed back correctly
3071 from remote delivery subprocesses. The field in question was not being
3072 subsequently used, so this caused to problems in practice.
3074 13. Added new log selectors queue_time and deliver_time.
3076 14. Might have fixed a bug in maildirsizefile handling that threw up
3077 "unexpected character" debug warnings, and recalculated the data
3078 unnecessarily. In any case, I expanded the warning message to give more
3081 15. Added the message "Restricted characters in address" to the statements in
3082 the default ACL that block characters like @ and % in local parts.
3084 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3085 Three changes have been made:
3087 (a) There was a serious bug; a negative response to MAIL caused the whole
3088 recipient domain to be cached as invalid, thereby blocking all messages
3089 to all local parts at the same domain, from all senders. This bug has
3090 been fixed. The domain is no longer cached after a negative response to
3091 MAIL if the sender used is not empty.
3093 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3096 (c) A new callout option, "use_sender" has been added for people who want
3097 the modified behaviour.
3103 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3106 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3107 indeed breaks things for older releases.
3109 3. Added additional logging to the case where there is a problem reading data
3110 from a filter that is running in a subprocess using a pipe, in order to
3111 try to track down a specific problem.
3113 4. Testing facility fudge: when running in the test harness and attempting
3114 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3115 getting "No route to host". Convert this to a timeout.
3117 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3120 6. Some OS don't have socklen_t but use size_t instead. This affects the
3121 fifth argument of getsockopt() amongst other things. This is now
3122 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3123 can be set for individual OS. I have set it for SunOS5, OSF1, and
3124 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3125 some earlier ones do not.
3127 7. Change 4.30/15 was not doing the test caselessly.
3129 8. The standard form for an IPv6 address literal was being rejected by address
3130 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3131 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3132 this, as well as the form without the "IPv6" on the front (but only when
3133 address literals are enabled, of course).
3135 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3137 10. Exim crashed if a message with an empty sender address specified by -f
3138 encountered a router with an errors_to setting. This could be provoked only
3139 by a command such as
3143 where an empty string was supplied; "<>" did not hit this bug.
3145 11. Installed PCRE release 4.5.
3147 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3148 remained set. It is now erased.
3150 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3151 times from message ids (which are base 36 rather than the normal 62).
3153 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3154 were being counted as actual protocol errors, and logged if the log
3155 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3156 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3157 and DATA following a set of rejected RCPTs do not count as protocol errors.
3158 In other words, Exim assumes they were pipelined, though this may not
3159 actually be the case. Of course, in all cases the client gets an
3160 appropriate error code.
3162 15. If a lookup fails in an ACL condition, a message about the failure may
3163 be available; it is used if testing the ACL cannot continue, because most
3164 such messages specify what the cause of the deferral is. However, some
3165 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3166 that caused an old message to be retained and used if a later statement
3167 caused a defer, replacing the real cause of the deferral.
3169 16. If an IP address had so many PTR records that the DNS lookup buffer
3170 was not large enough to hold them, Exim could crash while trying to process
3171 the truncated data. It now detects and logs this case.
3173 17. Further to 4.21/58, another change has been made: if (and only if) the
3174 first line of a message (the first header line) ends with CRLF, a bare LF
3175 in a subsequent header line has a space inserted after it, so as not to
3176 terminate the header.
3178 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3179 unnecessarily, used atoi() instead of strtol(), and didn't check the
3180 termination when getting file sizes from file names by regex.
3182 19. Completely re-implemented the support for maildirsize files, in the light
3183 of a number of problems with the previous contributed implementation
3184 (4.30/29). In particular:
3186 . If the quota is zero, the maildirsize file is maintained, but no quota is
3189 . If the maildir directory does not exist, it is created before any attempt
3190 to write a maildirsize file.
3192 . The quota value in the file is just a cache; if the quota is changed in
3193 the transport, the new value overrides.
3195 . A regular expression is available for excluding directories from the
3198 20. The autoreply transport checks the characters in options that define the
3199 message's headers; it allows continued headers, but it was checking with
3200 isspace() after an embedded newline instead of explicitly looking for a
3203 21. If all the "regular" hosts to which an address was routed had passed their
3204 expiry times, and had not reached their retry times, the address was
3205 bounced, even if fallback hosts were defined. Now Exim should go on to try
3208 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3209 equivalent code in the SMTP transport. Some hosts send humungous responses
3210 to HELO/EHLO, more than 1024 it seems.
3212 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3213 gives compiler warnings in some environments. I've now done it "properly",
3216 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3217 (because of problems with the built-in one) was declared to return uschar *
3218 instead of char *, causing compiler failure.
3220 25. Fixed a file descriptor leak when processing alias/forward files.
3222 26. Fixed a minor format string issue in dbfn.c.
3224 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3226 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3227 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3230 29. When a sender address is verified, it is cached, to save repeating the test
3231 when there is more than one recipient in a message. However, when the
3232 verification involves a callout, it is possible for different callout
3233 options to be set for different recipients. It is too complicated to keep
3234 track of this in the cache, so now Exim always runs a verification when a
3235 callout is required, relying on the callout cache for the optimization.
3236 The overhead is duplication of the address routing, but this should not be
3239 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3240 to be verified with callout=postmaster, and the main callout worked but the
3241 postmaster check failed, the verification correctly failed. However, if a
3242 subsequent RCPT command asked for sender verification *without* the
3243 postmaster check, incorrect caching caused this verification also to fail,
3246 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3247 it was not caching the DNS options (qualify_single, search_parents) that
3248 were used when the lookup failed. A subsequent lookup with different
3249 options therefore always gave the same answer, though there were cases
3250 where it should not have. (Example: a "domains = !$mx_any" option on a
3251 dnslookup router: the "domains" option is always processed without any
3252 widening, but the router might have qualify_single set.) Now Exim uses the
3253 cached value only when the same options are set.
3255 32. Added John Jetmore's "exipick" utility to the distribution.
3257 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3258 than a timeout (e.g. a certificate is required, and is not provided), an
3259 Exim server now closes the connection immediately. Previously it waited for
3260 the client to close - but if the client is SSL, it seems that they each
3261 wait for each other, leading to a delay before one of them times out.
3263 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3264 maintained 0.8.x compatibility because I don't think many are using it, and
3265 it is clearly obsolete.
3267 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3270 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3271 client certificate was expired. A simple patch fixes this, though I don't
3272 understand the full logic of why the verify callback is called multiple
3275 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3276 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3277 which causes problems with some clients (such as the Certicom SSL Plus
3278 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3279 disables the coutermeasure allowing Eudora to connect."
3281 38. Exim was not checking that a write() to a log file succeeded. This could
3282 lead to Bad Things if a log got too big, in particular if it hit a file
3283 size limit. Exim now panics and dies if it cannot write to a log file, just
3284 as it does if it cannot open a log file.
3286 39. Modified OS/Makefile-Linux so that it now contains
3288 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3290 The two -D definitions ensure that Exim is compiled with large file
3291 support, which makes it possible to handle log files that are bigger than
3294 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3295 instance) a domain was checked against a named list that involved a lookup,
3296 causing $domain_data to be set, then another domain was checked against the
3297 same list, then the first domain was re-checked, the value of $domain_data
3298 after the final check could be wrong. In particular, if the second check
3299 failed, it could be set empty. This bug probably also applied to
3302 41. The strip_trailing_dot option was not being applied to the address given
3303 with the -f command-line option.
3305 42. The code for reading a message's header from the spool was incrementing
3306 $received_count, but never initializing it. This meant that the value was
3307 incorrect (doubled) while delivering a message in the same process in which
3308 it was received. In the most common configuration of Exim, this never
3309 happens - a fresh exec is done - but it can happen when
3310 deliver_drop_privilege is set.
3312 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3313 it now includes up to 150 characters of the unexpected data in the log
3316 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3317 and building data strings. The size of both of these buffers was 10 000
3318 bytes - far larger than anybody would *ever* want, thought I. Needless to
3319 say, somebody hit the limit. I have increased the maximum line length to
3320 20 000 and the maximum data length of concatenated lines to 100 000. I have
3321 also fixed two bugs, because there was no checking on these buffers. Tsk,
3322 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3323 buffer is too small.
3325 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3326 lsearch lookups. Now it does.
3328 46. When parsing a route_list item in a manualroute router, a fixed-length
3329 buffer was used for the list of hosts. I made this 1024 bytes long,
3330 thinking that nobody would ever have a list of hosts that long. Wrong.
3331 Somebody had a whole pile of complicated expansion conditions, and the
3332 string was silently truncated, leading to an expansion error. It turns out
3333 that it is easier to change to an unlimited length (owing to other changes
3334 that have happened since this code was originally written) than to build
3335 structure for giving a limitation error. The length of the item that
3336 expands into the list of hosts is now unlimited.
3338 47. The lsearch lookup could not handle data where the length of text line was
3339 more than 4095 characters. Such lines were truncated, leading to shortened
3340 data being returned. It should now handle lines of any length.
3342 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3343 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3346 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3347 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3348 no longer happens in comment lines. A list of replacements is now placed
3349 at the head of all of the source files, except those whose only change is
3350 to replace PERL_COMMAND in the very first #! line.
3352 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3353 messages on the queue, with a bottom-up merge sort, using code contributed
3354 by Michael Haardt. This should make operations like -bp somewhat faster on
3355 large queues. It won't affect queue runners, except when queue_run_in_order
3358 51. Installed eximstats 1.31 in the distribution.
3360 52. Added support for SRV lookups to the dnslookup router.
3362 53. If an ACL referred to $message_body or $message_body_end, the value was not
3363 reset for any messages that followed in the same SMTP session.
3365 54. The store-handling optimization for building very long strings was not
3366 differentiating between the different store pools. I don't think this
3367 actually made any difference in practice, but I've tidied it.
3369 55. While running the routers to verify a sender address, $sender_address
3370 was still set to the sender address. This is wrong, because when routing to
3371 send a bounce to the sender, it would be empty. Therefore, I have changed
3372 it so that, while verifying a sender address, $sender_address is set to <>.
3373 (There is no change to what happens when verifying a recipient address.)
3375 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3376 target A or AAAA records (if not already returned) without resetting the
3377 qualify_single or search_parents options of the DNS resolver. These are
3378 inappropriate in this case because the targets of MX and SRV records must
3379 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3380 target that, when qualified, matched something in the local domain. These
3381 two options are now turned off when doing these lookups.
3383 57. It seems that at least some releases of Reiserfs (which does not have the
3384 concept of a fixed number of inodes) returns zero and not -1 for the
3385 number of available inodes. This interacted badly with check_spool_inodes,
3386 which assumed that -1 was the "no such thing" setting. What I have done is
3387 to check that the total number of inodes is greater than zero before doing
3388 the test of how many are available.
3390 58. When a "warn" ACL statement has a log_message modifier, the message is
3391 remembered, and not repeated. This is to avoid a lot of repetition when a
3392 message has many recipients that cause the same warning to be written.
3393 Howewer, Exim was preserving the list of already written lines for an
3394 entire SMTP session, which doesn't seem right. The memory is now reset if a
3395 new message is started.
3397 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3398 debugging output unless log_rewrite was also set.
3400 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3401 of (int)(handle) when we know that handle contains (void *)(-1).
3403 61. The Exim daemon panic-logs an error return when it closes the incoming
3404 connection. However "connection reset by peer" seems to be common, and
3405 isn't really an error worthy of noting specially, so that particular error
3408 62. When Exim is trying to find all the local interfaces, it used to panic and
3409 die if the ioctl to get the interface flags failed. However, it seems that
3410 on at least one OS (Solaris 9) it is possible to have an interface that is
3411 included in the list of interfaces, but for which you get a failure error
3412 for this call. This happens when the interface is not "plumbed" into a
3413 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3414 failure of the "get flags" call assumes that the interface is down.
3416 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3417 makes life easier for people who are doing arithmetic on fields extracted
3418 from dates, where you often get leading zeros that should not be
3419 interpreted as octal.
3421 64. Added qualify_domain to the redirect router, to override the global
3424 65. If a pathologically long header line contained very many addresses (the
3425 report of this problem mentioned 10 000) and each of them was rewritten,
3426 Exim could use up a very large amount of memory. (It kept on making new
3427 copies of the header line as it rewrote, and never released the old ones.)
3428 At the expense of a bit more processing, the header rewriting function has
3429 been changed so that it no longer eats memory in this way.
3431 66. The generation of the Received: header has been moved from the time that a
3432 message starts to be received, to the time that it finishes. The timestamp
3433 in the Received: header should now be very close to that of the <= log
3434 line. There are two side-effects of this change:
3436 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3437 logged header lines no longer include the local Received: line, because
3438 it has not yet been created. The same applies to a copy of the message
3439 that is returned to a non-SMTP sender when a message is rejected.
3441 (b) When a filter file is tested using -bf, no additional Received: header
3442 is added to the test message. After some thought, I decided that this
3445 This change does not affect the value of $received_for. It is still set
3446 after address rewriting, but before local_scan() is called.
3448 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3450 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3451 gave an unhelpful panic error message, and a defer error. I have managed to
3452 change this behaviour so that it now rejects any supplied certificate,
3453 which seems right, as the list of acceptable certificates is empty.
3455 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3456 gave an unhelpful defer error. I have not managed to make this reject any
3457 supplied certificates, but the error message it gives is "no certificate
3458 supplied", which is not helpful.
3460 70. exigrep's output now also includes lines that are not associated with any
3461 message, but which match the given pattern. Implemented by a patch from
3462 Martin Sluka, which also tidied up the Perl a bit.
3464 71. Recipient callout verification, like sender verification, was using <> in
3465 the MAIL FROM command. This isn't really the right thing, since the actual
3466 sender may affect whether the remote host accepts the recipient or not. I
3467 have changed it to use the actual sender in the callout; this means that
3468 the cache record is now keyed on a recipient/sender pair, not just the
3469 recipient address. There doesn't seem to be a real danger of callout loops,
3470 since a callout by the remote host to check the sender would use <>.
3471 [SEE ABOVE: changed after hitting problems.]
3473 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3474 temporary errors. However, in the case of such a code being given after
3475 the end of a data transmission (i.e. after ".") Exim was failing to write
3476 a retry record for the message. (Yes, there was some broken host that was
3477 actually sending 8xx at this point.)
3479 73. An unknown lookup type in a host list could cause Exim to panic-die when
3480 the list was checked. (An example that provoked this was putting <; in the
3481 middle of a list instead of at the start.) If this happened during a DATA
3482 ACL check, a -D file could be left lying around. This kind of configuration
3483 error no longer causes Exim to die; instead it causes a defer errror. The
3484 incident is still logged to the main and panic logs.
3486 74. Buglet left over from Exim 3 conversion. The message "too many messages
3487 in one connection" was written to the rejectlog but not the mainlog, except
3488 when address rewriting (yes!) was being logged.
3490 75. Added write_rejectlog option.
3492 76. When a system filter was run not as root (that is, when system_filter_user
3493 was set), the values of the $n variables were not being returned to the
3494 main process; thus, they were not subsequently available in the $sn
3497 77. Added +return_path_on_delivery log selector.
3499 78. A connection timeout was being treated differently from recipients deferred
3500 when testing hosts_max_try with a message that was older than the host's
3501 retry timeout. (The host should not be counted, thus allowing all hosts to
3502 be tried at least once before bouncing.) This may have been the cause of an
3503 occasionally reported bug whereby a message would remain on the queue
3504 longer than the retry timeout, but would be bounced if a delivery was
3505 forced. I say "may" because I never totally pinned down the problem;
3506 setting up timeout/retry tests is difficult. See also the next item.
3508 79. The ultimate address timeout was not being applied to errors that involved
3509 a combination of host plus message (for example, a timeout on a MAIL
3510 command). When an address resolved to a number of possible hosts, and they
3511 were not all tried for each delivery (e.g. because of hosts_max_try), a
3512 message could remain on the queue longer than the retry timeout.
3514 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3517 81. Fixed an obscure SMTP outgoing bug which required at least the following
3518 conditions: (a) there was another message waiting for the same server;
3519 (b) the server returned 5xx to all RCPT commands in the first message so
3520 that the message was not completed; (c) the server dropped the connection
3521 or gave a negative response to the RSET that Exim sends to abort the
3522 transaction. The observed case was a dropped connection after DATA that had
3523 been sent in pipelining mode. That is, the server had advertised PIPELINING
3524 but was not implementing it correctly. The effect of the bug was incorrect
3525 behaviour, such as trying another host, and this could lead to a crash.
3531 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3532 and daemon.c were passed as pointers to ints; they should have been
3533 pointers to socklen_t variables (which are typically unsigned ints).
3535 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3538 3. Fixed a really odd bug that affected only the testing scheme; patching a
3539 certain fixed string in the binary changed the value of another string that
3540 happened to be identical to the end of the original first string.
3542 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3543 name", it returns that address as the IP address. On some operating
3544 systems (e.g. Solaris), it also passes back the IP address string as the
3545 "host name". However, on others (e.g. Linux), it passes back an empty
3546 string. Exim wasn't checking for this, and was changing the host name to an
3547 empty string, assuming it had been canonicized.
3549 5. Although rare, it is permitted to have more than one PTR record for a given
3550 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3551 all the names associated with an address, because they do in Solaris.
3552 However, it seems that they do not in Linux for data that comes from the
3553 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3554 I found this out when I moved to a new Linux workstation and tried to run
3555 the Exim test suite.
3557 To get round this problem I have changed the code so that it now does its
3558 own call to the DNS to look up PTR records when searching for a host name.
3559 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3560 addresses that are only in /etc/hosts are still found.
3562 This behaviour is, however, controlled by an option called host_lookup_
3563 order, which defaults to "bydns:byaddr". If people want to use the other
3564 order, or indeed, just use one or the other means of lookup, they can
3565 specify it in this variable.
3567 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3568 some operating systems, this comes back from gethostbyaddr() as an empty
3569 string, and this is what Exim used to test for. However, it seems that in
3570 other systems, "." is yielded. Exim now tests for this case too.
3572 7. The values of check_spool_space and check_log_space are now held internally
3573 as a number of kilobytes instead of an absolute number of bytes. If a
3574 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3575 kilobyte. This means that much larger values can be stored.
3577 8. Exim monitor: an attempt to get the action menu when not actually pointing
3578 at a message produces an empty menu entitled "No message selected". This
3579 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3580 no entries in it ("Shell widget menu has zero width and/or height"). So I
3581 have added a single, blank menu entry in this case.
3583 9. Added ${quote_local_part.
3585 10. MIME decoding is now applied to the contents of Subject: header lines when
3588 11. Now that a reference to $sender_host_address automatically causes a reverse
3589 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3590 host lookup before query-style lookups in lists that might use this
3591 variable. This has therefore been abolished, and the "net-" prefix is no
3592 longer necessary for query-style lookups.
3594 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3595 have been a typo for LFLAGS, so it has been changed.
3597 13. The install script calls Exim with "-C /dev/null" in order to find the
3598 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3599 to be output. Howeve, since Exim outputs its version number before the
3600 error, it didn't break the script. It just looked ugly. I fixed this by
3601 always allowing "-C /dev/null" if the caller is root.
3603 14. Ignore overlarge ACL variable number when reading spool file - insurance
3604 against a later release with more variables having written the file.
3606 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3607 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3608 this, as well as the form without the "IPv6" on the front.
3610 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3611 OS/Makefile-Darwin file.
3613 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3614 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3616 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3617 variable. However, some OS use ints for inodes. I've added cast to long int
3618 to get rid of the compiler warning.
3620 19. I had forgotten to lock out "/../" in configuration file names when
3621 ALT_CONFIG_PREFIX was set.
3623 20. Routers used for verification do not need to specify transports. However,
3624 if such a router generated a host list, and callout was configured, Exim
3625 crashed, because it could not find a port number from the (non-existent)
3626 transport. It now assumes port 25 in this circumstance.
3628 21. Added the -t option to exigrep.
3630 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3631 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3632 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3633 them is set, LOOKUP_LSEARCH is forced.
3635 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3637 24. Added sender and host information to the "rejected by local_scan()" log
3638 line; previously there was no indication of these.
3640 25. Added .include_if_exists.
3642 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3643 Linux. It turns out that not all file systems support this. Apparently some
3644 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3645 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3646 not supported on the file descriptor, is now ignored when Exim is trying to
3647 sync a directory. This applies only to Linux.
3649 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3651 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3652 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3655 29. Added support for maildirsize files from supplied patch (modified a bit).
3657 30. The use of :fail: followed by an empty string could lead Exim to respond to
3658 sender verification failures with (e.g.):
3660 550 Verification failed for <xxx>
3661 550 Sender verify failed
3663 where the first response line was missing the '-' that indicates it is not
3664 the final line of the response.
3666 31. The loop for finding the name of the user that called Exim had a hardwired
3667 limit of 10; it now uses the value of finduser_retries, which is used for
3668 all other user lookups.
3670 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3673 33. Exim was neglecting to zero errno before one call of strtol() when
3674 expanding a string and expecting an integer value. On some systems this
3675 resulted in spurious "integer overflow" errors. Also, it was casting the
3676 result into an int without checking.
3678 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3679 did not work. The code looks as if it has *never* worked, though it appears
3680 to have been documented since at least releast 1.62. I have made it work.
3682 35. The "timeout_DNS" error in retry rules, also documented since at least
3683 1.62, also never worked. As it isn't clear exactly what this means, and
3684 clearly it isn't a major issue, I have abolished the feature by treating it
3685 as "timeout", and writing a warning to the main and panic logs.
3687 36. The display of retry rules for -brt wasn't always showing the error code
3690 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3691 timeout_connect_A, timeout_connect_MX.
3693 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3694 to the empty sender.
3696 39. The daemon was not analysing the content of -oX till after it had closed
3697 stderr and disconnected from the controlling terminal. This meant that any
3698 syntax errors were only noted on the panic log, and the return code from
3699 the command was 0. By re-arranging the code a little, I've made the
3700 decoding happen first, so such errors now appear on stderr, and the return
3701 code is 1. However, the actual setting up of the sockets still happens in
3702 the disconnected process, so errors there are still only recorded on the
3705 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3706 connections (as happens on some IP stacks) was logged at start up time as
3707 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3708 it from "IPv6 and IPv4", which means that two separate sockets are being
3711 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3712 says whether AF_INET or AF_INET6 was passed as an argument.
3714 42. Exiwhat output was messed up when time zones were included in log
3717 43. Exiwhat now gives more information about the daemon's listening ports,
3718 and whether -tls-on-connect was used.
3720 44. The "port" option of the smtp transport is now expanded.
3722 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3723 silently ignored. Now an error message is written to the main and panic
3726 46. There's a new ACL modifier called "logwrite" which writes to a log file
3727 as soon as it is encountered.
3729 47. Added $local_user_uid and $local_user_gid at routing time.
3731 48. Exim crashed when trying to verify a sender address that was being
3734 49. Exim was recognizing only a space character after ".include". It now also
3735 recognizes a tab character.
3737 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3738 extracting the relevant information from the specification. The man page no
3739 longer contains scrambled data for the -d option, and I've added a section
3740 at the front about calling Exim under different names.
3742 51. Added "extra_headers" argument to the "mail" command in filter files.
3744 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3747 53. Installed eximstats 1.29.
3749 54. Added transport_filter_timeout as a generic transport option.
3751 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3752 Cc: header lines. This was required by RFC 822, but it not required by RFC
3755 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3756 incoming messages that did not have them. Now it does so only if the
3757 message originates locally, that is, if there is no associated remote host
3758 address. When Resent- header lines are present, this applies to the Resent-
3759 lines rather than the non-Resent- lines.
3761 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3762 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3764 58. Messages for configuration errors now include the name of the main
3765 configuration file - useful now that there may be more than one file in a
3766 list (.included file names were always shown).
3768 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3769 for those rare installations that do not start the daemon as root or run it
3770 setuid root. I've cut out the call to initgroups() if the daemon is not
3773 60. The Exim user and group can now be bound into the binary as text strings
3774 that are looked up at the start of Exim's processing.
3776 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3778 62. Added $mailstore_basename variable.
3780 63. Installed patch to sieve.c from Michael Haardt.
3782 64. When Exim failed to open the panic log after failing to open the main log,
3783 the original message it was trying to log was written to stderr and debug
3784 output, but if they were not available (the usual case in production), it
3785 was lost. Now it is written to syslog before the two lines that record the
3786 failures to open the logs.
3788 65. Users' Exim filters run in subprocesses under the user's uid. It is
3789 possible for a "deliver" command or an alias in a "personal" command to
3790 provoke an address rewrite. If logging of address rewriting is configured,
3791 this fails because the process is not running as root or exim. There may be
3792 a better way of dealing with this, but for the moment (because 4.30 needs
3793 to be released), I have disabled address rewrite logging when running a
3794 filter in a non-root, non-exim process.
3800 1. The buildconfig auxiliary program wasn't quoting the value set for
3801 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3802 not defined. This bug was masked in 4.22 by the effect that was fixed in
3805 2. Some messages that were rejected after a message id was allocated were
3806 shown as "incomplete" by exigrep. It no longer does this for messages that
3807 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3809 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3810 have allow_domain_literals set, the ID did not get logged in the <= line.
3811 Domain literals are now always recognized in Message-ID: header lines.
3813 4. The first argument for a ${extract expansion item is the key name or field
3814 number. Leading and trailing spaces in this item were not being ignored,
3815 causing some misleading effects.
3817 5. When deliver_drop_privilege was set, single queue runner processes started
3818 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3819 same command in the process it spins off) were not dropping privilege.
3821 6. When the daemon running as "exim" started a queue runner, it always
3822 re-executed Exim in the spun-off process. This is a waste of effort when
3823 deliver_drop_privilege is set. The new process now just calls the
3824 queue-runner function directly.
3830 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3833 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3834 ignored. Though the use of -oP was forcing the writing of a pid file, it
3835 was always written to the default place.
3837 3. If the message "no IP address found for host xxxx" is generated during
3838 incoming verification, it is now followed by identification of the incoming
3839 connection (so you can more easily find what provoked it).
3841 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3843 5. Added some features to "harden" Exim a bit more against certain attacks:
3845 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3846 be put in Local/Makefile. This is like the never_users runtime option,
3847 but it cannot be overridden. The default setting is "root".
3849 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3850 prefix string with which any file named in a -C command line option
3853 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3854 is retained for -C and -D only if the caller of Exim is root. Without
3855 it, the exim user may also use -C and -D and retain privilege.
3857 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3858 command line option is disabled.
3860 6. Macro names set by the -D option must start with an upper case letter, just
3861 like macro names defined in the configuration file.
3863 7. Added "dereference=" facility to LDAP.
3865 8. Two instances of the typo "uknown" in the source files are fixed.
3867 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3868 the Configure-Makefile script screwed up while processing it.
3870 10. Incorporated PCRE 4.4.
3872 11. The SMTP synchronization check was not operating right at the start of an
3873 SMTP session. For example, it could not catch a HELO sent before the client
3874 waited for the greeting. There is now a check for outstanding input at the
3875 point when the greeting is written. Because of the duplex, asynchronous
3876 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3877 way, but not yet received, when the check is performed.
3879 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3880 on TCP/IP sockets, because this apparently causes some broken clients to
3883 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3884 unchanged) from the Cygwin maintainer.
3886 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3887 was supported instead of testing for mbx. Effectively a typo.
3889 15. The spa authenticator server code was not checking that the input it
3890 received was valid base64.
3892 16. The debug output line for the "set" modifier in ACLs was not showing the
3893 name of the variable that was being set.
3895 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3897 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3898 lookup on its own. Something else was needed to trigger the lookup. For
3899 example, a match in host_lookup or the need for a host name in a host list.
3900 Now, if $sender_host_name is referenced and the host name has not yet been
3901 looked up, a lookup is performed. If the lookup fails, the variable remains
3902 empty, and $host_lookup_failed is set to "1".
3904 19. Added "eqi" as a case-independent comparison operator.
3906 20. The saslauthd authentication condition could segfault if neither service
3907 nor realm was specified.
3909 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3910 error message that was logged was misleading, and incoming SMTP
3911 connections were dropped. The message is now more accurate, and temporary
3912 errors are given to SMTP connections.
3914 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3915 (except RSET) with a 421 error, until QUIT is received. However, it was
3916 failing to send a response to QUIT.
3918 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3919 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3920 ACL is now run after the helo_try_verify_hosts code.
3922 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3923 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3924 case-independent, so other case variants are also recognized). Apparently
3925 some systems use these upper case variants.
3927 25. If more than two messages were waiting for the same host, and a transport
3928 filter was specified for the transport, Exim sent two messages over the
3929 same TCP/IP connection, and then failed with "socket operation on non-
3930 socket" when it tried to send the third.
3932 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3934 27. The extern definition of crypt16() in expand.c was not being excluded when
3935 the OS had its own crypt16() function.
3937 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3938 as a preferred synonym for return_size_limit, both as an option and as an
3941 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3943 30. Changed the default configuration ACL to relax the local part checking rule
3944 for addresses that are not in any local domains. For these addresses,
3945 slashes and pipe symbols are allowed within local parts, but the sequence
3946 /../ is explicitly forbidden.
3948 31. SPA server authentication was not clearing the challenge buffer before
3951 32. log_message in a "warn" ACL statement was writing to the reject log as
3952 well as to the main log, which contradicts the documentation and doesn't
3953 seem right (because no rejection is happening). So I have stopped it.
3955 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3956 However, I am unable to do any testing of this.
3958 34. Fixed an infelicity in the appendfile transport. When checking directories
3959 for a mailbox, to see if any needed to be created, it was accidentally
3960 using path names with one or more superfluous leading slashes; tracing
3961 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3963 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3964 contents are added to the log line that is written for every discarded
3965 recipient. (Previously a log_message setting was ignored.)
3967 36. The ${quote: operator now quotes the string if it is empty.
3969 37. The install script runs exim in order to find its version number. If for
3970 some reason other than non-existence or emptiness, which it checks, it
3971 could not run './exim', it was installing it with an empty version number,
3972 i.e. as "exim-". This error state is now caught, and the installation is
3975 38. An argument was missing from the function that creates an error message
3976 when Exim fails to connect to the socket for saslauthd authentication.
3977 This could cause Exim to crash, or give a corrupted message.
3979 39. Added isip, isip4, and isip6 to ${if conditions.
3981 40. The ACL variables $acl_xx are now saved with the message, and can be
3982 accessed later in routers, transports, and filters.
3984 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3985 strings in the file are not string-expanded.
3987 42. If a MAIL command specified a SIZE value that was too large to fit into an
3988 int variable, the check against message_size_limit failed. Such values are
3989 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3990 day this will have to be increased, but I don't think I want to be around
3991 when emails are that large.
3998 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3999 iconv() is not standard in FreeBSD.
4001 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4002 IPv6 enabled. The observed symptom was a segmentation fault on return from
4003 the function os_common_find_running_interfaces() in src/os.c.
4005 3. In the check_special_case() function in daemon.c I had used "errno" as an
4006 argument name, which causes warnings on some systems. This was basically a
4007 typo, since it was named "eno" in the comments!
4009 4. The code that waits for the clock to tick (at a resolution of some fraction
4010 of a second) so as to ensure message-id uniqueness was always waiting for
4011 at least one whole tick, when it could have waited for less. [This is
4012 almost certainly not relevant at current processor speeds, where it is
4013 unlikely to ever wait at all. But we try to future-proof.]
4015 5. The function that sleeps for a time interval that includes fractions of a
4016 second contained a race. It did not block SIGALRM between setting the
4017 timer, and suspending (a couple of lines later). If the interval was short
4018 and the sigsuspend() was delayed until after it had expired, the suspension
4019 never ended. On busy systems this could lead to processes getting stuck for
4022 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4023 process, before it forks any delivery processes. The open lookup caching
4024 mechanism meant that the open file or database connection was passed into
4025 the delivery process. The problem was that delivery processes always tidy
4026 up cached lookup data. This could cause a problem for the next delivery
4027 process started by the queue runner, because the external queue runner
4028 process does not know about the closure. So the next delivery process
4029 still has data in the lookup cache. In the case of a file lookup, there was
4030 no problem because closing a file descriptor in a subprocess doesn't affect
4031 the parent. However, if the lookup was caching a connection to a database,
4032 the connection was closed, and the second delivery process was likely to
4033 see errors such as "PGSQL: query failed: server closed the connection
4034 unexpectedly". The problem has been fixed by closing all cached lookups
4035 in a queue runner before running a delivery process.
4037 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4038 seem to have the "const" qualifier which it has on other OS. I've
4041 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4042 *of the same type* (client or server) with the same public name that an
4043 error should be diagnosed.
4045 9. When Exim looked up a host name for an IP address, but failed to find the
4046 original IP address when looking up the host name (a safety check), it
4047 output the message "<ip address> does not match any IP for NULL", which was
4048 confusing, to say the least. The bug was that the host name should have
4049 appeared instead of "NULL".
4051 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4052 user that is built into the binary, and the -C or -D options is used, root
4053 privilege is dropped before the configuration file is read. In addition,
4054 logging is switched to stderr instead of the normal log files. If the
4055 configuration then re-defines the Exim user, the unprivileged environment
4056 is probably not what is expected, so Exim logs a panic warning message (but
4059 However, if deliver_drop_privilege is set, the unprivileged state may well
4060 be exactly what is intended, so the warning has been cut out in that case,
4061 and Exim is allowed to try to write to its normal log files.
4067 1. smtp_return_error_details was not giving details for temporary sender
4068 or receiver verification errors.
4070 2. Diagnose a configuration error if two authenticators have the same public
4073 3. Exim used not to create the message log file for a message until the first
4074 delivery attempt. This could be confusing when incoming messages were held
4075 for policy or load reasons. The message log file is now created at the time
4076 the message is received, and an initial "Received" line is written to it.
4078 4. The automatically generated man page for command line options had a minor
4079 bug that caused no ill effects; however, a more serious problem was that
4080 the procedure for building the man page automatically didn't always
4081 operate. Consequently, release 4.20 contains an out-of-date version. This
4082 shouldn't happen again.
4084 5. When building Exim with embedded Perl support, the script that builds the
4085 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4086 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4088 6. The freeze_tell option was not being used for messages that were frozen on
4089 arrival, either by an ACL or by local_scan().
4091 7. Added the smtp_incomplete_transaction log selector.
4093 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4094 was accepting AUTH without a new EHLO.
4096 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4097 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4100 10. Logging of TCP/IP connections (when configured) now happens in the main
4101 daemon process instead of the child process, so that the TCP/IP connection
4102 count is more accurate (but it can never be perfect).
4104 11. The use of "drop" in a nested ACL was not being handled correctly in the
4105 outer ACL. Now, if condition failure induced by the nested "drop" causes
4106 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4107 or "require"), the connection is dropped.
4109 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4110 that yield "discard" can now be used with an "accept" or a "discard" verb,
4111 but an error is generated for any others (because I can't see a useful way
4112 to define what should happen).
4114 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4115 are now processed in the same way as lines in the Exim configuration file.
4116 In particular, continuation lines are supported.
4118 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4120 15. Added -ti meaning -t -i.
4122 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4123 domains, and warn by logging if others are found.
4125 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4126 returned from GIFCONF when Exim is trying to find the list of interfaces on
4127 a host. The code in os.c has been modified to copy each ifreq to an aligned
4128 structure in all cases.
4130 Also, in some cases, the returned ifreq's were being copied to a 'struct
4131 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4132 means the last couple of bytes of an IPv6 address could be chopped if the
4133 ifreq contained only a normal sockaddr (14 bytes storage).
4135 18. Named domain lists were not supported in the hosts_treat_as_local option.
4136 An entry such as +xxxx was not recognized, and was treated as a literal
4139 19. Ensure that header lines added by a DATA ACL are included in the reject log
4140 if the ACL subsequently rejects the message.
4142 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4143 MD5 (which is deprecated).
4145 21. When testing a filter file using -bf, Exim was writing a message when it
4146 took the sender from a "From " line in the message, but it was not doing so
4147 when it took $return_path from a Return-Path: header line. It now does.
4149 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4150 with a valid header line field name (a series of printing characters
4151 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4153 23. Changed "disc" in the source to "disk" to conform to the documentation and
4154 the book and for uniformity.
4156 24. Ignore Sendmail's -Ooption=value command line item.
4158 25. When execve() failed while trying to run a command in a pipe transport,
4159 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4160 could be confused with a return value of 69 from the command itself. This
4161 has been changed to 127, the value the shell returns if it is asked to run
4162 a non-existent command. The wording for the related log line suggests a
4163 non-existent command as the problem.
4165 26. If received_header_text expands to an empty string, do not add a Received:
4166 header line to the message. (Well, it adds a token one on the spool, but
4167 marks it "old" so that it doesn't get used or transmitted.)
4169 27. Installed eximstats 1.28 (addition of -nt option).
4171 28. There was no check for failure on the call to getsockname() in the daemon
4172 code. This can fail if there is a shortage of resources on the system, with
4173 ENOMEM, for example. A temporary error is now given on failure.
4175 29. Contrary to the C standard, it seems that in some environments, the
4176 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4177 program. Exim now does this explicitly; it affects the formatting of
4178 timestamps using strftime().
4180 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4181 complaints. I've now initialized all the variables, to avoid this.
4183 32. Header lines added by a system filter were not being "seen" during
4184 transport-time rewrites.
4186 33. The info_callback() function passed to OpenSSL is set up with type void
4187 (*)(SSL *, int, int), as described somewhere. However, when calling the
4188 function (actually a macro) that sets it up, the type void(*)() is
4189 expected. I've put in a cast to prevent warnings from picky compilers.
4191 34. If a DNS black list lookup found a CNAME record, but there were no A
4192 records associated with the domain it pointed at, Exim crashed.
4194 35. If a DNS black list lookup returned more than one A record, Exim ignored
4195 all but the first. It now scans all returned addresses if a particular IP
4196 value is being sought. In this situation, the contents of the
4197 $dnslist_value variable are a list of all the addresses, separated by a
4200 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4201 to accept a host name and all its aliases if the forward lookup for any of
4202 them yielded the IP address of the incoming connection. Now it accepts only
4203 those names whose forward lookup yields the correct IP address. Any other
4204 names are discarded. This closes a loophole whereby a rogue DNS
4205 administrator could create reverse DNS records to break through a
4206 wildcarded host restriction in an ACL.
4208 37. If a user filter or a system filter that ran in a subprocess used any of
4209 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4210 the wrong values were passed to the pipe command ($thisaddress had the
4211 value of $0, $0 had the value of $1, etc). This bug was introduced by
4212 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4214 38. Improved the line breaking for long SMTP error messages from ACLs.
4215 Previously, if there was no break point between 40 and 75 characters, Exim
4216 left the rest of the message alone. Two changes have been made: (a) I've
4217 reduced the minimum length to 35 characters; (b) if it can't find a break
4218 point between 35 and 75 characters, it looks ahead and uses the first one
4219 that it finds. This may give the occasional overlong line, but at least the
4220 remaining text gets split now.
4222 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4223 file descriptors might be low, and that setting 1000 would always raise it.
4224 It turns out that in some environments, the limit is already over 1000 and
4225 that lowering it causes trouble. So now Exim takes care not to decrease it.
4227 40. When delivering a message, the value of $return_path is set to $sender_
4228 address at the start of routing (routers may change the value). By an
4229 oversight, this default was not being set up when an address was tested by
4230 -bt or -bv, which affected the outcome if any router or filter referred to
4233 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4234 log only when "message" or "log_message" are set. However, if one of the
4235 conditions was an address verification, or a call to a nested ACL, the
4236 messages generated by the underlying test were being passed through. This
4237 no longer happens. The underlying message is available in $acl_verify_
4238 message for both "message" and "log_message" expansions, so it can be
4239 passed through if needed.
4241 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4242 new expansion $bh_ to give the encoded byte string without charset
4243 translation. Translation happens only if iconv() is available; HAVE_ICONV
4244 indicates this at build time. HEADERS_CHARSET gives the charset to
4245 translate to; headers_charset can change it in the configuration, and
4246 "headers charset" can change it in an individual filter file.
4248 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4249 that creates RFC 2047 encoded "words" labels them as that charset instead
4250 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4251 expansion operator; (ii) when Exim creates a From: line for a local
4252 message; (iii) when a header line is rewritten to include a "phrase" part.
4254 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4255 buggy, causing it to skip the first lines of messages whose message ID
4256 ended in 'D'. This would not have bitten before Exim release 4.14, because
4257 message IDs were unlikely to end in 'D' before then. The effect was to have
4258 incorrect size information for certain domains.
4260 45. #include "config.h" was missing at the start of the crypt16.c module. This
4261 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4264 46. If there was a timeout during a "random" callout check, Exim treated it as
4265 a failure of the random address, and carried on sending RSET and the real
4266 address. If the delay was just some slowness somewhere, the response to the
4267 original RCPT would be taken as a response to RSET and so on, causing
4268 mayhem of various kinds.
4270 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4271 when I implemented it. It didn't allow for the fact that some option values
4272 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4273 the right test for positive values.
4275 48. Domain names in DNS records are case-independent. Exim always looks them up
4276 in lower case. Some resolvers return domain names in exactly the case they
4277 appear in the zone file, that is, they may contain uppercase letters. Not
4278 all resolvers do this - some return always lower case. Exim was treating a
4279 change of case by a resolver as a change of domain, similar to a widening
4280 of a domain abbreviation. This triggered its re-routing code and so it was
4281 trying to route what was effectively the same domain again. This normally
4282 caused routing to fail (because the router wouldn't handle the domain
4283 twice). Now Exim checks for this case specially, and just changes the
4284 casing of the domain that it ultimately uses when it transmits the message
4287 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4290 50. If a filter generated a file delivery with a non-absolute name (possible if
4291 no home directory exists for the router), the forbid_file option was not
4294 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4295 the existing equality matching.
4297 52. Exim was using ints instead of ino_t variables in some places where it was
4298 dealing with inode numbers.
4300 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4301 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4302 called TMPDIR, and if it finds it is different, it changes its value.
4304 54. The smtp_printf() function is now made available to local_scan() so
4305 additional output lines can be written before returning. There is also an
4306 smtp_fflush() function to enable the detection of a dropped connection.
4307 The variables smtp_input and smtp_batched_input are exported to
4310 55. Changed the default runtime configuration: the message "Unknown user"
4311 has been removed from the ACL, and instead placed on the localuser router,
4312 using the cannot_route_message feature. This means that any verification
4313 failures that generate their own messages won't get overridden. Similarly,
4314 the "Unrouteable address" message that was in the ACL for unverifiable
4315 relay addresses has also been removed.
4317 56. Added hosts_avoid_esmtp to the smtp transport.
4319 57. The exicyclog script was not checking for the esoteric option
4320 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4321 will work only if exicyclog is run under the appropriate euid.
4323 58. Following a discussion on the list, the rules by which Exim recognises line
4324 endings on incoming messages have been changed. The -dropcr and drop_cr
4325 options are now no-ops, retained only for backwards compatibility. The
4326 following line terminators are recognized: LF CRLF CR. However, special
4327 processing applies to CR:
4329 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4330 nor a local message in the state where . is a terminator.
4332 (ii) If a bare CR is encountered in a header line, an extra space is added
4333 after the line terminator so as not to end the header. The reasoning
4334 behind this is that bare CRs in header lines are most likely either
4335 to be mistakes, or people trying to play silly games.
4337 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4338 was being incorrectly given as 18 bytes larger than it should have been.
4339 This is a VOB (very old bug).
4341 60. This may never have affected anything current, but just in case it has:
4342 When the local host is found other than at the start of a list of hosts,
4343 the local host, those with the same MX, and any that follow, are discarded.
4344 When the list in question was part of a longer list of hosts, the following
4345 hosts (not currently being processed) were also being discarded. This no
4346 longer happens. I'm not sure if this situation could ever has previously
4349 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4352 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4353 header. This is something that is recommended in a new Internet Draft, and
4354 is something that is documented as being done by Sendmail. There are two
4355 possible values. For messages generated by the autoreply transport, Exim
4358 Auto-Submitted: auto-replied
4360 whereas for all other generated messages (e.g. bounces) it adds
4362 Auto-Submitted: auto-generated
4364 63. The "personal" condition in filters now includes a test for the
4365 Auto-Submitted: header. If it contains the string "auto-" the message it
4366 not considered personal.
4368 64. Added rcpt_include_affixes as a generic transport option.
4370 65. Added queue_only_override (default true).
4372 66. Added the syslog_duplication option.
4374 67. If what should have been the first header line of a message consisted of
4375 a space followed by a colon, Exim was mis-interpreting it as a header line.
4376 It isn't of course - it is syntactically invalid and should therefore be
4377 treated as the start of the message body. The misbehaviour could have
4378 caused a number of strange effects, including loss of data in subsequent
4379 header lines, and spool format errors.
4381 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4382 client host had authenticated. This control can now be exercised by an ACL
4383 for more flexibility.
4385 69. By default, callouts do not happen when testing with -bh. There is now a
4386 variant, -bhc, which does actually run the callout code, including
4387 consulting and updating the callout cache.
4389 70. Added support for saslauthd authentication, courtesy of Alexander
4392 71. If statvfs() failed on the spool or log directories while checking their
4393 size for availability, Exim confusingly gave the error "space shortage".
4394 Furthermore, in debugging mode it crashed with a floating point exception.
4395 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4396 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4397 serious problem, Exim now writes to the main and panic logs when this
4398 happens, with details of the failure. It then refuses to accept the
4399 incoming message, giving the message "spool directory problem" or "log
4400 directory problem" with a 421 code for SMTP messages.
4402 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4403 0, 1, and 2 exist, because some OS complain for execs without them (see
4404 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4405 descriptors. However, the code omitted to check that the open succeeded,
4406 causing mysterious errors if for some reason the permissions on /dev/null
4407 got screwed. Now Exim writes a message to the main and panic logs, and
4408 bombs out if it can't open /dev/null.
4410 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4411 interact so that it is all more flexible. It is supposed to remain
4412 backwards compatible. Also added extra_local_interfaces.
4414 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4415 to bomb out with an assertion failure - to the client this appears as a
4416 connection drop. This problem occurs in the part of the code that was taken
4417 from the Samba project. Fortunately, the assertion is in a very simple
4418 function, so I have fixed this by reproducing the function inline in the
4419 one place where it is called, and arranging for authentication to fail
4420 instead of killing the process with assert().
4422 75. The SPA client code was not working when the server requested OEM rather
4423 than Unicode encoding.
4425 76. Added code to make require_files with a specific uid setting more usable in
4426 the case where statting the file as root fails - usually a non-root-mounted
4427 NFS file system. When this happens and the failure is EACCES, Exim now
4428 forks a subprocess and does the per-uid checking as the relevant uid.
4430 77. Added process_log_path.
4432 78. If log_file_path was not explicitly set, a setting of check_log_space or
4433 check_log_inodes was ignored.
4435 79. If a space check for the spool or log partitions fails, the incident is now
4436 logged. Of course, in the latter case the data may get lost...
4438 80. Added the %p formatting code to string_format() so that it can be used to
4439 print addresses in debug_print(). Adjusted all the address printing in the
4440 debugging in store.c to use %p rather than %d.
4442 81. There was a concern that a line of code in smtp_in.c could overflow a
4443 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4444 initially expressed, the concern was not well-founded, because trailing
4445 spaces are removed early. However, if the trailing spaces were followed by
4446 a NULL, they did not get removed, so the overflow was possible. Two fixes
4449 (a) I re-wrote the offending code in a cleaner fashion.
4450 (b) If an incoming SMTP command contains a NULL character, it is rejected
4453 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4454 runs initgroups(), so that if the Exim user is in any additional groups,
4455 they will be used during message reception.
4461 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob