1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.443 2006/12/11 14:15:59 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.
341 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
343 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
344 long ago, but noone interested enough thought of it. Let's just say that
345 the "1.1" means that there are some new functions that weren't there at
346 some point in the past.
348 PH/51 Error processing for expansion failure of helo_data from an smtp
349 transport during callout processing was broken.
355 SC/01 Use a glob alias rather than an array ref in eximstats generated
356 parser. This improves both readability and performance.
358 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
359 Don't display local sender or destination tables in eximstats unless
360 there is data to show.
361 Added average volumes into the eximstats top table text output.
363 SC/03 Collect data on the number of addresses (recipients) as well
364 as the number of messages in eximstats.
366 TF/01 Correct an error in the documentation for the redirect router. Exim
367 does (usually) call initgroups() when daemonizing.
369 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
370 with consistent privilege compared to when running as a daemon.
372 TF/03 Note in the spec that $authenticated_id is not set for local
373 submissions from trusted users.
375 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
376 Thanks to Dean Brooks <dean@iglou.com> for the patch.
378 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
379 by adding some example configuration directives to the default
380 configuration file. A little bit of work is required to uncomment the
381 directives and define how usernames and passwords are checked, but
382 there is now a framework to start from.
384 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
385 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
386 without this. I don't know how relevant this is to other LDAP libraries.
388 PH/02 Add the verb name to the "unknown ACL verb" error.
390 PH/03 Magnus Holmgren's patch for filter_prepend_home.
392 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
394 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
395 directory not expanded when it should be if an expanded home directory
396 was set for the address (which is overridden by the transport).
398 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
401 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
402 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
403 because it is too late at that time, and has no effect.
405 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
406 security issue with \' (bugzilla #107). I could not use the
407 PQescapeStringConn() function, because it needs a PGconn value as one of
410 PH/08 When testing addresses using -bt, indicate those final addresses that
411 are duplicates that would not cause an additional delivery. At least one
412 person was confused, thinking that -bt output corresponded to deliveries.
413 (Suppressing duplicates isn't a good idea as you lose the information
414 about possibly different redirections that led to the duplicates.)
416 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
417 systems where poll() doesn't work, in particular OS X.
419 PH/10 Added more information to debugging output for retry time not reached.
421 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
422 operations in malware.c.
424 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
427 PH/13 If write_rejectlog was set false when logging was sent to syslog with
428 syslog_duplication set false, log lines that would normally be written
429 both the the main log and to the reject log were not written to syslog at
432 PH/14 In the default configuration, change the use of "message" in ACL warn
433 statements to "add_header".
435 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
436 not followed by a command (e.g. "seen endif").
438 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
439 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
442 PH/17 Added extra conditions to the default value of delay_warning_condition
446 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
447 { match{$h_precedence:}{(?i)bulk|list|junk} } \
448 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
451 The Auto-Submitted: and various List- headers are standardised, whereas I
452 don't think Precedence: ever was.
454 PH/18 Refactored debugging code in route_finduser() to show more information,
455 in particular, the error code if getpwnam() issues one.
457 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
458 This is apparently needed in addition to the PH/07 change above to avoid
459 any possible encoding problems.
461 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
462 but not after initializing Perl.
464 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
465 output them only if debugging. By default they are written stderr,
466 apparently, which is not desirable.
468 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
471 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
474 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
476 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
477 authenticated or an ident call has been made. Suppress the default
478 values for $authenticated_id and $authenticated_sender (but permit -oMai
479 and -oMas) when testing with -bh.
481 PH/24 Re-jigged the order of the tests in the default configuration so that the
482 tests for valid domains and recipients precede the DNS black list and CSA
483 tests, on the grounds that those ones are more expensive.
485 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
486 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
487 This bug exists in every version of Exim that I still have, right back to
490 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
491 However, an attempt to turn on case-sensitivity in a regex key by
492 including (?-i) didn't work because the subject string was already
493 lowercased, and the effects were non-intuitive. It turns out that a
494 one-line patch can be used to allow (?-i) to work as expected.
500 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
501 other effects) broke the use of negated acl sub-conditions.
503 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
506 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
507 "Deny" causes Exim to reject the incoming connection with a 554 error.
508 Unfortunately, if there is a major crisis, such as a disk failure,
509 tcp-wrappers gives "deny", whereas what one would like would be some
510 kind of temporary error. A kludge has been added to help with this.
511 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
512 554 error is used if errno is still zero or contains ENOENT (which occurs
513 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
516 PH/03 Add -lutil to the default FreeBSD LIBS setting.
518 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
519 errors. Otherwise a message that provokes a temporary error (when other
520 messages do not) can cause a whole host to time out.
522 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
523 addresses were routed directly to files or pipes from a redirect router.
524 File deliveries just didn't batch; pipe deliveries might have suffered
527 PH/06 A failure to get a lock for a hints database would erroneously always say
528 "Failed to get write lock", even when it was really a read lock.
530 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
531 of 0600. This has been changed to use the value of the lockfile_mode
532 option (which defaults to 0600).
534 PH/08 Applied small patch from the Sieve maintainer.
536 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
537 folder from quota calculations, a direct delivery into this folder messed
538 up the contents of the maildirsize file. This was because the regex was
539 used only to exclude .Trash (or whatever) when the size of the mailbox
540 was calculated. There was no check that a delivery was happening into an
541 excluded directory. This bug has been fixed by ignoring all quota
542 processing for deliveries into excluded directories.
544 PH/10 Added the maildirfolder_create_regex option to appendfile.
550 PH/01 The code for finding all the local interface addresses on a FreeBSD
551 system running IPv6 was broken. This may well have applied to all BSD
552 systems, as well as to others that have similar system calls. The broken
553 code found IPv4 interfaces correctly, but gave incorrect values for the
554 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
555 that it would not match correctly against @[] and not recognize the IPv6
558 PH/02 The ipliteral router was not recognizing addresses of the form user@
559 [ipv6:....] because it didn't know about the "ipv6:" prefix.
561 PH/03 Added disable_ipv6.
563 PH/04 Changed $reply_address to use the raw form of the headers instead of the
564 decoded form, because it is most often used to construct To: headers
565 lines in autoreplies, and the decoded form may well be syntactically
566 invalid. However, $reply_address has leading white space removed, and all
567 newlines turned into spaces so that the autoreply transport does not
570 PH/05 If group was specified without a user on a router, and no group or user
571 was specified on a transport, the group from the router was ignored.
573 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
574 for visible compile-time settings that can be used to change these
575 numbers, for those that want even more. Backwards compatibility with old
576 spool files has been maintained. However, going back to a previous Exim
577 release will lost any variables that are in spool files.
579 PH/07 Two small changes when running in the test harness: increase delay when
580 passing a TCP/IP connection to a new process, in case the original
581 process has to generate a bounce, and remove special handling of
582 127.0.0.2 (sic), which is no longer necessary.
584 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
585 be the same on different OS.
587 PH/09 Moved a debug statement in filter processing to avoid a race problem when
590 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
591 whether --show-vars was specified or not
593 JJ/02 exipick: Added support for new ACL variable spool format introduced
596 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
597 syntactically invalid From: or Reply-to: line, and a filter used this to
598 generate an autoreply, and therefore failed to obtain an address for the
599 autoreply, Exim could try to deliver to a non-existent relative file
600 name, causing unrelated and misleading errors. What now happens is that
601 it logs this as a hard delivery error, but does not attempt to create a
604 PH/11 The exinext utility has a -C option for testing purposes, but although
605 the given file was scanned by exinext itself; it wasn't being passed on
606 when Exim was called.
608 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
609 an end-of-file indication when reading a command response.
611 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
612 compiled. In many other places in Exim, IPv6 addresses are always
613 recognized, so I have changed this. It also means that IPv4 domain
614 literals of the form [IPV4:n.n.n.n] are now always recognized.
616 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
617 used if the router is not running as root, for example, when verifying at
618 ACL time, or when using -bh. The debugging output from this situation was
619 non-existent - all you got was a failure to exec. I have made two
622 (a) Failures to set uid/gid, the current directory, or a process leader
623 in a subprocess such as that created by queryprogram now generate
624 suitable debugging ouput when -d is set.
626 (b) The queryprogram router detects when it is not running as root,
627 outputs suitable debugging information if -d is set, and then runs
628 the subprocess without attempting to change uid/gid.
630 PH/15 Minor change to Makefile for building test_host (undocumented testing
633 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
634 additional section of a DNS packet that returns MX or SRV records.
635 Instead, it always explicitly searches for A/AAAA records. This avoids
636 major problems that occur when a DNS server includes only records of one
637 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
638 fixed another bug: if SRV records were looked up and the corresponding
639 address records were *not* found in the additional section, the port
640 values from the SRV records were lost.
642 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
643 using the correct key (the original address) when searching the retry
644 rules in order to find which one to use for generating the retry hint.
646 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
647 adding the default one. Similarly, if it contains a Reply-To: header, the
648 errors_reply_to option, if set, is not used.
650 PH/19 When calculating a retry time, Exim used to measure the "time since
651 failure" by looking at the "first failed" field in the retry record. Now
652 it does not use this if it is later than than the arrival time of the
653 message. Instead it uses the arrival time. This makes for better
654 behaviour in cases where some deliveries succeed, thus re-setting the
655 "first failed" field. An example is a quota failure for a huge message
656 when small messages continue to be delivered. Without this change, the
657 "time since failure" will always be short, possible causing more frequent
658 delivery attempts for the huge message than are intended.
659 [Note: This change was subsequently modified - see PH/04 for 4.62.]
661 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
662 $1, $2, $3) because the numerical variables can be reset during some
663 expansion items (e.g. "match"), thereby losing the authentication data.
665 PH/21 Make -bV show the size of off_t variables so that the test suite can
666 decide whether to run tests for quotas > 2G.
668 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
669 mailbox_size, and mailbox_filecount in the appendfile transport. If a
670 filecount value is greater than 2G or if a quota value is greater than 2G
671 on a system where the size of off_t is not greater than 4, a panic error
674 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
675 never match. The debug and -bh output now contains an explicit error
676 message indicating a malformed IPv4 address or mask.
678 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
679 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
682 PH/25 Do not write to syslog when running in the test harness. The only
683 occasion when this arises is a failure to open the main or panic logs
684 (for which there is an explicit test).
686 PH/26 Added the /no_tell option to "control=freeze".
688 PH/27 If a host name lookup failed very early in a connection, for example, if
689 the IP address matched host_lookup and the reverse lookup yielded a name
690 that did not have a forward lookup, an error message of the form "no IP
691 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
692 could be logged. Now it outputs the IP address instead of "NULL".
694 PH/28 An enabling patch from MH: add new function child_open_exim2() which
695 allows the sender and the authenticated sender to be set when
696 submitting a message from within Exim. Since child_open_exim() is
697 documented for local_scan(), the new function should be too.
699 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
700 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
701 results in an empty string is now treated as unset.
703 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
705 PH/31 Added sender_verify_fail logging option.
707 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
708 needed by RFC 822 but not by RFC 2822 was commented out. I have now
709 tidied the source and removed it altogether.
711 PH/33 When a queue run was abandoned because the load average was too high, a
712 log line was always written; now it is written only if the queue_run log
713 selector is set. In addition, the log line for abandonment now contains
714 information about the queue run such as the pid. This is always present
715 in "start" and "stop" lines but was omitted from the "abandon" line.
717 PH/34 Omit spaces between a header name and the colon in the error message that
718 is given when verify = headers_syntax fails (if there are lots of them,
719 the message gets confusing).
721 PH/35 Change the default for dns_check_names_pattern to allow slashes within
722 names, as there are now some PTR records that contain slashes. This check
723 is only to protect against broken name servers that fall over on strange
724 characters, so the fact that it applies to all lookups doesn't matter.
726 PH/36 Now that the new test suite is complete, we can remove some of the
727 special code in Exim that was needed for the old test suite. For example,
728 sorting DNS records because real resolvers return them in an arbitrary
729 order. The new test suite's fake resolver always returns records in the
732 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
733 bounces) except when queue_only is set, to avoid logging races between
734 the different processes.
736 PH/38 Panic-die if .include specifies a non-absolute path.
738 PH/39 A tweak to the "H" retry rule from its user.
740 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
741 a label. They prevented compilation on older perls.
743 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
744 a warning to be raised on newish perls.
746 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
747 on queue. Changes to match documented behaviour of showing count of
748 messages matching specified criteria.
750 PH/40 Changed the default ident timeout from 30s to 5s.
752 PH/41 Added support for the use of login_cap features, on those BSD systems
753 that have them, for controlling the resources used by pipe deliveries.
755 PH/42 The content-scanning code uses fopen() to create files in which to put
756 message data. Previously it was not paying any attention to the mode of
757 the files. Exim runs with umask(0) because the rest of the code creates
758 files with open(), and sets the required mode explicitly. Thus, these
759 files were ending up world-writeable. This was not a big issue, because,
760 being within the spool directory, they were not world-accessible. I have
761 created a function called modefopen, which takes an additional mode
762 argument. It sets umask(777), creates the file, chmods it to the required
763 mode, then resets the umask. All the relevant calls to fopen() in the
764 content scanning code have been changed to use this function.
766 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
767 to 24 hours. This avoids potential overflow problems when processing G
768 and H retry rules. I suspect nobody ever tinkers with this value.
770 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
772 PH/45 When the plaintext authenticator is running as a client, the server's
773 challenges are checked to ensure they are valid base64 strings. By
774 default, the authentication attempt is cancelled if an invalid string is
775 received. Setting client_ignore_invalid_base64 true ignores these errors.
776 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
777 they are received. Thus, the responses can be made to depend on the
778 challenges. If an invalid string is ignored, an empty string is placed in
781 PH/46 Messages that are created by the autoreply transport now contains a
782 References: header, in accordance with RFCs 2822 and 3834.
784 PH/47 Added authenticated_sender_force to the smtp transport.
786 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
788 PH/49 Installed latest patch from the Sieve maintainer.
790 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
791 was also set, the appendfile transport was unnecessarily scanning a
792 directory of message files (e.g. for maildir delivery) to find the count
793 of files (along with the size), even though it did not need this
794 information. It now does the scan only if it needs to find either the
795 size of the count of files.
797 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
799 PH/52 Two bugs concerned with error handling when the smtp transport is
802 (i) Exim was not creating retry information for temporary errors given
803 for individual recipients after the DATA command when the smtp transport
804 was used in LMTP mode. This meant that they could be retried too
805 frequently, and not timed out correctly.
807 (ii) Exim was setting the flag that allows error details to be returned
808 for LMTP errors on RCPT commands, but not for LMTP errors for individual
809 recipients that were returned after the DATA command.
811 PH/53 This is related to PH/52, but is more general: for any failing address,
812 when detailed error information was permitted to be returned to the
813 sender, but the error was temporary, then after the final timeout, only
814 "retry timeout exceeded" was returned. Now it returns the full error as
815 well as "retry timeout exceeded".
817 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
818 do this, and (what is worse) MTAs that accept it.
820 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
821 will now be deprecated.
823 PH/56 New os.c-cygwin from the Cygwin maintainer.
825 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
826 formats (previously only available in exim formats via -bpr, -bpru,
827 and -bpra. Now also available in native and exiqgrep formats)
829 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
830 with very large, slow to parse queues
832 JJ/08 exipick: added ! as generic prefix to negate any criteria format
834 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
836 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
837 responses to authentication challenges, though it was showing the
838 challenges; (ii) I've removed the CR characters from the debug output for
841 PH/58 Allow for the insertion of a newline as well as a space when a string
842 is turned into more than one encoded-word during RFC 2047 encoding. The
843 Sieve code now uses this.
845 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
846 data_4xx, lost_connection, tls_required.
848 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
849 message was being sent as an SMTP response.
851 PH/61 Add -l and -k options to exicyclog.
853 PH/62 When verifying, if an address was redirected to one new address, so that
854 verification continued, and the new address failed or deferred after
855 having set something in $address_data, the value of $address_data was not
856 passed back to the ACL. This was different to the case when no
857 redirection occurred. The value is now passed back in both cases.
859 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
860 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
861 use login_cap.h, so on its own it isn't the distinguishing feature. The
862 new name refers directly to the setclassresources() function.
864 PH/65 Added configuration files for NetBSD3.
866 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
868 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
869 is preferred over IPv4.
871 PH/68 The bounce_return_message and bounce_return_body options were not being
872 honoured for bounces generated during the reception of non-SMTP messages.
873 In particular, this applied to messages rejected by the ACL. This bug has
874 been fixed. However, if bounce_return_message is true and bounce_return_
875 body is false, the headers that are returned for a non-SMTP message
876 include only those that have been read before the error was detected.
877 (In the case of an ACL rejection, they have all been read.)
879 PH/69 The HTML version of the specification is now built in a directory called
880 spec_html instead of spec.html, because the latter looks like a path with
881 a MIME-type, and this confuses some software.
883 PH/70 Catch two compiler warnings in sieve.c.
885 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
886 function verify_get_ident() calls ip_connect() to connect a socket, but
887 if the "connect()" function timed out, ip_connect() used to close the
888 socket. However, verify_get_ident() also closes the socket later, and in
889 between Exim writes to the log, which may get opened at this point. When
890 the socket was closed in ip_connect(), the log could get the same file
891 descriptor number as the socket. This naturally causes chaos. The fix is
892 not to close the socket in ip_connect(); the socket should be closed by
893 the function that creates it. There was only one place in the code where
894 this was missing, in the iplookup router, which I don't think anybody now
895 uses, but I've fixed it anyway.
897 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
898 well as to direct DNS lookups. Otherwise the handling of names in host
899 lists is inconsistent and therefore confusing.
905 PH/01 Two changes to the default runtime configuration:
907 (1) Move the checks for relay_from_hosts and authenticated clients from
908 after to before the (commented out) DNS black list checks.
910 (2) Add control=submission to the relay_from_hosts and authenticated
911 clients checks, on the grounds that messages accepted by these
912 statements are most likely to be submissions.
914 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
916 (1) Generate an error if the third argument for the ${prvs expansion is
919 (2) Treat a missing third argument of ${prvscheck as if it were an empty
922 (3) Reset the variables that are obtained from the first argument of
923 ${prvscheck and used in the second argument before leaving the code,
924 because their memory is reclaimed, so using them afterwards may do
927 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
928 one (it's much easier than Tom thought :-).
930 (5) Because of (4), we can now allow for the use of $prvscheck_result
931 inside the third argument.
933 PH/03 For some reason, the default setting of PATH when running a command from
934 a pipe transport was just "/usr/bin". I have changed it to
937 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
938 anything to be listed in the output from -bV.
940 PH/05 When a filter generated an autoreply, the entire To: header line was
941 quoted in the delivery log line, like this:
943 => >A.N.Other <ano@some.domain> <original@ddress> ...
945 This has been changed so that it extracts the operative address. There
946 may be more than one such address. If so, they are comma-separated, like
949 => >ano@some.domain,ona@other.domain <original@ddress> ...
951 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
952 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
953 not being looked up in the rDNS to get a host name, Exim was showing the
954 IP address twice in Received: lines, even though the IP addresses were
955 identical. For example:
957 Received: from [1.2.3.4] (helo=[1.2.3.4])
959 However, if the real host name was known, it was omitting the HELO data
960 if it matched the actual IP address. This has been tidied up so that it
961 doesn't show the same IP address twice.
963 PH/07 When both +timestamp and +memory debugging was on, the value given by
964 $tod_xxx expansions could be wrong, because the tod_stamp() function was
965 called by the debug printing, thereby overwriting the timestamp buffer.
966 Debugging no longer uses the tod_stamp() function when +timestamp is set.
968 PH/08 When the original message was included in an autoreply transport, it
969 always said "this is a copy of the message, including all the headers",
970 even if body_only or headers_only was set. It now gives an appropriate
973 PH/09 Applied a patch from the Sieve maintainer which:
975 o fixes some comments
976 o adds the (disabled) notify extension core
977 o adds some debug output for the result of if/elsif tests
978 o points to the current vacation draft in the documentation
979 and documents the missing references header update
983 o fixes a bug in processing the envelope test (when testing
984 multiple envelope elements, the last element determinted the
987 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
988 Electronic Mail") by including:
990 Auto-submitted: auto-generated
992 in the messages that it generates (bounce messages and others, such as
993 warnings). In the case of bounce messages for non-SMTP mesages, there was
994 also a typo: it was using "Auto_submitted" (underscore instead of
995 hyphen). Since every message generated by Exim is necessarily in response
996 to another message, thes have all been changed to:
998 Auto-Submitted: auto-replied
1000 in accordance with these statements in the RFC:
1002 The auto-replied keyword:
1004 - SHOULD be used on messages sent in direct response to another
1005 message by an automatic process,
1007 - MUST NOT be used on manually-generated messages,
1009 - MAY be used on Delivery Status Notifications (DSNs) and Message
1010 Disposition Notifications (MDNs),
1012 - MUST NOT be used on messages generated by automatic or periodic
1013 processes, except for messages which are automatic responses to
1016 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1017 to the default Received: header definition.
1019 PH/12 Added log selector acl_warn_skipped (default on).
1021 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1022 variables because (a) they are in the wrong storage pool and (b) even if
1023 they were copied, it wouldn't work properly because of the caching.
1025 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1026 checking when decoding. Apparently there are clients that generate
1027 overlong encoded strings. Why am I not surprised?
1029 PH/15 If the first argument of "${if match_address" was not empty, but did not
1030 contain an "@" character, Exim crashed. Now it writes a panic log message
1031 and treats the condition as false.
1033 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1035 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1036 extension "envelope-auth". The code is finished and in agreement with
1037 other implementations, but there is no documentation so far and in fact,
1038 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1039 not changing the active code.
1041 Print executed "if" and "elsif" statements when debugging is used. This
1042 helps a great deal to understand what a filter does.
1044 Document more things not specified clearly in RFC3028. I had all this
1045 sorted out, when out of a sudden new issues came to my mind. Oops."
1047 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1050 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1051 canonical form (as documented). However, after a host name lookup from
1052 the IP address, check_host() was doing a simple string comparison with
1053 addresses acquired from the DNS when checking that the found name did
1054 have the original IP as one of its addresses. Since any found IPv6
1055 addresses are likely to be in abbreviated form, the comparison could
1056 fail. Luckily, there already exists a function for doing the comparison
1057 by converting both addresses to binary, so now that is used instead of
1058 the text comparison.
1060 PH/20 There was another similar case to PH/19, when a complete host name was
1061 given in a host list; looking up its IP address could give an abbreviated
1062 form, whereas the current host's name might or might not be abbreviated.
1063 The same fix has been applied.
1069 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1070 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1073 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1074 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1076 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1078 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1079 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1080 Also, for "Auto-Submitted", treat anything other than "no" as
1081 non-personal, in accordance with RFC 3834. (Previously it treated
1082 anything starting "auto-" as non-personal.)
1084 TF/01 The control=submission/name=... option had a problem with syntax
1085 errors if the name included a slash character. The /name= option
1086 now slurps the rest of the string, so it can include any characters
1087 but it must come last in the list of options (after /sender_retain
1090 PH/05 Some modifications to the interface to the fake nameserver for the new
1098 TK/01 Added the "success_on_redirect" address verification option. See
1099 NewStuff for rationale and an example.
1101 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1103 PH/02 Patch to exigrep to allow it to work on syslog lines.
1105 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1106 fread() to skip over the body file's header line, because in Cygwin the
1107 header line is locked and is inaccessible.
1109 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1110 co-exist for some time) to make it clear that it is the Exim ID that is
1111 referenced, not the Message-ID: header line.
1113 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1114 string_format() function, because snprintf() does not exist on all
1117 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1118 ${stat: expansion item.
1120 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1121 protocol synchronization error", to keep the pedants happy.
1123 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1124 well as for IRIX systems, when gcc is being used. See the host.c source
1127 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1129 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1132 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1133 transport and to the smtp transport in LMTP mode.
1135 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1137 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1138 run a filter in a subprocess. This could lead to confusion in subsequent
1139 lookups in the parent process. There should also be a search_tidyup() at
1140 the end of the subprocess.
1142 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1143 only if the host matched helo_try_verify_hosts, which caused the
1144 verification to occur when the EHLO/HELO command was issued. The ACL just
1145 tested the remembered result. Now, if a previous verification attempt has
1146 not happened, "verify = helo" does it there and then.
1148 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1150 TK/03 Fix log output including CR from clamd.
1152 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1153 exist provoked a memory error which could cause a segfault.
1155 PH/15 Installed PCRE 6.2
1157 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1159 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1160 of the problem. Specifically, suggested +O2 rather than +O1 for the
1163 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1165 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1166 record, the port was not correctly logged when the outgoing_port log
1167 selector was set (it logged the transort's default port).
1169 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1170 fallback_hosts, and "hosts" in the smtp transport.
1172 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1173 host errors such as "Connection refused".
1175 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1176 authentication with radiusclient 0.4.9:
1178 - Error returned from rc_read_config was caught wrongly
1179 - Username/password not passed on to radius server due to wrong length.
1181 The presumption is that some radiusclient API changes for 4.51/PH/17
1182 were not taken care of correctly. The code is still untested by me (my
1183 Linux distribution still has 0.3.2 of radiusclient), but it was
1184 contributed by a Radius user.
1186 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1187 expanding the "port" option of the smtp transport.
1189 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1190 while reading a MIME header. Thanks to Tom Hughes for a patch.
1192 PH/24 Include config.h inside local_scan.h so that configuration settings are
1195 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1196 that in an ACL for RCPT (for example), you can examine exactly what was
1199 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1200 commands, but it was not correctly comparing the address with the actual
1201 client host address. Thus, it would show the EHLO address in Received:
1202 header lines when this was not necessary.
1204 PH/27 Added the % operator to ${eval:}.
1206 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1207 it should be ignoring failures (because with -C, for example, it has lost
1208 privilege). It wasn't ignoring creation failures other than "already
1211 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1214 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1215 because an input line was too long, either on its own, or by virtue of
1216 too many continuations, the temporary file was not being removed, and the
1217 return code was incorrect.
1219 PH/31 Missing "BOOL" in function definition in filtertest.c.
1221 PH/32 Applied Sieve patches from the maintainer.
1223 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1225 PH/33 Added "verify = not_blind".
1227 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1228 Local/Makefile (with some defaults set). These are used in built scripts
1229 such as exicyclog, but they have never been used in the exim_install
1230 script (though there are many overriding facilities there). I have
1231 arranged that the exim_install script now takes note of these two
1234 PH/35 Installed configuration files for Dragonfly.
1236 PH/36 When a locally submitted message by a trusted user did not contain a
1237 From: header, and the sender address was obtained from -f or from an SMTP
1238 MAIL command, and the trusted user did not use -F to supply a sender
1239 name, $originator_name was incorrectly used when constructing a From:
1240 header. Furthermore, $originator_name was used for submission mode
1241 messages from external hosts without From: headers in a similar way,
1242 which is clearly wrong.
1244 PH/37 Added control=suppress_local_fixups.
1246 PH/38 When log_selector = +received_sender was set, and the addition of the
1247 sender made the log line's construction buffer exactly full, or one byte
1248 less than full, an overflow happened when the terminating "\n" was
1251 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1252 when the result of a list match is failure because a DNS lookup failed.
1254 PH/40 RM_COMMAND is now used in the building process.
1256 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1257 the "build-* directories that it finds.
1259 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1260 address in a domain literal was a prefix of an interface address.
1262 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1263 when verifying a sender address, unless rewrite_headers is false.
1265 PH/44 Wrote a long comment about why errors_to addresses are verified as
1266 recipients, not senders.
1268 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1269 the ratelimit ACL was added.
1271 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1273 PH/46 Added extra information about PostgreSQL errors to the error string.
1275 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1276 avoiding the need to install special zones in a real server. This is
1277 backwards compatible; if it can't find the fake resolver, it drops back.
1278 Thus, both old and new test suites can be run.
1280 TF/02 Added util/ratelimit.pl
1282 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1283 clock is set back in time.
1285 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1286 Candler <B.Candler@pobox.com>.
1288 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1289 OK for addresses that are the result of redirections.
1291 PH/48 A number of further additions for the benefit of the new test suite,
1292 including a fake gethostbyname() that interfaces to the fake DNS resolver
1295 TF/06 The fix for widen_domains has also been applied to qualify_single and
1296 search_parents which are the other dnslookup options that can cause
1299 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1302 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1304 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1305 service" if it encountered an MX record with an empty target hostname.
1306 The message is now "an MX or SRV record indicated no SMTP service".
1308 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1309 if the DNS of the sending site is misconfigured. This is quite a
1310 common situation. This change restores the behaviour of treating a
1311 helo verification defer as a failure.
1313 PH/51 If self=fail was set on a router, the bounce message did not include the
1314 actual error message.
1320 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1322 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1323 command itself ended in error, the underlying message about the transport
1324 filter timeout was being overwritten with the pipe command error. Now the
1325 underlying error message should be appended to the second error message.
1327 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1329 PH/02 Reduce the amount of output that "make" produces by default. Full output
1330 can still be requested.
1332 PH/03 The warning log line about a condition test deferring for a "warn" verb
1333 was being output only once per connection, rather than after each
1334 occurrence (because it was using the same function as for successful
1335 "warn" verbs). This seems wrong, so I have changed it.
1337 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1338 it should not have, which might have caused a crash in the right
1339 circumstances, but probably never did.
1341 PH/04 Installed a modified version of Tony Finch's patch to make submission
1342 mode fix the return path as well as the Sender: header line, and to
1343 add a /name= option so that you can make the user's friendly name appear
1346 TF/03 Added the control = fakedefer ACL modifier.
1348 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1349 Mark Lowes for thorough testing.
1351 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1353 TK/03 Merged latest SRS patch from Miles Wilton.
1355 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1356 with the definition in sysexits.h (which is #included earlier).
1357 Fortunately, Exim does not actually use EX_OK. The code used to try to
1358 preserve the sysexits.h value, by assumimg that macro definitions were
1359 scanned for macro replacements. I have been disabused of this notion,
1360 so now the code just undefines EX_OK before #including unistd.h.
1362 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1363 in the smtp transport. When a block could not be written in a single
1364 write() function, the timeout was being re-applied to each part-write.
1365 This seems wrong - if the receiver was accepting one byte at a time it
1366 would take for ever. The timeout is now adjusted when this happens. It
1367 doesn't have to be particularly precise.
1369 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1370 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1372 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1373 without a domain if the check to <postmaster@domain> fails.
1375 SC/01 Eximstats: added -xls and the ability to specify output files
1376 (patch written by Frank Heydlauf).
1378 SC/02 Eximstats: use FileHandles for outputing results.
1380 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1382 SC/04 Eximstats: fixed display of large numbers with -nvr option
1384 SC/05 Eximstats: fixed merging of reports with empty tables.
1386 SC/06 Eximstats: added the -include_original_destination flag
1388 SC/07 Eximstats: removed tabs and trailing whitespace.
1390 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1392 TK/06 MBOX spool code: Add real "From " MBOX separator line
1393 so the .eml file is really in mbox format (even though
1394 most programs do not really care). Patch from Alex Miller.
1396 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1397 The latter is generated from $received_to and is only set if the
1398 message has one envelope recipient. SA can use these headers,
1399 obviously out-of-the-box. Patch from Alex Miller.
1401 PH/08 The ${def test on a variable was returning false if the variable's
1402 value was "0", contrary to what the specification has always said!
1403 The result should be true unless the variable is empty.
1405 PH/09 The syntax error of a character other than { following "${if
1406 def:variable_name" (after optional whitespace) was not being diagnosed.
1407 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1408 accidental colon was present, for example, could give incorrect results.
1410 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1411 result is used (not including appendfile, where other changes are about
1414 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1415 This involved changing a lot of size variables from int to off_t. It
1416 should work with maildirs and everything.
1418 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1419 spamd dying while we are connected to it.
1421 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1424 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1425 function for BDB 4.3.
1427 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1429 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1430 the sources for PCRE have been re-arranged and more files are now
1433 PH/15 The code I had for printing potentially long long variables in PH/11
1434 above was not the best (it lost precision). The length of off_t variables
1435 is now inspected at build time, and an appropriate printing format (%ld
1436 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1437 to be "long long int" or "long int". This is needed for the internal
1438 formatting function string_vformat().
1440 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1441 the configuration file to be ":syslog", then the script "guesses" where
1442 the logs files are, rather than using the compiled in default. In our
1443 case the guess is not the same as the compiled default, so the script
1444 suddenly stopped working when I started to use syslog. The patch checks
1445 to see if log_file_path is "". If so, it attempts to read it from exim
1446 with no configuration file to get the compiled in version, before it
1447 falls back to the previous guessing code."
1449 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1450 implementing BATV in an Exim configuration. See NewStuff for the gory
1453 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1454 Makefile that are specific to HP-UX.
1456 PH/18 If the "use_postmaster" option was set for a recipient callout together
1457 with the "random" option, the postmaster address was used as the MAIL
1458 FROM address for the random test, but not for the subsequent recipient
1459 test. It is now used for both.
1461 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1462 patch removes a few documentation additions to RFC 3028, because the
1463 latest draft now contains them. It adds the new en;ascii-case comparator
1464 and a new error check for 8bit text in MIME parts. Comparator and
1465 require names are now matched exactly. I enabled the subaddress
1466 extension, but it is not well tested yet (read: it works for me)."
1468 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1469 rework some of the code of TK/09 above to avoid the hardwired use of
1470 "%lld" and "long long". Replaced the call to snprintf() with a call to
1473 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1474 records point to non-existent hosts", "retry timeout exceeded", and
1475 "retry time not reached for any host after a long failure period".
1477 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1478 experimental DomainKeys support:
1480 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1481 (2) On an error such as an illegally used "control", the wrong name for
1482 the control was given.
1484 These problems did NOT occur unless DomainKeys support was compiled.
1486 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1488 PH/24 Added ${if match_ip condition.
1490 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1491 databases so that it will be absolutely obvious if a crash occurs in the
1492 DB library. This is a regular occurrence (often caused by mis-matched
1495 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1496 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1497 picked up on a user's system that detects such things. There doesn't seem
1498 to be a gcc warning option for this - only an attribute that has to be
1499 put on the function's prototype. It seems that in Fedora Core 4 they have
1500 set this on a number of new functions. No doubt there will be more in due
1503 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1504 specify a transport. However, if an address that was verified by such a
1505 router was the subject of a callout, Exim crashed because it tried to
1506 read the rcpt_include_affixes from the non-existent transport. Now it
1507 just assumes that the setting of that option is false. This bug was
1508 introduced by 4.51/PH/31.
1510 PH/28 Changed -d+all to exclude +memory, because that information is very
1511 rarely of interest, but it makes the output a lot bigger. People tend to
1512 do -d+all out of habit.
1514 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1515 code in os-type was giving problems when libc.so lives in lib64, like on
1518 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1519 aren't the modern standard, and it seems that some systems' include files
1520 don't always have them. Exim was already checking for some of the newer
1521 ones like T_AAAA, and defining it itself. I've added checks for all the
1522 record types that Exim uses.
1524 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1525 not automatically generating a new one, as it is supposed to. This
1526 prevented TLS from working. If the file did exist, but contained invalid
1527 data, a new version was generated, as expected. It was only the case of a
1528 non-existent file that was broken.
1530 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1531 with a change in libdomainkeys > 0.64.
1533 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1534 from DNS. If the selector record carries the flag, it now has
1535 precedence over the domain-wide flag.
1537 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1539 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1540 the use of a port name that isn't defined in /etc/services) occurred, the
1541 message was deferred as in a normal delivery, and thus remained on the
1542 spool, instead of being failed because of the mua_wrapper setting. This
1543 is now fixed, and I tidied up some of the mua_wrapper messages at the
1546 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1547 the messages in a hash of arrays rather than using individual hashes.
1548 This is a bit cleaner and results in dramatic memory savings, albeit
1549 at a slight CPU cost.
1551 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1552 as requested by Marc Sherman.
1554 SC/10 Eximstats: added histograms for user specified patterns as requested
1557 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1559 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1560 fopen() in the content-scanning modules that did not already have it.
1566 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1567 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1569 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1571 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1573 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1574 to test Sieve filters that use "vacation".
1576 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1577 that changes the way the GnuTLS parameters are stored in the cache file.
1578 The new format can be generated externally. For backward compatibility,
1579 if the data in the cache doesn't make sense, Exim assumes it has read an
1580 old-format file, and it generates new data and writes a new file. This
1581 means that you can't go back to an older release without removing the
1584 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1585 work if there are any delivery delays because "one_time" forces the
1586 parent to be marked "delivered", so its unseen clone is never tried
1587 again. For this reason, Exim now forbids the simultaneous setting of
1590 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1591 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1592 entry if you want to know the details. The fix, however, neglected to
1593 consider the case when local delivery batching is involved. The test for
1594 "previously delivered" was not happening when checking to see if an
1595 address could be batched with a previous (undelivered) one; under
1596 certain circumstances this could lead to multiple deliveries to the same
1599 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1600 in its include files, and this causes problems building Exim.
1602 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1603 header_syntax) but Exim was just ignoring anything given after a slash.
1604 In particular, this caused confusion with an attempt to use "verify =
1605 reverse_host_lookup/defer_ok". An error is now given when options are
1606 supplied for verify items that do not have them. (Maybe reverse_host_
1607 lookup should have a defer_ok option, but that's a different point.)
1609 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1610 defined by RFC 821) to 2048, because there were problems with some AUTH
1611 commands, and RFC 1869 says the size should be increased for extended
1612 SMTP commands that take arguments.
1614 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1617 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1618 "unknown" error; now it says that the functionality isn't in the binary.
1620 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1621 an address' error message when a string expansion fails (syntax or
1622 whatever). Otherwise the password may appear in the log. Following change
1623 PH/42 below, there is no longer a chance of it appearing in a bounce
1626 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1628 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1629 of its addresses was ever tried. (Bugzilla bug #2.)
1631 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1632 the result incorrectly in the debug output. (It correctly added a newline
1633 to what was transported.)
1635 TF/01 Added $received_time.
1637 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1638 commented out examples of how to interface to a virus scanner and to
1639 SpamAssassin. Also added commented examples of av_scanner and
1640 spamd_address settings.
1642 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1643 and controls are allowed in which ACLs. There were a couple of minor
1644 errors. Some of the entries in the conditions table (which is a table of
1645 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1646 negation of where the condition IS allowed.
1648 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1650 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1651 header file does not have a version number, so I've had to invent a new
1652 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1653 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1654 radiusclient), but it was contributed by a Radius user.
1656 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1657 files or directories, for OpenSSL.
1659 PH/19 When an Exim process that is running as root has to create an Exim log
1660 file, it does so in a subprocess that runs as exim:exim so as to get the
1661 ownership right at creation (otherwise, other Exim processes might see
1662 the file with the wrong ownership). There was no test for failure of this
1663 fork() call, which would lead to the process getting stuck as it waited
1664 for a non-existent subprocess. Forks do occasionally fail when resources
1665 run out. I reviewed all the other calls to fork(); they all seem to check
1668 PH/20 When checking for unexpected SMTP input at connect time (before writing
1669 the banner), Exim was not dealing correctly with a non-positive return
1670 from the read() function. If the client had disconnected by this time,
1671 the result was a log entry for a synchronization error with an empty
1672 string after "input=" when read() returned zero. If read() returned -1
1673 (an event I could not check), uninitialized data bytes were printed.
1674 There were reports of junk text (parts of files, etc) appearing after
1677 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1679 PH/22 Added support for macro redefinition, and (re)definition in between
1680 driver and ACL definitions.
1682 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1683 forgetting to use the resulting value; it was using the unexpanded value.
1685 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1686 hadn't been configured. The fix is from Juergen Kreileder, who
1687 understands it better than I do:
1689 "Here's what I see happening with three configured cyrus_sasl
1690 authenticators configured (plain, login, cram-md5):
1692 On startup auth_cyrus_sasl_init() gets called for each of these.
1693 This means three calls to sasl_listmech() without a specified mech_list.
1694 => SASL tests which mechs of all available mechs actually work
1695 => three warnings about OTP not working
1696 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1698 With the patch, sasl_listmech() also gets called three times. But now
1699 SASL's mech_list option is set to the server_mech specified in the the
1700 authenticator. Or in other words, the answer from sasl_listmech()
1701 gets limited to just the mech you're testing for (which is different
1703 => the return list contains just 'plain' or 'login', 'cram-md5' or
1704 nothing depending on the value of ob->server_mech.
1706 I've just tested the patch: Authentication still works fine,
1707 unavailable mechs specified in the exim configuration are still
1708 caught, and the auth.log warnings about OTP are gone."
1710 PH/25 When debugging is enabled, the contents of the command line are added
1711 to the debugging output, even when log_selector=+arguments is not
1714 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1715 answer is "GNU", and only if the return is "GNU/something" is the answer
1718 PH/27 $acl_verify_message is now set immediately after the failure of a
1719 verification in an ACL, and so is available in subsequent modifiers. In
1720 particular, the message can be preserved by coding like this:
1722 warn !verify = sender
1723 set acl_m0 = $acl_verify_message
1725 Previously, $acl_verify_message was set only while expanding "message"
1726 and "log_message" when a very denied access.
1728 PH/28 Modified OS/os.c-Linux with
1730 -#ifndef OS_LOAD_AVERAGE
1731 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1733 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1734 nomenclature these days.)
1736 PH/29 Installed patch from the Sieve maintainer that adds the options
1737 sieve_useraddress and sieve_subaddress to the redirect router.
1739 PH/30 In these circumstances:
1740 . Two addresses routed to the same list of hosts;
1741 . First host does not offer TLS;
1742 . First host accepts first address;
1743 . First host gives temporary error to second address;
1744 . Second host offers TLS and a TLS session is established;
1745 . Second host accepts second address.
1746 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1747 and peerdn, if requested) that were in fact used only for the second
1750 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1751 attention to any local part prefix or suffix that was matched by the
1752 router that accepted the address. It now behaves in the same way as it
1753 does for delivery: the affixes are removed from the local part unless
1754 rcpt_include_affixes is set on the transport.
1756 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1757 timeout during the DATA phase of an incoming message.
1759 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1760 applied a patch sanctioned by the Sieve maintainer.
1762 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1763 the uid or gid is negative. A case of a negative gid caused this to be
1764 noticed. The fix allows for either to be negative.
1766 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1767 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1770 PH/36 The change PH/12 above was broken. Fixed it.
1772 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1773 the grounds that routing the same address twice would always produce the
1774 same answer. This might have been true once, but it is certainly no
1775 longer true now. Routing a child address may depend on the previous
1776 routing that produced that child. Some complicated redirection strategies
1777 went wrong when messages had multiple recipients, and made Exim's
1778 behaviour dependent on the order in which the addresses were given.
1780 I have moved the duplicate checking until after the routing is complete.
1781 Exim scans the addresses that are assigned to local and remote
1782 transports, and removes any duplicates. This means that more work will be
1783 done, as duplicates will always all be routed, but duplicates are
1784 presumably rare, so I don't expect this is of any significance.
1786 For deliveries to pipes, files, and autoreplies, the duplicate checking
1787 still happens during the routing process, since they are not going to be
1790 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1791 It corrects a timeout issue with spamd. This is Ian's comment: "The
1792 background is that sometimes spamd either never reads data from a
1793 connection it has accepted, or it never writes response data. The exiscan
1794 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1795 blindly assumes that writes won't block so it may never time out."
1797 PH/39 Allow G after quota size as well as K and M.
1799 PH/40 The value set for $authenticated_id in an authenticator may not contain
1800 binary zeroes or newlines because the value is written to log lines and
1801 to spool files. There was no check on this. Now the value is run through
1802 the string_printing() function so that such characters are converted to
1803 printable escape sequences.
1805 PH/41 $message_linecount is a new variable that contains the total number of
1806 lines in the message. Compare $body_linecount, which is the count for the
1809 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1810 bounce and delay warning messages, except in certain special cases, which
1813 (a) An SMTP error message from a remote host;
1814 (b) A message specified in a :fail: redirection;
1815 (c) A message specified in a "fail" command in a system filter;
1816 (d) A message specified in a FAIL return from the queryprogram router;
1817 (e) A message specified by the cannot_route_message router option.
1819 In these cases only, Exim does include the error details in bounce and
1820 warning messages. There are also a few cases where bland messages such
1821 as "unrouteable address" or "local delivery error" are given.
1823 PH/43 $value is now also set for the "else" part of a ${run expansion.
1825 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1826 being worked on, but at least Exim now implements the latest version to
1829 PH/45 In a pipe transport, although a timeout while waiting for the pipe
1830 process to complete was treated as a delivery failure, a timeout while
1831 writing the message to the pipe was logged, but erroneously treated as a
1832 successful delivery. Such timeouts include transport filter timeouts. For
1833 consistency with the overall process timeout, these timeouts are now
1834 treated as errors, giving rise to delivery failures by default. However,
1835 there is now a new Boolean option for the pipe transport called
1836 timeout_defer, which, if set TRUE, converts the failures into defers for
1837 both kinds of timeout. A transport filter timeout is now identified in
1840 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1841 systems where "make" and "gmake" are different, calling "gmake" at top
1842 level broke things. I've arranged for the value of $(MAKE) to be passed
1843 from the Makefile to this script so that it can call the same version of
1847 A note about Exim versions 4.44 and 4.50
1848 ----------------------------------------
1850 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1851 changes of various kinds. As a consequence, a big documentation update was
1852 needed. This delayed the release for rather longer than seemed good, especially
1853 in the light of a couple of (minor) security issues. Therefore, the changes
1854 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1855 release. So 4.44 and 4.50 are in effect two different branches that both start
1858 I have left the 4.50 change log unchanged; it contains all the changes since
1859 4.43. The change log for 4.44 is below; many of its items are identical to
1860 those for 4.50. This seems to be the most sensible way to preserve the
1861 historical information.
1867 1. Minor wording change to the doc/README.SIEVE file.
1869 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1870 computation of the current number of files was incorrect.
1872 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1873 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1876 4. Give more explanation in the error message when the command for a transport
1877 filter fails to execute.
1879 5. There are several places where Exim runs a non-Exim command in a
1880 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1881 was being done only for the command run by the queryprogram router. It is
1882 now done for all such subprocesses. The other cases are: ${run, transport
1883 filters, and the commands run by the lmtp and pipe transports.
1885 6. Added CONFIGURE_GROUP build-time option.
1887 7. Some older OS have a limit of 256 on the maximum number of file
1888 descriptors. Exim was using setrlimit() to set 1000 as a large value
1889 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1890 systems. I've change it so that if it can't get 1000, it tries for 256.
1892 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1893 was an oversight, and furthermore, ever since the addition of extra
1894 controls (e.g. 4.43/32), the checks on when to allow different forms of
1895 "control" were broken. There should now be diagnostics for all cases when a
1896 control that does not make sense is encountered.
1898 9. Added the /retain_sender option to "control=submission".
1900 10. $recipients is now available in the predata ACL (oversight).
1902 11. Tidy the search cache before the fork to do a delivery from a message
1903 received from the command line. Otherwise the child will trigger a lookup
1904 failure and thereby defer the delivery if it tries to use (for example) a
1905 cached ldap connection that the parent has called unbind on.
1907 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1908 of $address_data from the recipient verification was clobbered by the
1909 sender verification.
1911 13. The value of address_data from a sender verification is now available in
1912 $sender_address_data in subsequent conditions in the ACL statement.
1914 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1916 15. Added a new option "connect=<time>" to callout options, to set a different
1919 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1920 was its contents. (It was OK if the option was not defined at all.)
1922 17. A "Completed" log line is now written for messages that are removed from
1923 the spool by the -Mrm option.
1925 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1926 information about exactly what failed.
1928 19. Added -dd to debug only the daemon process.
1930 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1931 handles timeouts, both on the server side and network timeouts. Renamed the
1932 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1934 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1935 It is now set to "smtps".
1937 22. $host_address is now set to the target address during the checking of
1938 ignore_target_hosts.
1940 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1941 being passed; this would have caused $sender_host_name to have been used if
1942 matching the list had actually called for a host name (not very likely,
1943 since this list is usually IP addresses). A host name is now passed as
1946 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1947 code that specifies a non-restarting handler (typically sigaction() in
1948 modern systems) in an attempt to fix a rare and obscure crash bug.
1950 25. Narrowed the window for a race in the daemon that could cause it to ignore
1951 SIGCHLD signals. This is not a major problem, because they are used only to
1952 wake it up if nothing else does.
1954 26. A malformed maildirsize file could cause Exim to calculate negative values
1955 for the mailbox size or file count. Odd effects could occur as a result.
1956 The maildirsize information is now recalculated if the size or filecount
1959 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1960 support for a long time. Removed HAVE_SYS_VFS_H.
1962 28. Installed the latest version of exipick from John Jetmore.
1964 29. In an address list, if the pattern was not a regular expression, an empty
1965 subject address (from a bounce message) matched only if the pattern was an
1966 empty string. Non-empty patterns were not even tested. This was the wrong
1967 because it is perfectly reasonable to use an empty address as part of a
1968 database query. An empty address is now tested by patterns that are
1969 lookups. However, all the other forms of pattern expect the subject to
1970 contain a local part and a domain, and therefore, for them, an empty
1971 address still always fails if the pattern is not itself empty.
1973 30. Exim went into a mad DNS loop when attempting to do a callout where the
1974 host was specified on an smtp transport, and looking it up yielded more
1975 than one IP address.
1977 31. Re-factored the code for checking spool and log partition space into a
1978 function that finds that data and another that does the check. The former
1979 is then used to implement four new variables: $spool_space, $log_space,
1980 $spool_inodes, and $log_inodes.
1982 32. The RFC2047 encoding function was originally intended for short strings
1983 such as real names; it was not keeping to the 75-character limit for
1984 encoded words that the RFC imposes. It now respects the limit, and
1985 generates multiple encoded words if necessary. To be on the safe side, I
1986 have increased the buffer size for the ${rfc2047: expansion operator from
1989 33. It is now permitted to omit both strings after an "if" condition; if the
1990 condition is true, the result is "true". As before, when the second string
1991 is omitted, a false condition yields an empty string. This makes it less
1992 cumbersome to write custom ACL and router conditions.
1994 34. Failure to deliver a bounce message always caused it to be frozen, even if
1995 there was an errors_to setting on the router. The errors_to setting is now
1998 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1999 canonical form (fully expanded) before being placed in
2000 $sender_host_address.
2002 36. The table in the code that translates DNS record types into text (T_A to
2003 "A" for instance) was missing entries for NS and CNAME. It is just possible
2004 that this could have caused confusion if both these types were looked up
2005 for the same domain, because the text type is used as part of Exim's
2006 per-process caching. But the chance of anyone hitting this buglet seems
2009 37. The dnsdb lookup has been extended in a number of ways.
2011 (1) There is a new type, "zns", which walks up the domain tree until it
2012 finds some nameserver records. It should be used with care.
2014 (2) There is a new type, "mxh", which is like "mx" except that it returns
2015 just the host names, not the priorities.
2017 (3) It is now possible to give a list of domains (or IP addresses) to be
2018 looked up. The behaviour when one of the lookups defers can be
2019 controlled by a keyword.
2021 (4) It is now possible to specify the separator character for use when
2022 multiple records are returned.
2024 38. The dnslists ACL condition has been extended: it is now possible to supply
2025 a list of IP addresses and/or domains to be looked up in a particular DNS
2028 39. Added log_selector=+queue_time_overall.
2030 40. When running the queue in the test harness, wait just a tad after forking a
2031 delivery process, to get repeatability of debugging output.
2033 41. Include certificate and key file names in error message when GnuTLS fails
2034 to set them up, because the GnuTLS error message doesn't include the name
2035 of the failing file when there is a problem reading it.
2037 42. Allow both -bf and -bF in the same test run.
2039 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2041 44. The "Exiscan patch" is now merged into the mainline Exim source.
2043 45. Sometimes the final signoff response after QUIT could fail to get
2044 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2045 < 0 before doing a fflush(). This bug looks as though it goes back to the
2046 introduction of TLS in release 3.20, but "sometimes" must have been rare
2047 because the tests only now provoked it.
2049 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2050 (this can affect the format of dates).
2052 47. exim_tidydb, when checking for the continued existence of a message for
2053 which it has found a message-specific retry record, was not finding
2054 messages that were in split spool directories. Consequently, it was
2055 deleting retry records that should have stayed in existence.
2057 48. Steve fixed some bugs in eximstats.
2059 49. The SPA authentication driver was not abandoning authentication and moving
2060 on to the next authenticator when an expansion was forced to fail,
2061 contradicting the general specification for all authenticators. Instead it
2062 was generating a temporary error. It now behaves as specified.
2064 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2065 (the order specifies the preference for clients). The order is now AES256,
2066 AES128, 3DES, ARCFOUR128.
2068 51. Small patch to Sieve code - explicitly set From: when generating an
2071 52. Exim crashed if a remote delivery caused a very long error message to be
2072 recorded - for instance if somebody sent an entire SpamAssassin report back
2073 as a large number of 550 error lines. This bug was coincidentally fixed by
2074 increasing the size of one of Exim's internal buffers (big_buffer) that
2075 happened as part of the Exiscan merge. However, to be on the safe side, I
2076 have made the code more robust (and fixed the comments that describe what
2079 53. Now that there can be additional text after "Completed" in log lines (if
2080 the queue_time_overall log selector is set), a one-byte patch to exigrep
2081 was needed to allow it to recognize "Completed" as not the last thing in
2084 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2085 patch that reportedly fixes this has been added. I am not expert enough to
2086 create a test for it. This is what the patch creator wrote:
2088 "I found a little strange behaviour of ldap code when working with
2089 Windows 2003 AD Domain, where users was placed in more than one
2090 Organization Units. When I tried to give exim partial DN, the exit code
2091 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2092 But simultaneously result of request was absolutely normal ldap result,
2093 so I produce this patch..."
2095 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2096 REFERENCE, so I have modified the code to exclude the patch when that macro
2099 55. Some experimental protocols are using DNS PTR records for new purposes. The
2100 keys for these records are domain names, not reversed IP addresses. The
2101 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2102 leaves it alone. Component reversal etc. now happens only for IP addresses.
2105 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2107 57. Double the size of the debug message buffer (to 2048) so that more of very
2108 long debug lines gets shown.
2110 58. The exicyclog utility now does better if the number of log files to keep
2111 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2113 59. Two changes related to the smtp_active_hostname option:
2115 (1) $smtp_active_hostname is now available as a variable.
2116 (2) The default for smtp_banner uses $smtp_active_hostname instead
2117 of $primary_hostname.
2119 60. The host_aton() function is supposed to be passed a string that is known
2120 to be a valid IP address. However, in the case of IPv6 addresses, it was
2121 not checking this. This is a hostage to fortune. Exim now panics and dies
2122 if the condition is not met. A case was found where this could be provoked
2123 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2124 components; fortuitously, this particular loophole had already been fixed
2125 by change 4.50/55 above.
2127 If there are any other similar loopholes, the new check in host_aton()
2128 itself should stop them being exploited. The report I received stated that
2129 data on the command line could provoke the exploit when Exim was running as
2130 exim, but did not say which command line option was involved. All I could
2131 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2132 running as the user.
2135 61. There was a buffer overflow vulnerability in the SPA authentication code
2136 (which came originally from the Samba project). I have added a test to the
2137 spa_base64_to_bits() function which I hope fixes it.
2140 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2141 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2143 63. The daemon start-up calls getloadavg() while still root for those OS that
2144 need the first call to be done as root, but it missed one case: when
2145 deliver_queue_load_max is set with deliver_drop_privilege. This is
2146 necessary for the benefit of the queue runner, because there is no re-exec
2147 when deliver_drop_privilege is set.
2149 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2150 This has been fixed.
2152 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2153 was in use, was not putting the data itself into the right store pool;
2154 consequently, it could be overwritten for a subsequent message in the same
2155 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2158 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2160 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2161 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2162 the function were treating the return as a boolean value, which happened to
2163 work because 0=false and not-0=true, but is not correct code.
2165 68. The host_aton() function was not handling scoped IPv6 addresses (those
2166 with, for example, "%eth0" on the end) correctly.
2168 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2169 negated items (that is, ~something) in unsigned ints. Some compilers
2170 apparently mutter when there is no cast.
2172 70. If an address verification called from an ACL failed, and did not produce a
2173 user-specific message (i.e. there was only a "system" message), nothing was
2174 put in $acl_verify_message. In this situation, it now puts the system
2177 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2178 session; change 4.31/43 added the unwanted input to the log line - except
2179 that it did not do this in the start of session case. It now does.
2181 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2182 This is wrong and can cause the other end to generate a synchronization
2183 error if it is another Exim or anything else that does the synchronization
2184 check. A QUIT command is no longer sent after a timeout.
2186 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2187 during host lookups.
2189 74. The defer_ok option of callout verification was not working if it was used
2190 when verifying addresses in header lines, that is, for this case:
2192 verify = header_sender/callout=defer_ok
2194 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2195 those file descriptors could be used for SMTP connections. If anything
2196 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2197 could be sent to the SMTP client, causing chaos. The daemon now opens
2198 stdin, stdout, and stderr to /dev/null when it puts itself into the
2201 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2202 log by default. The user can override this with suitable Perl magic.
2204 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2205 the log message when discard triggers, was not working for the DATA ACL or
2206 for the non-SMTP ACL.
2208 78. Error message wording change in sieve.c.
2210 79. If smtp_accept_max_per_host was set, the number of connections could be
2211 restricted to fewer than expected, because the daemon was trying to set up
2212 a new connection before checking whether the processes handling previous
2213 connections had finished. The check for completed processes is now done
2214 earlier. On busy systems, this bug wouldn't be noticed because something
2215 else would have woken the daemon, and it would have reaped the completed
2218 80. If a message was submitted locally by a user whose login name contained one
2219 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2220 It caused a spool format error. I have fixed the spool reading code. A
2221 related problem was that the "from" clause in the Received: line became
2222 illegal because of the space(s). It is now covered by ${quote_local_part.
2224 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2227 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2229 83. Patch from Sieve maintainer to fix unterminated string problem in
2230 "vacation" handling.
2232 84. Some minor changes to the Linux configuration files to help with other
2233 OS variants using glibc.
2235 85. One more patch for Sieve to update vacation handling to latest spec.
2238 ----------------------------------------------------
2239 See the note above about the 4.44 and 4.50 releases.
2240 ----------------------------------------------------
2246 1. Change 4.43/35 introduced a bug that caused file counts to be
2247 incorrectly computed when quota_filecount was set in an appendfile
2250 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2251 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2254 3. Give more explanation in the error message when the command for a transport
2255 filter fails to execute.
2257 4. There are several places where Exim runs a non-Exim command in a
2258 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2259 was being done only for the command run by the queryprogram router. It is
2260 now done for all such subprocesses. The other cases are: ${run, transport
2261 filters, and the commands run by the lmtp and pipe transports.
2263 5. Some older OS have a limit of 256 on the maximum number of file
2264 descriptors. Exim was using setrlimit() to set 1000 as a large value
2265 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2266 systems. I've change it so that if it can't get 1000, it tries for 256.
2268 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2269 was an oversight, and furthermore, ever since the addition of extra
2270 controls (e.g. 4.43/32), the checks on when to allow different forms of
2271 "control" were broken. There should now be diagnostics for all cases when a
2272 control that does not make sense is encountered.
2274 7. $recipients is now available in the predata ACL (oversight).
2276 8. Tidy the search cache before the fork to do a delivery from a message
2277 received from the command line. Otherwise the child will trigger a lookup
2278 failure and thereby defer the delivery if it tries to use (for example) a
2279 cached ldap connection that the parent has called unbind on.
2281 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2282 of $address_data from the recipient verification was clobbered by the
2283 sender verification.
2285 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2286 was its contents. (It was OK if the option was not defined at all.)
2288 11. A "Completed" log line is now written for messages that are removed from
2289 the spool by the -Mrm option.
2291 12. $host_address is now set to the target address during the checking of
2292 ignore_target_hosts.
2294 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2295 being passed; this would have caused $sender_host_name to have been used if
2296 matching the list had actually called for a host name (not very likely,
2297 since this list is usually IP addresses). A host name is now passed as
2300 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2301 code that specifies a non-restarting handler (typically sigaction() in
2302 modern systems) in an attempt to fix a rare and obscure crash bug.
2304 15. Narrowed the window for a race in the daemon that could cause it to ignore
2305 SIGCHLD signals. This is not a major problem, because they are used only to
2306 wake it up if nothing else does.
2308 16. A malformed maildirsize file could cause Exim to calculate negative values
2309 for the mailbox size or file count. Odd effects could occur as a result.
2310 The maildirsize information is now recalculated if the size or filecount
2313 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2314 support for a long time. Removed HAVE_SYS_VFS_H.
2316 18. Updated exipick to current release from John Jetmore.
2318 19. Allow an empty sender to be matched against a lookup in an address list.
2319 Previously the only cases considered were a regular expression, or an
2322 20. Exim went into a mad DNS lookup loop when doing a callout where the
2323 host was specified on the transport, if the DNS lookup yielded more than
2326 21. The RFC2047 encoding function was originally intended for short strings
2327 such as real names; it was not keeping to the 75-character limit for
2328 encoded words that the RFC imposes. It now respects the limit, and
2329 generates multiple encoded words if necessary. To be on the safe side, I
2330 have increased the buffer size for the ${rfc2047: expansion operator from
2333 22. Failure to deliver a bounce message always caused it to be frozen, even if
2334 there was an errors_to setting on the router. The errors_to setting is now
2337 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2338 canonical form (fully expanded) before being placed in
2339 $sender_host_address.
2341 24. Updated eximstats to version 1.33
2343 25. Include certificate and key file names in error message when GnuTLS fails
2344 to set them up, because the GnuTLS error message doesn't include the name
2345 of the failing file when there is a problem reading it.
2347 26. Expand error message when OpenSSL has problems setting up cert/key files.
2350 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2351 (this can affect the format of dates).
2353 28. exim_tidydb, when checking for the continued existence of a message for
2354 which it has found a message-specific retry record, was not finding
2355 messages that were in split spool directories. Consequently, it was
2356 deleting retry records that should have stayed in existence.
2358 29. eximstats updated to version 1.35
2359 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2360 1.35 - bugfix such that pie charts by volume are generated correctly
2362 30. The SPA authentication driver was not abandoning authentication and moving
2363 on to the next authenticator when an expansion was forced to fail,
2364 contradicting the general specification for all authenticators. Instead it
2365 was generating a temporary error. It now behaves as specified.
2367 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2368 (the order specifies the preference for clients). The order is now AES256,
2369 AES128, 3DES, ARCFOUR128.
2371 31. Small patch to Sieve code - explicitly set From: when generating an
2374 32. Exim crashed if a remote delivery caused a very long error message to be
2375 recorded - for instance if somebody sent an entire SpamAssassin report back
2376 as a large number of 550 error lines. This bug was coincidentally fixed by
2377 increasing the size of one of Exim's internal buffers (big_buffer) that
2378 happened as part of the Exiscan merge. However, to be on the safe side, I
2379 have made the code more robust (and fixed the comments that describe what
2382 33. Some experimental protocols are using DNS PTR records for new purposes. The
2383 keys for these records are domain names, not reversed IP addresses. The
2384 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2385 leaves it alone. Component reversal etc. now happens only for IP addresses.
2388 34. The host_aton() function is supposed to be passed a string that is known
2389 to be a valid IP address. However, in the case of IPv6 addresses, it was
2390 not checking this. This is a hostage to fortune. Exim now panics and dies
2391 if the condition is not met. A case was found where this could be provoked
2392 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2393 components; fortuitously, this particular loophole had already been fixed
2394 by change 4.50/55 or 4.44/33 above.
2396 If there are any other similar loopholes, the new check in host_aton()
2397 itself should stop them being exploited. The report I received stated that
2398 data on the command line could provoke the exploit when Exim was running as
2399 exim, but did not say which command line option was involved. All I could
2400 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2401 running as the user.
2404 35. There was a buffer overflow vulnerability in the SPA authentication code
2405 (which came originally from the Samba project). I have added a test to the
2406 spa_base64_to_bits() function which I hope fixes it.
2409 36. The daemon start-up calls getloadavg() while still root for those OS that
2410 need the first call to be done as root, but it missed one case: when
2411 deliver_queue_load_max is set with deliver_drop_privilege. This is
2412 necessary for the benefit of the queue runner, because there is no re-exec
2413 when deliver_drop_privilege is set.
2415 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2416 was in use, was not putting the data itself into the right store pool;
2417 consequently, it could be overwritten for a subsequent message in the same
2418 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2421 38. Sometimes the final signoff response after QUIT could fail to get
2422 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2423 < 0 before doing a fflush(). This bug looks as though it goes back to the
2424 introduction of TLS in release 3.20, but "sometimes" must have been rare
2425 because the tests only now provoked it.
2431 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2432 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2433 Now it returns an integer. A number of calls were still expecting a T/F
2434 return. Fortuitously, in all cases, the tests worked in OK situations,
2435 which is the norm. However, things would have gone wrong on any write
2436 failures on the smtp file descriptor. This function is used when sending
2437 messages over SMTP and also when doing verify callouts.
2439 2. When Exim is called to do synchronous delivery of a locally submitted
2440 message (the -odf or -odi options), it no longer closes stderr before doing
2443 3. Implemented the mua_wrapper option.
2445 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2447 5. Implemented the functions header_remove(), header_testname(),
2448 header_add_at_position(), and receive_remove_recipient(), and exported them
2451 6. If an ACL "warn" statement specified the addition of headers, Exim already
2452 inserted X-ACL-Warn: at the start if there was no header name. However, it
2453 was not making this test for the second and subsequent header lines if
2454 there were newlines in the string. This meant that an invalid header could
2455 be inserted if Exim was badly configured.
2457 7. Allow an ACL "warn" statement to add header lines at the start or after all
2458 the Received: headers, as well as at the end.
2460 8. Added the rcpt_4xx retry error code.
2462 9. Added postmaster_mailfrom=xxx to callout verification option.
2464 10. Added mailfrom=xxxx to the callout verification option, for verify=
2467 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2468 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2470 12. Inserted some casts to stop certain compilers warning when using pointer
2471 differences as field lengths or precisions in printf-type calls (mostly
2472 affecting debugging statements).
2474 13. Added optional readline() support for -be (dynamically loaded).
2476 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2477 same clock tick as a message's arrival, so that its received time was the
2478 same as the "first fail" time on the retry record, and that message
2479 remained on the queue past the ultimate address timeout, every queue runner
2480 would try a delivery (because it was past the ultimate address timeout) but
2481 after another failure, the ultimate address timeout, which should have then
2482 bounced the address, did not kick in. This was a "< instead of <=" error;
2483 in most cases the first failure would have been in the next clock tick
2484 after the received time, and all would be well.
2486 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2487 being recognized when the domain list was tested by the match_domain
2488 condition in an expansion string.
2490 16. Added the ${str2b64: operator.
2492 17. Exim was always calling setrlimit() to set a large limit for the number of
2493 processes, without checking whether the existing limit was already
2494 adequate. (It did check for the limit on file descriptors.) Furthermore,
2495 errors from getrlimit() and setrlimit() were being ignored. Now they are
2496 logged to the main and panic logs, but Exim does carry on, to try to do its
2497 job under whatever limits there are.
2499 18. Imported PCRE 5.0.
2501 19. Trivial typo in log message " temporarily refused connection" (the leading
2504 20. If the log selector return_path_on_delivery was set and an address was
2505 redirected to /dev/null, the delivery process crashed because it assumed
2506 that a return path would always be set for a "successful" delivery. In this
2507 case, the whole delivery is bypassed as an optimization, and therefore no
2510 21. Internal re-arrangement: the function for sending a challenge and reading
2511 a response while authentication was assuming a zero-terminated challenge
2512 string. It's now changed to take a pointer and a length, to allow for
2513 binary data in such strings.
2515 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2517 23. Exim was not respecting finduser_retries when seeking the login of the
2518 uid under which it was called; it was always trying 10 times. (The default
2519 setting of finduser_retries is zero.) Also, it was sleeping after the final
2520 failure, which is pointless.
2522 24. Implemented tls_on_connect_ports.
2524 25. Implemented acl_smtp_predata.
2526 26. If the domain in control=submission is set empty, Exim assumes that the
2527 authenticated id is a complete email address when it generates From: or
2528 Sender: header lines.
2530 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2531 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2532 chown and chgrp in /bin and hostname in /usr/bin.
2534 28. Exim was keeping the "process log" file open after each use, just as it
2535 does for the main log. This opens the possibility of it remaining open for
2536 long periods when the USR1 signal hits a daemon. Occasional processlog
2537 errors were reported, that could have been caused by this. Anyway, it seems
2538 much more sensible not to leave this file open at all, so that is what now
2541 29. The long-running daemon process does not normally write to the log once it
2542 has entered its main loop, and it closes the log before doing so. This is
2543 so that log files can straightforwardly be renamed and moved. However,
2544 there are a couple of unusual error situations where the daemon does write
2545 log entries, and I had neglected to close the log afterwards.
2547 30. The text of an SMTP error response that was received during a remote
2548 delivery was being truncated at 512 bytes. This is too short for some of
2549 the long messages that one sometimes sees. I've increased the limit to
2552 31. It is now possible to make retry rules that apply only when a message has a
2553 specific sender, in particular, an empty sender.
2555 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2556 it possible to be selective about when SMTP synchronization is enforced.
2558 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2560 32. Implemented hosts_connection_nolog.
2562 33. Added an ACL for QUIT.
2564 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2567 35. Added mailbox_size and mailbox_filecount to appendfile.
2569 36. Added control = no_multiline_responses to ACLs.
2571 37. There was a bug in the logic of the code that waits for the clock to tick
2572 in the case where the clock went backwards by a substantial amount such
2573 that the microsecond fraction of "now" was more than the microsecond
2574 fraction of "then" (but the whole seconds number was less).
2576 38. Added support for the libradius Radius client library this is found on
2577 FreeBSD (previously only the radiusclient library was supported).
2583 1. When certain lookups returned multiple values in the form name=value, the
2584 quoting of the values was not always being done properly. Specifically:
2585 (a) If the value started with a double quote, but contained no whitespace,
2587 (b) If the value contained whitespace other than a space character (i.e.
2588 tabs or newlines or carriage returns) it was not quoted.
2589 This fix has been applied to the mysql and pgsql lookups by writing a
2590 separate quoting function and calling it from the lookup code. The fix
2591 should probably also be applied to nisplus, ibase and oracle lookups, but
2592 since I cannot test any of those, I have not disturbed their existing code.
2594 2. A hit in the callout cache for a specific address caused a log line with no
2595 reason for rejecting RCPT. Now it says "Previous (cached) callout
2596 verification failure".
2598 3. There was an off-by-one bug in the queryprogram router. An over-long
2599 return line was truncated at 256 instead of 255 characters, thereby
2600 overflowing its buffer with the terminating zero. As well as fixing this, I
2601 have increased the buffer size to 1024 (and made a note to document this).
2603 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2604 when Exim is waiting for an SMTP response from a remote server, Exim
2605 restarts its select() call on the socket, thereby resetting its timeout.
2606 This is not a problem when such interrupts are rare. Somebody set up a cron
2607 job to run exiwhat every 2 minutes, which is less than the normal select()
2608 timeout (5 or 10 minutes). This meant that the select() timeout never
2609 kicked in because it was always reset. I have fixed this by comparing the
2610 time when an interrupt arrives with the time at the start of the first call
2611 to select(). If more time than the timeout has elapsed, the interrupt is
2612 treated as a timeout.
2614 5. Some internal re-factoring in preparation for the addition of Sieve
2615 extensions (by MH). In particular, the "personal" test is moved to a
2616 separate function, and given an option for scanning Cc: and Bcc: (which is
2617 not set for Exim filters).
2619 6. When Exim created an email address using the login of the caller as the
2620 local part (e.g. when creating a From: or Sender: header line), it was not
2621 quoting the local part when it contained special characters such as @.
2623 7. Installed new OpenBSD configuration files.
2625 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2626 try to make them clearer.
2628 9. Callout options, other than the timeout value, were being ignored when
2629 verifying sender addresses in header lines. For example, when using
2631 verify = header_sender/callout=no_cache
2633 the cache was (incorrectly) being used.
2635 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2636 only the Cygwin environment.
2638 11. When return_path_on_delivery was set as a log selector, if different remote
2639 addresses in the same message used different return paths and parallel
2640 remote delivery occurred, the wrong values would sometimes be logged.
2641 (Whenever a remote delivery process finished, the return path value from
2642 the most recently started remote delivery process was logged.)
2644 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2645 lines when AUTH and/or TLS are in use. This is the "received protocol"
2646 field. Exim used to use "asmtp" for authenticated SMTP, without any
2647 indication (in the protocol name) for TLS use. Now it follows the RFC and
2648 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2649 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2650 names appear in log lines as well as in Received: header lines.
2652 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2653 extensions, and comparison tests, and to fix some bugs.
2655 14. Changes to the "personal" filter test:
2657 (1) The test was buggy in that it was just doing the equivalent of
2658 "contains" tests on header lines. For example, if a user's address was
2659 anne@some.where, the "personal" test would incorrectly be true for
2661 To: susanne@some.where
2663 This test is now done by extracting each address from the header in turn,
2664 and checking the entire address. Other tests that are part of "personal"
2665 are now done using regular expressions (for example, to check local parts
2666 of addresses in From: header lines).
2668 (2) The list of non-personal local parts in From: addresses has been
2669 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2670 taken from the Sieve specification recommendations.
2672 (3) If the message contains any header line starting with "List-" it is
2673 treated as non-personal.
2675 (4) The test for "circular" in the Subject: header line has been removed
2676 because it now seems ill-conceived.
2678 15. Minor typos in src/EDITME comments corrected.
2680 16. Installed latest exipick from John Jetmore.
2682 17. If headers_add on a router specified a text string that was too long for
2683 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2684 of string_sprintf() is now avoided.
2686 18. $message_body_size was not set (it was always zero) when running the DATA
2687 ACL and the local_scan() function.
2689 19. For the "mail" command in an Exim filter, no default was being set for
2690 the once_repeat time, causing a random time value to be used if "once" was
2691 specified. (If the value happened to be <= 0, no repeat happened.) The
2692 default is now 0s, meaning "never repeat". The "vacation" command was OK
2693 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2694 (I found it when inspecting the code).
2696 20. There is now an overall timeout for performing a callout verification. It
2697 defaults to 4 times the callout timeout, which applies to individual SMTP
2698 commands during the callout. The overall timeout applies when there is more
2699 than one host that can be tried. The timeout is checked before trying the
2700 next host. This prevents very long delays if there are a large number of
2701 hosts and all are timing out (e.g. when the network connections are timing
2702 out). The value of the overall timeout can be changed by specifying an
2703 additional sub-option for "callout", called "maxwait". For example:
2705 verify = sender/callout=5s,maxwait=20s
2707 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2708 to the end before writing, but this should make it even safer).
2710 22. Exim was forgetting that it had advertised PIPELINING for the second and
2711 subsequent messages on an SMTP connection. It was also not resetting its
2712 memory on STARTTLS and an internal HELO.
2714 23. When Exim logs an SMTP synchronization error within a session, it now
2715 records whether PIPELINING has been advertised or not.
2717 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2718 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2719 rather than long int.
2721 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2723 26. Added the never_mail option to autoreply.
2729 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2730 crash if the getsockname() call failed; this can happen if a connection is
2731 closed very soon after it is established. The problem was simply in the
2732 order in which certain operations were done, causing Exim to try to write
2733 to the SMTP stream before it had set up the file descriptor. The bug has
2734 been fixed by making things happen in the correct order.
2740 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2741 before the connection was closed, thus losing the rejection response.
2743 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2744 some early Solaris releases, but causes trouble in current releases where
2745 socklen_t is defined.
2747 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2750 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2753 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2754 that releasing the top of it at the end releases what was used for sub-
2755 expansions (unless the block got too big). However, discard this block if
2756 the first thing is a variable or header, so that we can use its block when
2757 it is dynamic (useful for very large $message_headers, for example).
2759 6. Lookups now cache *every* query, not just the most recent. A new, separate
2760 store pool is used for this. It can be recovered when all lookup caches are
2761 flushed. Lookups now release memory at the end of their result strings.
2762 This has involved some general refactoring of the lookup sources.
2764 7. Some code has been added to the store_xxx() functions to reduce the amount
2765 of flapping under certain conditions.
2767 8. log_incoming_interface used to affect only the <= reception log lines. Now
2768 it causes the local interface and port to be added to several more SMTP log
2769 lines, for example "SMTP connection from", and rejection lines.
2771 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2773 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2775 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2776 could be overwritten at the end of the current message (or the start of a
2777 new message if it was set in a HELO ACL). The value is now preserved for
2778 the duration of the SMTP connection.
2780 12. If a transport had a headers_rewrite setting, and a matching header line
2781 contained an unqualified address, that address was qualified, even if it
2782 did not match any rewriting rules. The underlying bug was that the values
2783 of the flags that permit the existence of unqualified sender and recipient
2784 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2785 non-local messages, and by -bnq for local messages) were not being
2786 preserved with the message after it was received.
2788 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2789 "next input=" as part of the text comprising the host identity instead of
2790 the correct text. The code was using the same buffer for two different
2791 strings. However, depending on which order the printing function evaluated
2792 its arguments, the bug did not always show up. Under Linux, for example, my
2793 test suite worked just fine.
2795 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2796 doesn't work with some older versions of Perl. It has been changed to "my",
2797 which in any case is probably the better facility to use.
2799 15. A really picky compiler found some instances of statements for creating
2800 error messages that either had too many or two few arguments for the format
2803 16. The size of the buffer for calls to the DNS resolver has been increased
2804 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2805 for addresses that have a lot of PTR records. This alleviates a problem; it
2806 does not fully solve it.
2808 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2809 buffer now truncates the list and logs the incident, which is the same
2810 action as happens when Exim is looking up a host name and its aliases.
2811 Previously in this situation something unpredictable would happen;
2812 sometimes it was "internal error: store_reset failed".
2814 18. If a server dropped the connection unexpectedly when an Exim client was
2815 using GnuTLS and trying to read a response, the client delivery process
2816 crashed while trying to generate an error log message.
2818 19. If a "warn" verb in an ACL added multiple headers to a message in a single
2819 string, for example:
2821 warn message = H1: something\nH2: something
2823 the text was added as a single header line from Exim's point of view
2824 though it ended up OK in the delivered message. However, searching for the
2825 second and subsequent header lines using $h_h2: did not work. This has been
2826 fixed. Similarly, if a system filter added multiple headers in this way,
2827 the routers could not see them.
2829 20. Expanded the error message when iplsearch is called with an invalid key to
2830 suggest using net-iplsearch in a host list.
2832 21. When running tests using -bh, any delays imposed by "delay" modifiers in
2833 ACLs are no longer actually imposed (and a message to that effect is
2836 22. If a "gecos" field in a passwd entry contained escaped characters, in
2837 particular, if it contained a \" sequence, Exim got it wrong when building
2838 a From: or a Sender: header from that name. A second bug also caused
2839 incorrect handling when an unquoted " was present following a character
2840 that needed quoting.
2842 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2843 was not being matched caselessly.
2845 24. Arranged for all hyphens in the exim.8 source to be escaped with
2848 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2849 properly. Recipient callout cache records were still being keyed to include
2850 the sender, even when use_sender was set false. This led to far more
2851 callouts that were necessary. The sender is no longer included in the key
2852 when use_sender is false.
2854 26. Added "control = submission" modifier to ACLs.
2856 27. Added the ${base62d: operator to decode base 62 numbers.
2858 28. dnsdb lookups can now access SRV records.
2860 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2861 the configuration file.
2863 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2864 (-v) mode. This makes the output for a verbose queue run more intelligible.
2866 31. Added a use_postmaster feature to recipient callouts.
2868 32. Added the $body_zerocount variable, containing the number of binary zero
2869 bytes in the message body.
2871 33. The time of last modification of the "new" subdirectory is now used as the
2872 "mailbox time last read" when there is a quota error for a maildir
2875 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2877 35. Added +ignore_unknown as a special item in host lists.
2879 36. Code for decoding IPv6 addresses in host lists is now included, even if
2880 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2881 address was recognized as an IP address, but was then not correctly decoded
2882 into binary, causing unexpected and incorrect effects when compared with
2889 1. Very minor rewording of debugging text in manualroute to say "list of
2890 hosts" instead of "hostlist".
2892 2. If verify=header_syntax was set, and a header line with an unqualified
2893 address (no domain) and a large number of spaces between the end of the
2894 name and the colon was received, the reception process suffered a buffer
2895 overflow, and (when I tested it) crashed. This was caused by some obsolete
2896 code that should have been removed. The fix is to remove it!
2898 3. When running in the test harness, delay a bit after writing a bounce
2899 message to get a bit more predictability in the log output.
2901 4. Added a call to search_tidyup() just before forking a reception process. In
2902 theory, someone could use a lookup in the expansion of smtp_accept_max_
2903 per_host which, without the tidyup, could leave open a database connection.
2905 5. Added the variables $recipient_data and $sender_data which get set from a
2906 lookup success in an ACL "recipients" or "senders" condition, or a router
2907 "senders" option, similar to $domain_data and $local_part_data.
2909 6. Moved the writing of debug_print from before to after the "senders" test
2912 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2913 problems for message scanning, either using a data ACL, or using
2914 local_scan() because the Received: header was not generated till after they
2915 were called (in order to set the time as the time of reception completion).
2916 I have revised the way this works. The header is now generated after the
2917 body is received, but before the ACL or local_scan() are called. After they
2918 are run, the timestamp in the header is updated.
2924 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2925 before starting a queue runner without re-exec. This happened only when
2926 deliver_drop_privilege was set or when the Exim user was set to root. The
2927 effect of the bug was that timeouts during subsequent deliveries caused
2928 crashes instead of being properly handled. The handler is now left at its
2929 default (and expected) setting.
2931 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2932 message, again when deliver_drop_privilege is set or Exim is run as root.
2933 The bug described in (1) was not present in this case, but the tidying up
2934 of the other signals was missing. I have made the two cases consistent.
2936 3. The ignore_target_hosts setting on a manualroute router was being ignored
2937 for hosts that were looked up using the /MX notation.
2939 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2942 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2943 operated on the sender address. After changing the $sender_address to <>
2944 for the sender address verify, Exim was re-instated it as the original
2945 (before rewriting) address, but remembering that it had rewritten it, so it
2946 wasn't rewriting it again. This bug also had the effect of breaking the
2947 sender address verification caching when the sender address was rewritten.
2949 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2950 This has been changed so that if the ip literal address matches
2951 ignore_target_hosts, the router declines.
2953 7. Added expansion conditions match_domain, match_address, and match_local_
2954 part (NOT match_host).
2956 8. The placeholder for the Received: header didn't have a length field set.
2958 9. Added code to Exim itself and to exim_lock to test for a specific race
2959 condition that could lead to file corruption when using MBX delivery. The
2960 issue is with the lockfile that is created in /tmp. If this file is removed
2961 after a process has opened it but before that process has acquired a lock,
2962 there is the potential for a second process to recreate the file and also
2963 acquire a lock. This could lead to two Exim processes writing to the file
2964 at the same time. The added code performs the same test as UW imapd; it
2965 checks after acquiring the lock that its file descriptor still refers to
2966 the same named file.
2968 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2969 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2972 11. Added the smtp_active_hostname option. If used, this will typically be made
2973 to depend on the incoming interface address. Because $interface_address is
2974 not set up until the daemon has forked a reception process, error responses
2975 that can happen earlier (such as "too many connections") no longer contain
2978 12. If an expansion in a condition on a "warn" statement fails because a lookup
2979 defers, the "warn" statement is abandoned, and the next ACL statement is
2980 processed. Previously this caused the whole ACL to be aborted.
2982 13. Added the iplsearch lookup type.
2984 14. Added ident_timeout as a log selector.
2986 15. Added tls_certificate_verified as a log selector.
2988 16. Added a global option tls_require_ciphers (compare the smtp transport
2989 option of the same name). This controls incoming TLS connections.
2991 17. I finally figured out how to make tls_require_ciphers do a similar thing
2992 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2993 before starting the TLS session.
2995 18. Tabs are now shown as \t in -bP output.
2997 19. If the log selector return_path_on_delivery was set, Exim crashed when
2998 bouncing a message because it had too many Received: header lines.
3000 20. If two routers both had headers_remove settings, and the first one included
3001 a superfluous trailing colon, the final name in the first list and the
3002 first name in the second list were incorrectly joined into one item (with a
3003 colon in the middle).
3009 1. Added -C and -D options to the exinext utility, mainly to make it easier
3010 to include in the automated testing, but these could be helpful when
3011 multiple configurations are in use.
3013 2. The exinext utility was not formatting the output nicely when there was
3014 an alternate port involved in the retry record key, nor when there was a
3015 message id as well (for retries that were specific to a specific message
3016 and a specific host). It was also confused by IPv6 addresses, because of
3017 the additional colons they contain. I have fixed the IPv4 problem, and
3018 patched it up to do a reasonable job for IPv6.
3020 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3021 delivery, the log line now contains "pipelined" if PIPELINING was used.
3023 4. An SMTP transport process used to panic and die if the bind() call to set
3024 an explicit outgoing interface failed. This has been changed; it is now
3025 treated in the same way as a connect() failure.
3027 5. A reference to $sender_host_name in the part of a conditional expansion
3028 that was being skipped was still causing a DNS lookup. This no longer
3031 6. The def: expansion condition was not recognizing references to header lines
3032 that used bh_ and bheader_.
3034 7. Added the _cache feature to named lists.
3036 8. The code for checking quota_filecount in the appendfile transport was
3037 allowing one more file than it should have been.
3039 9. For compatibility with Sendmail, the command line option
3047 and sets the incoming protocol and host name (for trusted callers). The
3048 host name and its colon can be omitted when only the protocol is to be set.
3049 Note the Exim already has two private options, -pd and -ps, that refer to
3050 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3051 "s", but I don't think that's a major issue.
3053 10. A number of refactoring changes to the code, none of which should affect
3056 (a) The number of logging options was getting close to filling up the
3057 32-bit word that was used as a bit map. I have split them into two classes:
3058 those that are passed in the argument to log_write(), and those that are
3059 only ever tested independently outside of that function. These are now in
3060 separate 32-bit words, so there is plenty of room for expansion again.
3061 There is no change in the user interface or the logging behaviour.
3063 (b) When building, for example, log lines, the code previously used a
3064 macro that called string_cat() twice, in order to add two strings. This is
3065 not really sufficiently general. Furthermore, there was one instance where
3066 it was actually wrong because one of the argument was used twice, and in
3067 one call a function was used. (As it happened, calling the function twice
3068 did not affect the overall behaviour.) The macro has been replaced by a
3069 function that can join an arbitrary number of extra strings onto a growing
3072 (c) The code for expansion conditions now uses a table and a binary chop
3073 instead of a serial search (which was left over from when there were very
3074 few conditions). Also, it now recognizes conditions like "pam" even when
3075 the relevant support is not compiled in: a suitably worded error message is
3076 given if an attempt is made to use such a condition.
3078 11. Added ${time_interval:xxxxx}.
3080 12. A bug was causing one of the ddress fields not to be passed back correctly
3081 from remote delivery subprocesses. The field in question was not being
3082 subsequently used, so this caused to problems in practice.
3084 13. Added new log selectors queue_time and deliver_time.
3086 14. Might have fixed a bug in maildirsizefile handling that threw up
3087 "unexpected character" debug warnings, and recalculated the data
3088 unnecessarily. In any case, I expanded the warning message to give more
3091 15. Added the message "Restricted characters in address" to the statements in
3092 the default ACL that block characters like @ and % in local parts.
3094 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3095 Three changes have been made:
3097 (a) There was a serious bug; a negative response to MAIL caused the whole
3098 recipient domain to be cached as invalid, thereby blocking all messages
3099 to all local parts at the same domain, from all senders. This bug has
3100 been fixed. The domain is no longer cached after a negative response to
3101 MAIL if the sender used is not empty.
3103 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3106 (c) A new callout option, "use_sender" has been added for people who want
3107 the modified behaviour.
3113 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3116 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3117 indeed breaks things for older releases.
3119 3. Added additional logging to the case where there is a problem reading data
3120 from a filter that is running in a subprocess using a pipe, in order to
3121 try to track down a specific problem.
3123 4. Testing facility fudge: when running in the test harness and attempting
3124 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3125 getting "No route to host". Convert this to a timeout.
3127 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3130 6. Some OS don't have socklen_t but use size_t instead. This affects the
3131 fifth argument of getsockopt() amongst other things. This is now
3132 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3133 can be set for individual OS. I have set it for SunOS5, OSF1, and
3134 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3135 some earlier ones do not.
3137 7. Change 4.30/15 was not doing the test caselessly.
3139 8. The standard form for an IPv6 address literal was being rejected by address
3140 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3141 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3142 this, as well as the form without the "IPv6" on the front (but only when
3143 address literals are enabled, of course).
3145 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3147 10. Exim crashed if a message with an empty sender address specified by -f
3148 encountered a router with an errors_to setting. This could be provoked only
3149 by a command such as
3153 where an empty string was supplied; "<>" did not hit this bug.
3155 11. Installed PCRE release 4.5.
3157 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3158 remained set. It is now erased.
3160 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3161 times from message ids (which are base 36 rather than the normal 62).
3163 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3164 were being counted as actual protocol errors, and logged if the log
3165 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3166 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3167 and DATA following a set of rejected RCPTs do not count as protocol errors.
3168 In other words, Exim assumes they were pipelined, though this may not
3169 actually be the case. Of course, in all cases the client gets an
3170 appropriate error code.
3172 15. If a lookup fails in an ACL condition, a message about the failure may
3173 be available; it is used if testing the ACL cannot continue, because most
3174 such messages specify what the cause of the deferral is. However, some
3175 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3176 that caused an old message to be retained and used if a later statement
3177 caused a defer, replacing the real cause of the deferral.
3179 16. If an IP address had so many PTR records that the DNS lookup buffer
3180 was not large enough to hold them, Exim could crash while trying to process
3181 the truncated data. It now detects and logs this case.
3183 17. Further to 4.21/58, another change has been made: if (and only if) the
3184 first line of a message (the first header line) ends with CRLF, a bare LF
3185 in a subsequent header line has a space inserted after it, so as not to
3186 terminate the header.
3188 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3189 unnecessarily, used atoi() instead of strtol(), and didn't check the
3190 termination when getting file sizes from file names by regex.
3192 19. Completely re-implemented the support for maildirsize files, in the light
3193 of a number of problems with the previous contributed implementation
3194 (4.30/29). In particular:
3196 . If the quota is zero, the maildirsize file is maintained, but no quota is
3199 . If the maildir directory does not exist, it is created before any attempt
3200 to write a maildirsize file.
3202 . The quota value in the file is just a cache; if the quota is changed in
3203 the transport, the new value overrides.
3205 . A regular expression is available for excluding directories from the
3208 20. The autoreply transport checks the characters in options that define the
3209 message's headers; it allows continued headers, but it was checking with
3210 isspace() after an embedded newline instead of explicitly looking for a
3213 21. If all the "regular" hosts to which an address was routed had passed their
3214 expiry times, and had not reached their retry times, the address was
3215 bounced, even if fallback hosts were defined. Now Exim should go on to try
3218 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3219 equivalent code in the SMTP transport. Some hosts send humungous responses
3220 to HELO/EHLO, more than 1024 it seems.
3222 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3223 gives compiler warnings in some environments. I've now done it "properly",
3226 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3227 (because of problems with the built-in one) was declared to return uschar *
3228 instead of char *, causing compiler failure.
3230 25. Fixed a file descriptor leak when processing alias/forward files.
3232 26. Fixed a minor format string issue in dbfn.c.
3234 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3236 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3237 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3240 29. When a sender address is verified, it is cached, to save repeating the test
3241 when there is more than one recipient in a message. However, when the
3242 verification involves a callout, it is possible for different callout
3243 options to be set for different recipients. It is too complicated to keep
3244 track of this in the cache, so now Exim always runs a verification when a
3245 callout is required, relying on the callout cache for the optimization.
3246 The overhead is duplication of the address routing, but this should not be
3249 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3250 to be verified with callout=postmaster, and the main callout worked but the
3251 postmaster check failed, the verification correctly failed. However, if a
3252 subsequent RCPT command asked for sender verification *without* the
3253 postmaster check, incorrect caching caused this verification also to fail,
3256 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3257 it was not caching the DNS options (qualify_single, search_parents) that
3258 were used when the lookup failed. A subsequent lookup with different
3259 options therefore always gave the same answer, though there were cases
3260 where it should not have. (Example: a "domains = !$mx_any" option on a
3261 dnslookup router: the "domains" option is always processed without any
3262 widening, but the router might have qualify_single set.) Now Exim uses the
3263 cached value only when the same options are set.
3265 32. Added John Jetmore's "exipick" utility to the distribution.
3267 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3268 than a timeout (e.g. a certificate is required, and is not provided), an
3269 Exim server now closes the connection immediately. Previously it waited for
3270 the client to close - but if the client is SSL, it seems that they each
3271 wait for each other, leading to a delay before one of them times out.
3273 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3274 maintained 0.8.x compatibility because I don't think many are using it, and
3275 it is clearly obsolete.
3277 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3280 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3281 client certificate was expired. A simple patch fixes this, though I don't
3282 understand the full logic of why the verify callback is called multiple
3285 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3286 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3287 which causes problems with some clients (such as the Certicom SSL Plus
3288 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3289 disables the coutermeasure allowing Eudora to connect."
3291 38. Exim was not checking that a write() to a log file succeeded. This could
3292 lead to Bad Things if a log got too big, in particular if it hit a file
3293 size limit. Exim now panics and dies if it cannot write to a log file, just
3294 as it does if it cannot open a log file.
3296 39. Modified OS/Makefile-Linux so that it now contains
3298 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3300 The two -D definitions ensure that Exim is compiled with large file
3301 support, which makes it possible to handle log files that are bigger than
3304 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3305 instance) a domain was checked against a named list that involved a lookup,
3306 causing $domain_data to be set, then another domain was checked against the
3307 same list, then the first domain was re-checked, the value of $domain_data
3308 after the final check could be wrong. In particular, if the second check
3309 failed, it could be set empty. This bug probably also applied to
3312 41. The strip_trailing_dot option was not being applied to the address given
3313 with the -f command-line option.
3315 42. The code for reading a message's header from the spool was incrementing
3316 $received_count, but never initializing it. This meant that the value was
3317 incorrect (doubled) while delivering a message in the same process in which
3318 it was received. In the most common configuration of Exim, this never
3319 happens - a fresh exec is done - but it can happen when
3320 deliver_drop_privilege is set.
3322 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3323 it now includes up to 150 characters of the unexpected data in the log
3326 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3327 and building data strings. The size of both of these buffers was 10 000
3328 bytes - far larger than anybody would *ever* want, thought I. Needless to
3329 say, somebody hit the limit. I have increased the maximum line length to
3330 20 000 and the maximum data length of concatenated lines to 100 000. I have
3331 also fixed two bugs, because there was no checking on these buffers. Tsk,
3332 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3333 buffer is too small.
3335 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3336 lsearch lookups. Now it does.
3338 46. When parsing a route_list item in a manualroute router, a fixed-length
3339 buffer was used for the list of hosts. I made this 1024 bytes long,
3340 thinking that nobody would ever have a list of hosts that long. Wrong.
3341 Somebody had a whole pile of complicated expansion conditions, and the
3342 string was silently truncated, leading to an expansion error. It turns out
3343 that it is easier to change to an unlimited length (owing to other changes
3344 that have happened since this code was originally written) than to build
3345 structure for giving a limitation error. The length of the item that
3346 expands into the list of hosts is now unlimited.
3348 47. The lsearch lookup could not handle data where the length of text line was
3349 more than 4095 characters. Such lines were truncated, leading to shortened
3350 data being returned. It should now handle lines of any length.
3352 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3353 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3356 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3357 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3358 no longer happens in comment lines. A list of replacements is now placed
3359 at the head of all of the source files, except those whose only change is
3360 to replace PERL_COMMAND in the very first #! line.
3362 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3363 messages on the queue, with a bottom-up merge sort, using code contributed
3364 by Michael Haardt. This should make operations like -bp somewhat faster on
3365 large queues. It won't affect queue runners, except when queue_run_in_order
3368 51. Installed eximstats 1.31 in the distribution.
3370 52. Added support for SRV lookups to the dnslookup router.
3372 53. If an ACL referred to $message_body or $message_body_end, the value was not
3373 reset for any messages that followed in the same SMTP session.
3375 54. The store-handling optimization for building very long strings was not
3376 differentiating between the different store pools. I don't think this
3377 actually made any difference in practice, but I've tidied it.
3379 55. While running the routers to verify a sender address, $sender_address
3380 was still set to the sender address. This is wrong, because when routing to
3381 send a bounce to the sender, it would be empty. Therefore, I have changed
3382 it so that, while verifying a sender address, $sender_address is set to <>.
3383 (There is no change to what happens when verifying a recipient address.)
3385 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3386 target A or AAAA records (if not already returned) without resetting the
3387 qualify_single or search_parents options of the DNS resolver. These are
3388 inappropriate in this case because the targets of MX and SRV records must
3389 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3390 target that, when qualified, matched something in the local domain. These
3391 two options are now turned off when doing these lookups.
3393 57. It seems that at least some releases of Reiserfs (which does not have the
3394 concept of a fixed number of inodes) returns zero and not -1 for the
3395 number of available inodes. This interacted badly with check_spool_inodes,
3396 which assumed that -1 was the "no such thing" setting. What I have done is
3397 to check that the total number of inodes is greater than zero before doing
3398 the test of how many are available.
3400 58. When a "warn" ACL statement has a log_message modifier, the message is
3401 remembered, and not repeated. This is to avoid a lot of repetition when a
3402 message has many recipients that cause the same warning to be written.
3403 Howewer, Exim was preserving the list of already written lines for an
3404 entire SMTP session, which doesn't seem right. The memory is now reset if a
3405 new message is started.
3407 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3408 debugging output unless log_rewrite was also set.
3410 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3411 of (int)(handle) when we know that handle contains (void *)(-1).
3413 61. The Exim daemon panic-logs an error return when it closes the incoming
3414 connection. However "connection reset by peer" seems to be common, and
3415 isn't really an error worthy of noting specially, so that particular error
3418 62. When Exim is trying to find all the local interfaces, it used to panic and
3419 die if the ioctl to get the interface flags failed. However, it seems that
3420 on at least one OS (Solaris 9) it is possible to have an interface that is
3421 included in the list of interfaces, but for which you get a failure error
3422 for this call. This happens when the interface is not "plumbed" into a
3423 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3424 failure of the "get flags" call assumes that the interface is down.
3426 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3427 makes life easier for people who are doing arithmetic on fields extracted
3428 from dates, where you often get leading zeros that should not be
3429 interpreted as octal.
3431 64. Added qualify_domain to the redirect router, to override the global
3434 65. If a pathologically long header line contained very many addresses (the
3435 report of this problem mentioned 10 000) and each of them was rewritten,
3436 Exim could use up a very large amount of memory. (It kept on making new
3437 copies of the header line as it rewrote, and never released the old ones.)
3438 At the expense of a bit more processing, the header rewriting function has
3439 been changed so that it no longer eats memory in this way.
3441 66. The generation of the Received: header has been moved from the time that a
3442 message starts to be received, to the time that it finishes. The timestamp
3443 in the Received: header should now be very close to that of the <= log
3444 line. There are two side-effects of this change:
3446 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3447 logged header lines no longer include the local Received: line, because
3448 it has not yet been created. The same applies to a copy of the message
3449 that is returned to a non-SMTP sender when a message is rejected.
3451 (b) When a filter file is tested using -bf, no additional Received: header
3452 is added to the test message. After some thought, I decided that this
3455 This change does not affect the value of $received_for. It is still set
3456 after address rewriting, but before local_scan() is called.
3458 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3460 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3461 gave an unhelpful panic error message, and a defer error. I have managed to
3462 change this behaviour so that it now rejects any supplied certificate,
3463 which seems right, as the list of acceptable certificates is empty.
3465 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3466 gave an unhelpful defer error. I have not managed to make this reject any
3467 supplied certificates, but the error message it gives is "no certificate
3468 supplied", which is not helpful.
3470 70. exigrep's output now also includes lines that are not associated with any
3471 message, but which match the given pattern. Implemented by a patch from
3472 Martin Sluka, which also tidied up the Perl a bit.
3474 71. Recipient callout verification, like sender verification, was using <> in
3475 the MAIL FROM command. This isn't really the right thing, since the actual
3476 sender may affect whether the remote host accepts the recipient or not. I
3477 have changed it to use the actual sender in the callout; this means that
3478 the cache record is now keyed on a recipient/sender pair, not just the
3479 recipient address. There doesn't seem to be a real danger of callout loops,
3480 since a callout by the remote host to check the sender would use <>.
3481 [SEE ABOVE: changed after hitting problems.]
3483 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3484 temporary errors. However, in the case of such a code being given after
3485 the end of a data transmission (i.e. after ".") Exim was failing to write
3486 a retry record for the message. (Yes, there was some broken host that was
3487 actually sending 8xx at this point.)
3489 73. An unknown lookup type in a host list could cause Exim to panic-die when
3490 the list was checked. (An example that provoked this was putting <; in the
3491 middle of a list instead of at the start.) If this happened during a DATA
3492 ACL check, a -D file could be left lying around. This kind of configuration
3493 error no longer causes Exim to die; instead it causes a defer errror. The
3494 incident is still logged to the main and panic logs.
3496 74. Buglet left over from Exim 3 conversion. The message "too many messages
3497 in one connection" was written to the rejectlog but not the mainlog, except
3498 when address rewriting (yes!) was being logged.
3500 75. Added write_rejectlog option.
3502 76. When a system filter was run not as root (that is, when system_filter_user
3503 was set), the values of the $n variables were not being returned to the
3504 main process; thus, they were not subsequently available in the $sn
3507 77. Added +return_path_on_delivery log selector.
3509 78. A connection timeout was being treated differently from recipients deferred
3510 when testing hosts_max_try with a message that was older than the host's
3511 retry timeout. (The host should not be counted, thus allowing all hosts to
3512 be tried at least once before bouncing.) This may have been the cause of an
3513 occasionally reported bug whereby a message would remain on the queue
3514 longer than the retry timeout, but would be bounced if a delivery was
3515 forced. I say "may" because I never totally pinned down the problem;
3516 setting up timeout/retry tests is difficult. See also the next item.
3518 79. The ultimate address timeout was not being applied to errors that involved
3519 a combination of host plus message (for example, a timeout on a MAIL
3520 command). When an address resolved to a number of possible hosts, and they
3521 were not all tried for each delivery (e.g. because of hosts_max_try), a
3522 message could remain on the queue longer than the retry timeout.
3524 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3527 81. Fixed an obscure SMTP outgoing bug which required at least the following
3528 conditions: (a) there was another message waiting for the same server;
3529 (b) the server returned 5xx to all RCPT commands in the first message so
3530 that the message was not completed; (c) the server dropped the connection
3531 or gave a negative response to the RSET that Exim sends to abort the
3532 transaction. The observed case was a dropped connection after DATA that had
3533 been sent in pipelining mode. That is, the server had advertised PIPELINING
3534 but was not implementing it correctly. The effect of the bug was incorrect
3535 behaviour, such as trying another host, and this could lead to a crash.
3541 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3542 and daemon.c were passed as pointers to ints; they should have been
3543 pointers to socklen_t variables (which are typically unsigned ints).
3545 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3548 3. Fixed a really odd bug that affected only the testing scheme; patching a
3549 certain fixed string in the binary changed the value of another string that
3550 happened to be identical to the end of the original first string.
3552 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3553 name", it returns that address as the IP address. On some operating
3554 systems (e.g. Solaris), it also passes back the IP address string as the
3555 "host name". However, on others (e.g. Linux), it passes back an empty
3556 string. Exim wasn't checking for this, and was changing the host name to an
3557 empty string, assuming it had been canonicized.
3559 5. Although rare, it is permitted to have more than one PTR record for a given
3560 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3561 all the names associated with an address, because they do in Solaris.
3562 However, it seems that they do not in Linux for data that comes from the
3563 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3564 I found this out when I moved to a new Linux workstation and tried to run
3565 the Exim test suite.
3567 To get round this problem I have changed the code so that it now does its
3568 own call to the DNS to look up PTR records when searching for a host name.
3569 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3570 addresses that are only in /etc/hosts are still found.
3572 This behaviour is, however, controlled by an option called host_lookup_
3573 order, which defaults to "bydns:byaddr". If people want to use the other
3574 order, or indeed, just use one or the other means of lookup, they can
3575 specify it in this variable.
3577 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3578 some operating systems, this comes back from gethostbyaddr() as an empty
3579 string, and this is what Exim used to test for. However, it seems that in
3580 other systems, "." is yielded. Exim now tests for this case too.
3582 7. The values of check_spool_space and check_log_space are now held internally
3583 as a number of kilobytes instead of an absolute number of bytes. If a
3584 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3585 kilobyte. This means that much larger values can be stored.
3587 8. Exim monitor: an attempt to get the action menu when not actually pointing
3588 at a message produces an empty menu entitled "No message selected". This
3589 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3590 no entries in it ("Shell widget menu has zero width and/or height"). So I
3591 have added a single, blank menu entry in this case.
3593 9. Added ${quote_local_part.
3595 10. MIME decoding is now applied to the contents of Subject: header lines when
3598 11. Now that a reference to $sender_host_address automatically causes a reverse
3599 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3600 host lookup before query-style lookups in lists that might use this
3601 variable. This has therefore been abolished, and the "net-" prefix is no
3602 longer necessary for query-style lookups.
3604 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3605 have been a typo for LFLAGS, so it has been changed.
3607 13. The install script calls Exim with "-C /dev/null" in order to find the
3608 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3609 to be output. Howeve, since Exim outputs its version number before the
3610 error, it didn't break the script. It just looked ugly. I fixed this by
3611 always allowing "-C /dev/null" if the caller is root.
3613 14. Ignore overlarge ACL variable number when reading spool file - insurance
3614 against a later release with more variables having written the file.
3616 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3617 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3618 this, as well as the form without the "IPv6" on the front.
3620 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3621 OS/Makefile-Darwin file.
3623 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3624 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3626 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3627 variable. However, some OS use ints for inodes. I've added cast to long int
3628 to get rid of the compiler warning.
3630 19. I had forgotten to lock out "/../" in configuration file names when
3631 ALT_CONFIG_PREFIX was set.
3633 20. Routers used for verification do not need to specify transports. However,
3634 if such a router generated a host list, and callout was configured, Exim
3635 crashed, because it could not find a port number from the (non-existent)
3636 transport. It now assumes port 25 in this circumstance.
3638 21. Added the -t option to exigrep.
3640 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3641 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3642 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3643 them is set, LOOKUP_LSEARCH is forced.
3645 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3647 24. Added sender and host information to the "rejected by local_scan()" log
3648 line; previously there was no indication of these.
3650 25. Added .include_if_exists.
3652 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3653 Linux. It turns out that not all file systems support this. Apparently some
3654 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3655 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3656 not supported on the file descriptor, is now ignored when Exim is trying to
3657 sync a directory. This applies only to Linux.
3659 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3661 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3662 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3665 29. Added support for maildirsize files from supplied patch (modified a bit).
3667 30. The use of :fail: followed by an empty string could lead Exim to respond to
3668 sender verification failures with (e.g.):
3670 550 Verification failed for <xxx>
3671 550 Sender verify failed
3673 where the first response line was missing the '-' that indicates it is not
3674 the final line of the response.
3676 31. The loop for finding the name of the user that called Exim had a hardwired
3677 limit of 10; it now uses the value of finduser_retries, which is used for
3678 all other user lookups.
3680 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3683 33. Exim was neglecting to zero errno before one call of strtol() when
3684 expanding a string and expecting an integer value. On some systems this
3685 resulted in spurious "integer overflow" errors. Also, it was casting the
3686 result into an int without checking.
3688 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3689 did not work. The code looks as if it has *never* worked, though it appears
3690 to have been documented since at least releast 1.62. I have made it work.
3692 35. The "timeout_DNS" error in retry rules, also documented since at least
3693 1.62, also never worked. As it isn't clear exactly what this means, and
3694 clearly it isn't a major issue, I have abolished the feature by treating it
3695 as "timeout", and writing a warning to the main and panic logs.
3697 36. The display of retry rules for -brt wasn't always showing the error code
3700 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3701 timeout_connect_A, timeout_connect_MX.
3703 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3704 to the empty sender.
3706 39. The daemon was not analysing the content of -oX till after it had closed
3707 stderr and disconnected from the controlling terminal. This meant that any
3708 syntax errors were only noted on the panic log, and the return code from
3709 the command was 0. By re-arranging the code a little, I've made the
3710 decoding happen first, so such errors now appear on stderr, and the return
3711 code is 1. However, the actual setting up of the sockets still happens in
3712 the disconnected process, so errors there are still only recorded on the
3715 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3716 connections (as happens on some IP stacks) was logged at start up time as
3717 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3718 it from "IPv6 and IPv4", which means that two separate sockets are being
3721 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3722 says whether AF_INET or AF_INET6 was passed as an argument.
3724 42. Exiwhat output was messed up when time zones were included in log
3727 43. Exiwhat now gives more information about the daemon's listening ports,
3728 and whether -tls-on-connect was used.
3730 44. The "port" option of the smtp transport is now expanded.
3732 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3733 silently ignored. Now an error message is written to the main and panic
3736 46. There's a new ACL modifier called "logwrite" which writes to a log file
3737 as soon as it is encountered.
3739 47. Added $local_user_uid and $local_user_gid at routing time.
3741 48. Exim crashed when trying to verify a sender address that was being
3744 49. Exim was recognizing only a space character after ".include". It now also
3745 recognizes a tab character.
3747 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3748 extracting the relevant information from the specification. The man page no
3749 longer contains scrambled data for the -d option, and I've added a section
3750 at the front about calling Exim under different names.
3752 51. Added "extra_headers" argument to the "mail" command in filter files.
3754 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3757 53. Installed eximstats 1.29.
3759 54. Added transport_filter_timeout as a generic transport option.
3761 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3762 Cc: header lines. This was required by RFC 822, but it not required by RFC
3765 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3766 incoming messages that did not have them. Now it does so only if the
3767 message originates locally, that is, if there is no associated remote host
3768 address. When Resent- header lines are present, this applies to the Resent-
3769 lines rather than the non-Resent- lines.
3771 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3772 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3774 58. Messages for configuration errors now include the name of the main
3775 configuration file - useful now that there may be more than one file in a
3776 list (.included file names were always shown).
3778 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3779 for those rare installations that do not start the daemon as root or run it
3780 setuid root. I've cut out the call to initgroups() if the daemon is not
3783 60. The Exim user and group can now be bound into the binary as text strings
3784 that are looked up at the start of Exim's processing.
3786 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3788 62. Added $mailstore_basename variable.
3790 63. Installed patch to sieve.c from Michael Haardt.
3792 64. When Exim failed to open the panic log after failing to open the main log,
3793 the original message it was trying to log was written to stderr and debug
3794 output, but if they were not available (the usual case in production), it
3795 was lost. Now it is written to syslog before the two lines that record the
3796 failures to open the logs.
3798 65. Users' Exim filters run in subprocesses under the user's uid. It is
3799 possible for a "deliver" command or an alias in a "personal" command to
3800 provoke an address rewrite. If logging of address rewriting is configured,
3801 this fails because the process is not running as root or exim. There may be
3802 a better way of dealing with this, but for the moment (because 4.30 needs
3803 to be released), I have disabled address rewrite logging when running a
3804 filter in a non-root, non-exim process.
3810 1. The buildconfig auxiliary program wasn't quoting the value set for
3811 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3812 not defined. This bug was masked in 4.22 by the effect that was fixed in
3815 2. Some messages that were rejected after a message id was allocated were
3816 shown as "incomplete" by exigrep. It no longer does this for messages that
3817 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3819 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3820 have allow_domain_literals set, the ID did not get logged in the <= line.
3821 Domain literals are now always recognized in Message-ID: header lines.
3823 4. The first argument for a ${extract expansion item is the key name or field
3824 number. Leading and trailing spaces in this item were not being ignored,
3825 causing some misleading effects.
3827 5. When deliver_drop_privilege was set, single queue runner processes started
3828 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3829 same command in the process it spins off) were not dropping privilege.
3831 6. When the daemon running as "exim" started a queue runner, it always
3832 re-executed Exim in the spun-off process. This is a waste of effort when
3833 deliver_drop_privilege is set. The new process now just calls the
3834 queue-runner function directly.
3840 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3843 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3844 ignored. Though the use of -oP was forcing the writing of a pid file, it
3845 was always written to the default place.
3847 3. If the message "no IP address found for host xxxx" is generated during
3848 incoming verification, it is now followed by identification of the incoming
3849 connection (so you can more easily find what provoked it).
3851 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3853 5. Added some features to "harden" Exim a bit more against certain attacks:
3855 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3856 be put in Local/Makefile. This is like the never_users runtime option,
3857 but it cannot be overridden. The default setting is "root".
3859 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3860 prefix string with which any file named in a -C command line option
3863 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3864 is retained for -C and -D only if the caller of Exim is root. Without
3865 it, the exim user may also use -C and -D and retain privilege.
3867 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3868 command line option is disabled.
3870 6. Macro names set by the -D option must start with an upper case letter, just
3871 like macro names defined in the configuration file.
3873 7. Added "dereference=" facility to LDAP.
3875 8. Two instances of the typo "uknown" in the source files are fixed.
3877 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3878 the Configure-Makefile script screwed up while processing it.
3880 10. Incorporated PCRE 4.4.
3882 11. The SMTP synchronization check was not operating right at the start of an
3883 SMTP session. For example, it could not catch a HELO sent before the client
3884 waited for the greeting. There is now a check for outstanding input at the
3885 point when the greeting is written. Because of the duplex, asynchronous
3886 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3887 way, but not yet received, when the check is performed.
3889 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3890 on TCP/IP sockets, because this apparently causes some broken clients to
3893 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3894 unchanged) from the Cygwin maintainer.
3896 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3897 was supported instead of testing for mbx. Effectively a typo.
3899 15. The spa authenticator server code was not checking that the input it
3900 received was valid base64.
3902 16. The debug output line for the "set" modifier in ACLs was not showing the
3903 name of the variable that was being set.
3905 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3907 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3908 lookup on its own. Something else was needed to trigger the lookup. For
3909 example, a match in host_lookup or the need for a host name in a host list.
3910 Now, if $sender_host_name is referenced and the host name has not yet been
3911 looked up, a lookup is performed. If the lookup fails, the variable remains
3912 empty, and $host_lookup_failed is set to "1".
3914 19. Added "eqi" as a case-independent comparison operator.
3916 20. The saslauthd authentication condition could segfault if neither service
3917 nor realm was specified.
3919 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3920 error message that was logged was misleading, and incoming SMTP
3921 connections were dropped. The message is now more accurate, and temporary
3922 errors are given to SMTP connections.
3924 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3925 (except RSET) with a 421 error, until QUIT is received. However, it was
3926 failing to send a response to QUIT.
3928 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3929 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3930 ACL is now run after the helo_try_verify_hosts code.
3932 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3933 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3934 case-independent, so other case variants are also recognized). Apparently
3935 some systems use these upper case variants.
3937 25. If more than two messages were waiting for the same host, and a transport
3938 filter was specified for the transport, Exim sent two messages over the
3939 same TCP/IP connection, and then failed with "socket operation on non-
3940 socket" when it tried to send the third.
3942 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3944 27. The extern definition of crypt16() in expand.c was not being excluded when
3945 the OS had its own crypt16() function.
3947 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3948 as a preferred synonym for return_size_limit, both as an option and as an
3951 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3953 30. Changed the default configuration ACL to relax the local part checking rule
3954 for addresses that are not in any local domains. For these addresses,
3955 slashes and pipe symbols are allowed within local parts, but the sequence
3956 /../ is explicitly forbidden.
3958 31. SPA server authentication was not clearing the challenge buffer before
3961 32. log_message in a "warn" ACL statement was writing to the reject log as
3962 well as to the main log, which contradicts the documentation and doesn't
3963 seem right (because no rejection is happening). So I have stopped it.
3965 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3966 However, I am unable to do any testing of this.
3968 34. Fixed an infelicity in the appendfile transport. When checking directories
3969 for a mailbox, to see if any needed to be created, it was accidentally
3970 using path names with one or more superfluous leading slashes; tracing
3971 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3973 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3974 contents are added to the log line that is written for every discarded
3975 recipient. (Previously a log_message setting was ignored.)
3977 36. The ${quote: operator now quotes the string if it is empty.
3979 37. The install script runs exim in order to find its version number. If for
3980 some reason other than non-existence or emptiness, which it checks, it
3981 could not run './exim', it was installing it with an empty version number,
3982 i.e. as "exim-". This error state is now caught, and the installation is
3985 38. An argument was missing from the function that creates an error message
3986 when Exim fails to connect to the socket for saslauthd authentication.
3987 This could cause Exim to crash, or give a corrupted message.
3989 39. Added isip, isip4, and isip6 to ${if conditions.
3991 40. The ACL variables $acl_xx are now saved with the message, and can be
3992 accessed later in routers, transports, and filters.
3994 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3995 strings in the file are not string-expanded.
3997 42. If a MAIL command specified a SIZE value that was too large to fit into an
3998 int variable, the check against message_size_limit failed. Such values are
3999 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4000 day this will have to be increased, but I don't think I want to be around
4001 when emails are that large.
4008 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4009 iconv() is not standard in FreeBSD.
4011 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4012 IPv6 enabled. The observed symptom was a segmentation fault on return from
4013 the function os_common_find_running_interfaces() in src/os.c.
4015 3. In the check_special_case() function in daemon.c I had used "errno" as an
4016 argument name, which causes warnings on some systems. This was basically a
4017 typo, since it was named "eno" in the comments!
4019 4. The code that waits for the clock to tick (at a resolution of some fraction
4020 of a second) so as to ensure message-id uniqueness was always waiting for
4021 at least one whole tick, when it could have waited for less. [This is
4022 almost certainly not relevant at current processor speeds, where it is
4023 unlikely to ever wait at all. But we try to future-proof.]
4025 5. The function that sleeps for a time interval that includes fractions of a
4026 second contained a race. It did not block SIGALRM between setting the
4027 timer, and suspending (a couple of lines later). If the interval was short
4028 and the sigsuspend() was delayed until after it had expired, the suspension
4029 never ended. On busy systems this could lead to processes getting stuck for
4032 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4033 process, before it forks any delivery processes. The open lookup caching
4034 mechanism meant that the open file or database connection was passed into
4035 the delivery process. The problem was that delivery processes always tidy
4036 up cached lookup data. This could cause a problem for the next delivery
4037 process started by the queue runner, because the external queue runner
4038 process does not know about the closure. So the next delivery process
4039 still has data in the lookup cache. In the case of a file lookup, there was
4040 no problem because closing a file descriptor in a subprocess doesn't affect
4041 the parent. However, if the lookup was caching a connection to a database,
4042 the connection was closed, and the second delivery process was likely to
4043 see errors such as "PGSQL: query failed: server closed the connection
4044 unexpectedly". The problem has been fixed by closing all cached lookups
4045 in a queue runner before running a delivery process.
4047 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4048 seem to have the "const" qualifier which it has on other OS. I've
4051 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4052 *of the same type* (client or server) with the same public name that an
4053 error should be diagnosed.
4055 9. When Exim looked up a host name for an IP address, but failed to find the
4056 original IP address when looking up the host name (a safety check), it
4057 output the message "<ip address> does not match any IP for NULL", which was
4058 confusing, to say the least. The bug was that the host name should have
4059 appeared instead of "NULL".
4061 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4062 user that is built into the binary, and the -C or -D options is used, root
4063 privilege is dropped before the configuration file is read. In addition,
4064 logging is switched to stderr instead of the normal log files. If the
4065 configuration then re-defines the Exim user, the unprivileged environment
4066 is probably not what is expected, so Exim logs a panic warning message (but
4069 However, if deliver_drop_privilege is set, the unprivileged state may well
4070 be exactly what is intended, so the warning has been cut out in that case,
4071 and Exim is allowed to try to write to its normal log files.
4077 1. smtp_return_error_details was not giving details for temporary sender
4078 or receiver verification errors.
4080 2. Diagnose a configuration error if two authenticators have the same public
4083 3. Exim used not to create the message log file for a message until the first
4084 delivery attempt. This could be confusing when incoming messages were held
4085 for policy or load reasons. The message log file is now created at the time
4086 the message is received, and an initial "Received" line is written to it.
4088 4. The automatically generated man page for command line options had a minor
4089 bug that caused no ill effects; however, a more serious problem was that
4090 the procedure for building the man page automatically didn't always
4091 operate. Consequently, release 4.20 contains an out-of-date version. This
4092 shouldn't happen again.
4094 5. When building Exim with embedded Perl support, the script that builds the
4095 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4096 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4098 6. The freeze_tell option was not being used for messages that were frozen on
4099 arrival, either by an ACL or by local_scan().
4101 7. Added the smtp_incomplete_transaction log selector.
4103 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4104 was accepting AUTH without a new EHLO.
4106 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4107 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4110 10. Logging of TCP/IP connections (when configured) now happens in the main
4111 daemon process instead of the child process, so that the TCP/IP connection
4112 count is more accurate (but it can never be perfect).
4114 11. The use of "drop" in a nested ACL was not being handled correctly in the
4115 outer ACL. Now, if condition failure induced by the nested "drop" causes
4116 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4117 or "require"), the connection is dropped.
4119 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4120 that yield "discard" can now be used with an "accept" or a "discard" verb,
4121 but an error is generated for any others (because I can't see a useful way
4122 to define what should happen).
4124 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4125 are now processed in the same way as lines in the Exim configuration file.
4126 In particular, continuation lines are supported.
4128 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4130 15. Added -ti meaning -t -i.
4132 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4133 domains, and warn by logging if others are found.
4135 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4136 returned from GIFCONF when Exim is trying to find the list of interfaces on
4137 a host. The code in os.c has been modified to copy each ifreq to an aligned
4138 structure in all cases.
4140 Also, in some cases, the returned ifreq's were being copied to a 'struct
4141 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4142 means the last couple of bytes of an IPv6 address could be chopped if the
4143 ifreq contained only a normal sockaddr (14 bytes storage).
4145 18. Named domain lists were not supported in the hosts_treat_as_local option.
4146 An entry such as +xxxx was not recognized, and was treated as a literal
4149 19. Ensure that header lines added by a DATA ACL are included in the reject log
4150 if the ACL subsequently rejects the message.
4152 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4153 MD5 (which is deprecated).
4155 21. When testing a filter file using -bf, Exim was writing a message when it
4156 took the sender from a "From " line in the message, but it was not doing so
4157 when it took $return_path from a Return-Path: header line. It now does.
4159 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4160 with a valid header line field name (a series of printing characters
4161 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4163 23. Changed "disc" in the source to "disk" to conform to the documentation and
4164 the book and for uniformity.
4166 24. Ignore Sendmail's -Ooption=value command line item.
4168 25. When execve() failed while trying to run a command in a pipe transport,
4169 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4170 could be confused with a return value of 69 from the command itself. This
4171 has been changed to 127, the value the shell returns if it is asked to run
4172 a non-existent command. The wording for the related log line suggests a
4173 non-existent command as the problem.
4175 26. If received_header_text expands to an empty string, do not add a Received:
4176 header line to the message. (Well, it adds a token one on the spool, but
4177 marks it "old" so that it doesn't get used or transmitted.)
4179 27. Installed eximstats 1.28 (addition of -nt option).
4181 28. There was no check for failure on the call to getsockname() in the daemon
4182 code. This can fail if there is a shortage of resources on the system, with
4183 ENOMEM, for example. A temporary error is now given on failure.
4185 29. Contrary to the C standard, it seems that in some environments, the
4186 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4187 program. Exim now does this explicitly; it affects the formatting of
4188 timestamps using strftime().
4190 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4191 complaints. I've now initialized all the variables, to avoid this.
4193 32. Header lines added by a system filter were not being "seen" during
4194 transport-time rewrites.
4196 33. The info_callback() function passed to OpenSSL is set up with type void
4197 (*)(SSL *, int, int), as described somewhere. However, when calling the
4198 function (actually a macro) that sets it up, the type void(*)() is
4199 expected. I've put in a cast to prevent warnings from picky compilers.
4201 34. If a DNS black list lookup found a CNAME record, but there were no A
4202 records associated with the domain it pointed at, Exim crashed.
4204 35. If a DNS black list lookup returned more than one A record, Exim ignored
4205 all but the first. It now scans all returned addresses if a particular IP
4206 value is being sought. In this situation, the contents of the
4207 $dnslist_value variable are a list of all the addresses, separated by a
4210 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4211 to accept a host name and all its aliases if the forward lookup for any of
4212 them yielded the IP address of the incoming connection. Now it accepts only
4213 those names whose forward lookup yields the correct IP address. Any other
4214 names are discarded. This closes a loophole whereby a rogue DNS
4215 administrator could create reverse DNS records to break through a
4216 wildcarded host restriction in an ACL.
4218 37. If a user filter or a system filter that ran in a subprocess used any of
4219 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4220 the wrong values were passed to the pipe command ($thisaddress had the
4221 value of $0, $0 had the value of $1, etc). This bug was introduced by
4222 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4224 38. Improved the line breaking for long SMTP error messages from ACLs.
4225 Previously, if there was no break point between 40 and 75 characters, Exim
4226 left the rest of the message alone. Two changes have been made: (a) I've
4227 reduced the minimum length to 35 characters; (b) if it can't find a break
4228 point between 35 and 75 characters, it looks ahead and uses the first one
4229 that it finds. This may give the occasional overlong line, but at least the
4230 remaining text gets split now.
4232 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4233 file descriptors might be low, and that setting 1000 would always raise it.
4234 It turns out that in some environments, the limit is already over 1000 and
4235 that lowering it causes trouble. So now Exim takes care not to decrease it.
4237 40. When delivering a message, the value of $return_path is set to $sender_
4238 address at the start of routing (routers may change the value). By an
4239 oversight, this default was not being set up when an address was tested by
4240 -bt or -bv, which affected the outcome if any router or filter referred to
4243 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4244 log only when "message" or "log_message" are set. However, if one of the
4245 conditions was an address verification, or a call to a nested ACL, the
4246 messages generated by the underlying test were being passed through. This
4247 no longer happens. The underlying message is available in $acl_verify_
4248 message for both "message" and "log_message" expansions, so it can be
4249 passed through if needed.
4251 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4252 new expansion $bh_ to give the encoded byte string without charset
4253 translation. Translation happens only if iconv() is available; HAVE_ICONV
4254 indicates this at build time. HEADERS_CHARSET gives the charset to
4255 translate to; headers_charset can change it in the configuration, and
4256 "headers charset" can change it in an individual filter file.
4258 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4259 that creates RFC 2047 encoded "words" labels them as that charset instead
4260 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4261 expansion operator; (ii) when Exim creates a From: line for a local
4262 message; (iii) when a header line is rewritten to include a "phrase" part.
4264 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4265 buggy, causing it to skip the first lines of messages whose message ID
4266 ended in 'D'. This would not have bitten before Exim release 4.14, because
4267 message IDs were unlikely to end in 'D' before then. The effect was to have
4268 incorrect size information for certain domains.
4270 45. #include "config.h" was missing at the start of the crypt16.c module. This
4271 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4274 46. If there was a timeout during a "random" callout check, Exim treated it as
4275 a failure of the random address, and carried on sending RSET and the real
4276 address. If the delay was just some slowness somewhere, the response to the
4277 original RCPT would be taken as a response to RSET and so on, causing
4278 mayhem of various kinds.
4280 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4281 when I implemented it. It didn't allow for the fact that some option values
4282 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4283 the right test for positive values.
4285 48. Domain names in DNS records are case-independent. Exim always looks them up
4286 in lower case. Some resolvers return domain names in exactly the case they
4287 appear in the zone file, that is, they may contain uppercase letters. Not
4288 all resolvers do this - some return always lower case. Exim was treating a
4289 change of case by a resolver as a change of domain, similar to a widening
4290 of a domain abbreviation. This triggered its re-routing code and so it was
4291 trying to route what was effectively the same domain again. This normally
4292 caused routing to fail (because the router wouldn't handle the domain
4293 twice). Now Exim checks for this case specially, and just changes the
4294 casing of the domain that it ultimately uses when it transmits the message
4297 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4300 50. If a filter generated a file delivery with a non-absolute name (possible if
4301 no home directory exists for the router), the forbid_file option was not
4304 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4305 the existing equality matching.
4307 52. Exim was using ints instead of ino_t variables in some places where it was
4308 dealing with inode numbers.
4310 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4311 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4312 called TMPDIR, and if it finds it is different, it changes its value.
4314 54. The smtp_printf() function is now made available to local_scan() so
4315 additional output lines can be written before returning. There is also an
4316 smtp_fflush() function to enable the detection of a dropped connection.
4317 The variables smtp_input and smtp_batched_input are exported to
4320 55. Changed the default runtime configuration: the message "Unknown user"
4321 has been removed from the ACL, and instead placed on the localuser router,
4322 using the cannot_route_message feature. This means that any verification
4323 failures that generate their own messages won't get overridden. Similarly,
4324 the "Unrouteable address" message that was in the ACL for unverifiable
4325 relay addresses has also been removed.
4327 56. Added hosts_avoid_esmtp to the smtp transport.
4329 57. The exicyclog script was not checking for the esoteric option
4330 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4331 will work only if exicyclog is run under the appropriate euid.
4333 58. Following a discussion on the list, the rules by which Exim recognises line
4334 endings on incoming messages have been changed. The -dropcr and drop_cr
4335 options are now no-ops, retained only for backwards compatibility. The
4336 following line terminators are recognized: LF CRLF CR. However, special
4337 processing applies to CR:
4339 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4340 nor a local message in the state where . is a terminator.
4342 (ii) If a bare CR is encountered in a header line, an extra space is added
4343 after the line terminator so as not to end the header. The reasoning
4344 behind this is that bare CRs in header lines are most likely either
4345 to be mistakes, or people trying to play silly games.
4347 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4348 was being incorrectly given as 18 bytes larger than it should have been.
4349 This is a VOB (very old bug).
4351 60. This may never have affected anything current, but just in case it has:
4352 When the local host is found other than at the start of a list of hosts,
4353 the local host, those with the same MX, and any that follow, are discarded.
4354 When the list in question was part of a longer list of hosts, the following
4355 hosts (not currently being processed) were also being discarded. This no
4356 longer happens. I'm not sure if this situation could ever has previously
4359 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4362 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4363 header. This is something that is recommended in a new Internet Draft, and
4364 is something that is documented as being done by Sendmail. There are two
4365 possible values. For messages generated by the autoreply transport, Exim
4368 Auto-Submitted: auto-replied
4370 whereas for all other generated messages (e.g. bounces) it adds
4372 Auto-Submitted: auto-generated
4374 63. The "personal" condition in filters now includes a test for the
4375 Auto-Submitted: header. If it contains the string "auto-" the message it
4376 not considered personal.
4378 64. Added rcpt_include_affixes as a generic transport option.
4380 65. Added queue_only_override (default true).
4382 66. Added the syslog_duplication option.
4384 67. If what should have been the first header line of a message consisted of
4385 a space followed by a colon, Exim was mis-interpreting it as a header line.
4386 It isn't of course - it is syntactically invalid and should therefore be
4387 treated as the start of the message body. The misbehaviour could have
4388 caused a number of strange effects, including loss of data in subsequent
4389 header lines, and spool format errors.
4391 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4392 client host had authenticated. This control can now be exercised by an ACL
4393 for more flexibility.
4395 69. By default, callouts do not happen when testing with -bh. There is now a
4396 variant, -bhc, which does actually run the callout code, including
4397 consulting and updating the callout cache.
4399 70. Added support for saslauthd authentication, courtesy of Alexander
4402 71. If statvfs() failed on the spool or log directories while checking their
4403 size for availability, Exim confusingly gave the error "space shortage".
4404 Furthermore, in debugging mode it crashed with a floating point exception.
4405 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4406 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4407 serious problem, Exim now writes to the main and panic logs when this
4408 happens, with details of the failure. It then refuses to accept the
4409 incoming message, giving the message "spool directory problem" or "log
4410 directory problem" with a 421 code for SMTP messages.
4412 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4413 0, 1, and 2 exist, because some OS complain for execs without them (see
4414 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4415 descriptors. However, the code omitted to check that the open succeeded,
4416 causing mysterious errors if for some reason the permissions on /dev/null
4417 got screwed. Now Exim writes a message to the main and panic logs, and
4418 bombs out if it can't open /dev/null.
4420 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4421 interact so that it is all more flexible. It is supposed to remain
4422 backwards compatible. Also added extra_local_interfaces.
4424 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4425 to bomb out with an assertion failure - to the client this appears as a
4426 connection drop. This problem occurs in the part of the code that was taken
4427 from the Samba project. Fortunately, the assertion is in a very simple
4428 function, so I have fixed this by reproducing the function inline in the
4429 one place where it is called, and arranging for authentication to fail
4430 instead of killing the process with assert().
4432 75. The SPA client code was not working when the server requested OEM rather
4433 than Unicode encoding.
4435 76. Added code to make require_files with a specific uid setting more usable in
4436 the case where statting the file as root fails - usually a non-root-mounted
4437 NFS file system. When this happens and the failure is EACCES, Exim now
4438 forks a subprocess and does the per-uid checking as the relevant uid.
4440 77. Added process_log_path.
4442 78. If log_file_path was not explicitly set, a setting of check_log_space or
4443 check_log_inodes was ignored.
4445 79. If a space check for the spool or log partitions fails, the incident is now
4446 logged. Of course, in the latter case the data may get lost...
4448 80. Added the %p formatting code to string_format() so that it can be used to
4449 print addresses in debug_print(). Adjusted all the address printing in the
4450 debugging in store.c to use %p rather than %d.
4452 81. There was a concern that a line of code in smtp_in.c could overflow a
4453 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4454 initially expressed, the concern was not well-founded, because trailing
4455 spaces are removed early. However, if the trailing spaces were followed by
4456 a NULL, they did not get removed, so the overflow was possible. Two fixes
4459 (a) I re-wrote the offending code in a cleaner fashion.
4460 (b) If an incoming SMTP command contains a NULL character, it is rejected
4463 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4464 runs initgroups(), so that if the Exim user is in any additional groups,
4465 they will be used during message reception.
4471 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob