1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.435 2006/11/16 16:21:58 steve 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.
316 SC/01 Use a glob alias rather than an array ref in eximstats generated
317 parser. This improves both readability and performance.
319 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
320 Don't display local sender or destination tables in eximstats unless
321 there is data to show.
322 Added average volumes into the eximstats top table text output.
324 SC/03 Collect data on the number of addresses (recipients) as well
325 as the number of messages in eximstats.
327 TF/01 Correct an error in the documentation for the redirect router. Exim
328 does (usually) call initgroups() when daemonizing.
330 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
331 with consistent privilege compared to when running as a daemon.
333 TF/03 Note in the spec that $authenticated_id is not set for local
334 submissions from trusted users.
336 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
337 Thanks to Dean Brooks <dean@iglou.com> for the patch.
339 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
340 by adding some example configuration directives to the default
341 configuration file. A little bit of work is required to uncomment the
342 directives and define how usernames and passwords are checked, but
343 there is now a framework to start from.
345 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
346 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
347 without this. I don't know how relevant this is to other LDAP libraries.
349 PH/02 Add the verb name to the "unknown ACL verb" error.
351 PH/03 Magnus Holmgren's patch for filter_prepend_home.
353 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
355 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
356 directory not expanded when it should be if an expanded home directory
357 was set for the address (which is overridden by the transport).
359 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
362 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
363 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
364 because it is too late at that time, and has no effect.
366 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
367 security issue with \' (bugzilla #107). I could not use the
368 PQescapeStringConn() function, because it needs a PGconn value as one of
371 PH/08 When testing addresses using -bt, indicate those final addresses that
372 are duplicates that would not cause an additional delivery. At least one
373 person was confused, thinking that -bt output corresponded to deliveries.
374 (Suppressing duplicates isn't a good idea as you lose the information
375 about possibly different redirections that led to the duplicates.)
377 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
378 systems where poll() doesn't work, in particular OS X.
380 PH/10 Added more information to debugging output for retry time not reached.
382 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
383 operations in malware.c.
385 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
388 PH/13 If write_rejectlog was set false when logging was sent to syslog with
389 syslog_duplication set false, log lines that would normally be written
390 both the the main log and to the reject log were not written to syslog at
393 PH/14 In the default configuration, change the use of "message" in ACL warn
394 statements to "add_header".
396 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
397 not followed by a command (e.g. "seen endif").
399 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
400 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
403 PH/17 Added extra conditions to the default value of delay_warning_condition
407 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
408 { match{$h_precedence:}{(?i)bulk|list|junk} } \
409 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
412 The Auto-Submitted: and various List- headers are standardised, whereas I
413 don't think Precedence: ever was.
415 PH/18 Refactored debugging code in route_finduser() to show more information,
416 in particular, the error code if getpwnam() issues one.
418 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
419 This is apparently needed in addition to the PH/07 change above to avoid
420 any possible encoding problems.
422 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
423 but not after initializing Perl.
425 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
426 output them only if debugging. By default they are written stderr,
427 apparently, which is not desirable.
429 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
432 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
435 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
437 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
438 authenticated or an ident call has been made. Suppress the default
439 values for $authenticated_id and $authenticated_sender (but permit -oMai
440 and -oMas) when testing with -bh.
442 PH/24 Re-jigged the order of the tests in the default configuration so that the
443 tests for valid domains and recipients precede the DNS black list and CSA
444 tests, on the grounds that those ones are more expensive.
446 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
447 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
448 This bug exists in every version of Exim that I still have, right back to
451 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
452 However, an attempt to turn on case-sensitivity in a regex key by
453 including (?-i) didn't work because the subject string was already
454 lowercased, and the effects were non-intuitive. It turns out that a
455 one-line patch can be used to allow (?-i) to work as expected.
461 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
462 other effects) broke the use of negated acl sub-conditions.
464 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
467 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
468 "Deny" causes Exim to reject the incoming connection with a 554 error.
469 Unfortunately, if there is a major crisis, such as a disk failure,
470 tcp-wrappers gives "deny", whereas what one would like would be some
471 kind of temporary error. A kludge has been added to help with this.
472 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
473 554 error is used if errno is still zero or contains ENOENT (which occurs
474 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
477 PH/03 Add -lutil to the default FreeBSD LIBS setting.
479 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
480 errors. Otherwise a message that provokes a temporary error (when other
481 messages do not) can cause a whole host to time out.
483 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
484 addresses were routed directly to files or pipes from a redirect router.
485 File deliveries just didn't batch; pipe deliveries might have suffered
488 PH/06 A failure to get a lock for a hints database would erroneously always say
489 "Failed to get write lock", even when it was really a read lock.
491 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
492 of 0600. This has been changed to use the value of the lockfile_mode
493 option (which defaults to 0600).
495 PH/08 Applied small patch from the Sieve maintainer.
497 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
498 folder from quota calculations, a direct delivery into this folder messed
499 up the contents of the maildirsize file. This was because the regex was
500 used only to exclude .Trash (or whatever) when the size of the mailbox
501 was calculated. There was no check that a delivery was happening into an
502 excluded directory. This bug has been fixed by ignoring all quota
503 processing for deliveries into excluded directories.
505 PH/10 Added the maildirfolder_create_regex option to appendfile.
511 PH/01 The code for finding all the local interface addresses on a FreeBSD
512 system running IPv6 was broken. This may well have applied to all BSD
513 systems, as well as to others that have similar system calls. The broken
514 code found IPv4 interfaces correctly, but gave incorrect values for the
515 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
516 that it would not match correctly against @[] and not recognize the IPv6
519 PH/02 The ipliteral router was not recognizing addresses of the form user@
520 [ipv6:....] because it didn't know about the "ipv6:" prefix.
522 PH/03 Added disable_ipv6.
524 PH/04 Changed $reply_address to use the raw form of the headers instead of the
525 decoded form, because it is most often used to construct To: headers
526 lines in autoreplies, and the decoded form may well be syntactically
527 invalid. However, $reply_address has leading white space removed, and all
528 newlines turned into spaces so that the autoreply transport does not
531 PH/05 If group was specified without a user on a router, and no group or user
532 was specified on a transport, the group from the router was ignored.
534 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
535 for visible compile-time settings that can be used to change these
536 numbers, for those that want even more. Backwards compatibility with old
537 spool files has been maintained. However, going back to a previous Exim
538 release will lost any variables that are in spool files.
540 PH/07 Two small changes when running in the test harness: increase delay when
541 passing a TCP/IP connection to a new process, in case the original
542 process has to generate a bounce, and remove special handling of
543 127.0.0.2 (sic), which is no longer necessary.
545 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
546 be the same on different OS.
548 PH/09 Moved a debug statement in filter processing to avoid a race problem when
551 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
552 whether --show-vars was specified or not
554 JJ/02 exipick: Added support for new ACL variable spool format introduced
557 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
558 syntactically invalid From: or Reply-to: line, and a filter used this to
559 generate an autoreply, and therefore failed to obtain an address for the
560 autoreply, Exim could try to deliver to a non-existent relative file
561 name, causing unrelated and misleading errors. What now happens is that
562 it logs this as a hard delivery error, but does not attempt to create a
565 PH/11 The exinext utility has a -C option for testing purposes, but although
566 the given file was scanned by exinext itself; it wasn't being passed on
567 when Exim was called.
569 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
570 an end-of-file indication when reading a command response.
572 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
573 compiled. In many other places in Exim, IPv6 addresses are always
574 recognized, so I have changed this. It also means that IPv4 domain
575 literals of the form [IPV4:n.n.n.n] are now always recognized.
577 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
578 used if the router is not running as root, for example, when verifying at
579 ACL time, or when using -bh. The debugging output from this situation was
580 non-existent - all you got was a failure to exec. I have made two
583 (a) Failures to set uid/gid, the current directory, or a process leader
584 in a subprocess such as that created by queryprogram now generate
585 suitable debugging ouput when -d is set.
587 (b) The queryprogram router detects when it is not running as root,
588 outputs suitable debugging information if -d is set, and then runs
589 the subprocess without attempting to change uid/gid.
591 PH/15 Minor change to Makefile for building test_host (undocumented testing
594 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
595 additional section of a DNS packet that returns MX or SRV records.
596 Instead, it always explicitly searches for A/AAAA records. This avoids
597 major problems that occur when a DNS server includes only records of one
598 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
599 fixed another bug: if SRV records were looked up and the corresponding
600 address records were *not* found in the additional section, the port
601 values from the SRV records were lost.
603 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
604 using the correct key (the original address) when searching the retry
605 rules in order to find which one to use for generating the retry hint.
607 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
608 adding the default one. Similarly, if it contains a Reply-To: header, the
609 errors_reply_to option, if set, is not used.
611 PH/19 When calculating a retry time, Exim used to measure the "time since
612 failure" by looking at the "first failed" field in the retry record. Now
613 it does not use this if it is later than than the arrival time of the
614 message. Instead it uses the arrival time. This makes for better
615 behaviour in cases where some deliveries succeed, thus re-setting the
616 "first failed" field. An example is a quota failure for a huge message
617 when small messages continue to be delivered. Without this change, the
618 "time since failure" will always be short, possible causing more frequent
619 delivery attempts for the huge message than are intended.
620 [Note: This change was subsequently modified - see PH/04 for 4.62.]
622 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
623 $1, $2, $3) because the numerical variables can be reset during some
624 expansion items (e.g. "match"), thereby losing the authentication data.
626 PH/21 Make -bV show the size of off_t variables so that the test suite can
627 decide whether to run tests for quotas > 2G.
629 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
630 mailbox_size, and mailbox_filecount in the appendfile transport. If a
631 filecount value is greater than 2G or if a quota value is greater than 2G
632 on a system where the size of off_t is not greater than 4, a panic error
635 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
636 never match. The debug and -bh output now contains an explicit error
637 message indicating a malformed IPv4 address or mask.
639 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
640 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
643 PH/25 Do not write to syslog when running in the test harness. The only
644 occasion when this arises is a failure to open the main or panic logs
645 (for which there is an explicit test).
647 PH/26 Added the /no_tell option to "control=freeze".
649 PH/27 If a host name lookup failed very early in a connection, for example, if
650 the IP address matched host_lookup and the reverse lookup yielded a name
651 that did not have a forward lookup, an error message of the form "no IP
652 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
653 could be logged. Now it outputs the IP address instead of "NULL".
655 PH/28 An enabling patch from MH: add new function child_open_exim2() which
656 allows the sender and the authenticated sender to be set when
657 submitting a message from within Exim. Since child_open_exim() is
658 documented for local_scan(), the new function should be too.
660 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
661 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
662 results in an empty string is now treated as unset.
664 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
666 PH/31 Added sender_verify_fail logging option.
668 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
669 needed by RFC 822 but not by RFC 2822 was commented out. I have now
670 tidied the source and removed it altogether.
672 PH/33 When a queue run was abandoned because the load average was too high, a
673 log line was always written; now it is written only if the queue_run log
674 selector is set. In addition, the log line for abandonment now contains
675 information about the queue run such as the pid. This is always present
676 in "start" and "stop" lines but was omitted from the "abandon" line.
678 PH/34 Omit spaces between a header name and the colon in the error message that
679 is given when verify = headers_syntax fails (if there are lots of them,
680 the message gets confusing).
682 PH/35 Change the default for dns_check_names_pattern to allow slashes within
683 names, as there are now some PTR records that contain slashes. This check
684 is only to protect against broken name servers that fall over on strange
685 characters, so the fact that it applies to all lookups doesn't matter.
687 PH/36 Now that the new test suite is complete, we can remove some of the
688 special code in Exim that was needed for the old test suite. For example,
689 sorting DNS records because real resolvers return them in an arbitrary
690 order. The new test suite's fake resolver always returns records in the
693 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
694 bounces) except when queue_only is set, to avoid logging races between
695 the different processes.
697 PH/38 Panic-die if .include specifies a non-absolute path.
699 PH/39 A tweak to the "H" retry rule from its user.
701 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
702 a label. They prevented compilation on older perls.
704 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
705 a warning to be raised on newish perls.
707 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
708 on queue. Changes to match documented behaviour of showing count of
709 messages matching specified criteria.
711 PH/40 Changed the default ident timeout from 30s to 5s.
713 PH/41 Added support for the use of login_cap features, on those BSD systems
714 that have them, for controlling the resources used by pipe deliveries.
716 PH/42 The content-scanning code uses fopen() to create files in which to put
717 message data. Previously it was not paying any attention to the mode of
718 the files. Exim runs with umask(0) because the rest of the code creates
719 files with open(), and sets the required mode explicitly. Thus, these
720 files were ending up world-writeable. This was not a big issue, because,
721 being within the spool directory, they were not world-accessible. I have
722 created a function called modefopen, which takes an additional mode
723 argument. It sets umask(777), creates the file, chmods it to the required
724 mode, then resets the umask. All the relevant calls to fopen() in the
725 content scanning code have been changed to use this function.
727 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
728 to 24 hours. This avoids potential overflow problems when processing G
729 and H retry rules. I suspect nobody ever tinkers with this value.
731 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
733 PH/45 When the plaintext authenticator is running as a client, the server's
734 challenges are checked to ensure they are valid base64 strings. By
735 default, the authentication attempt is cancelled if an invalid string is
736 received. Setting client_ignore_invalid_base64 true ignores these errors.
737 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
738 they are received. Thus, the responses can be made to depend on the
739 challenges. If an invalid string is ignored, an empty string is placed in
742 PH/46 Messages that are created by the autoreply transport now contains a
743 References: header, in accordance with RFCs 2822 and 3834.
745 PH/47 Added authenticated_sender_force to the smtp transport.
747 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
749 PH/49 Installed latest patch from the Sieve maintainer.
751 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
752 was also set, the appendfile transport was unnecessarily scanning a
753 directory of message files (e.g. for maildir delivery) to find the count
754 of files (along with the size), even though it did not need this
755 information. It now does the scan only if it needs to find either the
756 size of the count of files.
758 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
760 PH/52 Two bugs concerned with error handling when the smtp transport is
763 (i) Exim was not creating retry information for temporary errors given
764 for individual recipients after the DATA command when the smtp transport
765 was used in LMTP mode. This meant that they could be retried too
766 frequently, and not timed out correctly.
768 (ii) Exim was setting the flag that allows error details to be returned
769 for LMTP errors on RCPT commands, but not for LMTP errors for individual
770 recipients that were returned after the DATA command.
772 PH/53 This is related to PH/52, but is more general: for any failing address,
773 when detailed error information was permitted to be returned to the
774 sender, but the error was temporary, then after the final timeout, only
775 "retry timeout exceeded" was returned. Now it returns the full error as
776 well as "retry timeout exceeded".
778 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
779 do this, and (what is worse) MTAs that accept it.
781 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
782 will now be deprecated.
784 PH/56 New os.c-cygwin from the Cygwin maintainer.
786 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
787 formats (previously only available in exim formats via -bpr, -bpru,
788 and -bpra. Now also available in native and exiqgrep formats)
790 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
791 with very large, slow to parse queues
793 JJ/08 exipick: added ! as generic prefix to negate any criteria format
795 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
797 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
798 responses to authentication challenges, though it was showing the
799 challenges; (ii) I've removed the CR characters from the debug output for
802 PH/58 Allow for the insertion of a newline as well as a space when a string
803 is turned into more than one encoded-word during RFC 2047 encoding. The
804 Sieve code now uses this.
806 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
807 data_4xx, lost_connection, tls_required.
809 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
810 message was being sent as an SMTP response.
812 PH/61 Add -l and -k options to exicyclog.
814 PH/62 When verifying, if an address was redirected to one new address, so that
815 verification continued, and the new address failed or deferred after
816 having set something in $address_data, the value of $address_data was not
817 passed back to the ACL. This was different to the case when no
818 redirection occurred. The value is now passed back in both cases.
820 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
821 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
822 use login_cap.h, so on its own it isn't the distinguishing feature. The
823 new name refers directly to the setclassresources() function.
825 PH/65 Added configuration files for NetBSD3.
827 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
829 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
830 is preferred over IPv4.
832 PH/68 The bounce_return_message and bounce_return_body options were not being
833 honoured for bounces generated during the reception of non-SMTP messages.
834 In particular, this applied to messages rejected by the ACL. This bug has
835 been fixed. However, if bounce_return_message is true and bounce_return_
836 body is false, the headers that are returned for a non-SMTP message
837 include only those that have been read before the error was detected.
838 (In the case of an ACL rejection, they have all been read.)
840 PH/69 The HTML version of the specification is now built in a directory called
841 spec_html instead of spec.html, because the latter looks like a path with
842 a MIME-type, and this confuses some software.
844 PH/70 Catch two compiler warnings in sieve.c.
846 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
847 function verify_get_ident() calls ip_connect() to connect a socket, but
848 if the "connect()" function timed out, ip_connect() used to close the
849 socket. However, verify_get_ident() also closes the socket later, and in
850 between Exim writes to the log, which may get opened at this point. When
851 the socket was closed in ip_connect(), the log could get the same file
852 descriptor number as the socket. This naturally causes chaos. The fix is
853 not to close the socket in ip_connect(); the socket should be closed by
854 the function that creates it. There was only one place in the code where
855 this was missing, in the iplookup router, which I don't think anybody now
856 uses, but I've fixed it anyway.
858 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
859 well as to direct DNS lookups. Otherwise the handling of names in host
860 lists is inconsistent and therefore confusing.
866 PH/01 Two changes to the default runtime configuration:
868 (1) Move the checks for relay_from_hosts and authenticated clients from
869 after to before the (commented out) DNS black list checks.
871 (2) Add control=submission to the relay_from_hosts and authenticated
872 clients checks, on the grounds that messages accepted by these
873 statements are most likely to be submissions.
875 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
877 (1) Generate an error if the third argument for the ${prvs expansion is
880 (2) Treat a missing third argument of ${prvscheck as if it were an empty
883 (3) Reset the variables that are obtained from the first argument of
884 ${prvscheck and used in the second argument before leaving the code,
885 because their memory is reclaimed, so using them afterwards may do
888 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
889 one (it's much easier than Tom thought :-).
891 (5) Because of (4), we can now allow for the use of $prvscheck_result
892 inside the third argument.
894 PH/03 For some reason, the default setting of PATH when running a command from
895 a pipe transport was just "/usr/bin". I have changed it to
898 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
899 anything to be listed in the output from -bV.
901 PH/05 When a filter generated an autoreply, the entire To: header line was
902 quoted in the delivery log line, like this:
904 => >A.N.Other <ano@some.domain> <original@ddress> ...
906 This has been changed so that it extracts the operative address. There
907 may be more than one such address. If so, they are comma-separated, like
910 => >ano@some.domain,ona@other.domain <original@ddress> ...
912 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
913 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
914 not being looked up in the rDNS to get a host name, Exim was showing the
915 IP address twice in Received: lines, even though the IP addresses were
916 identical. For example:
918 Received: from [1.2.3.4] (helo=[1.2.3.4])
920 However, if the real host name was known, it was omitting the HELO data
921 if it matched the actual IP address. This has been tidied up so that it
922 doesn't show the same IP address twice.
924 PH/07 When both +timestamp and +memory debugging was on, the value given by
925 $tod_xxx expansions could be wrong, because the tod_stamp() function was
926 called by the debug printing, thereby overwriting the timestamp buffer.
927 Debugging no longer uses the tod_stamp() function when +timestamp is set.
929 PH/08 When the original message was included in an autoreply transport, it
930 always said "this is a copy of the message, including all the headers",
931 even if body_only or headers_only was set. It now gives an appropriate
934 PH/09 Applied a patch from the Sieve maintainer which:
936 o fixes some comments
937 o adds the (disabled) notify extension core
938 o adds some debug output for the result of if/elsif tests
939 o points to the current vacation draft in the documentation
940 and documents the missing references header update
944 o fixes a bug in processing the envelope test (when testing
945 multiple envelope elements, the last element determinted the
948 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
949 Electronic Mail") by including:
951 Auto-submitted: auto-generated
953 in the messages that it generates (bounce messages and others, such as
954 warnings). In the case of bounce messages for non-SMTP mesages, there was
955 also a typo: it was using "Auto_submitted" (underscore instead of
956 hyphen). Since every message generated by Exim is necessarily in response
957 to another message, thes have all been changed to:
959 Auto-Submitted: auto-replied
961 in accordance with these statements in the RFC:
963 The auto-replied keyword:
965 - SHOULD be used on messages sent in direct response to another
966 message by an automatic process,
968 - MUST NOT be used on manually-generated messages,
970 - MAY be used on Delivery Status Notifications (DSNs) and Message
971 Disposition Notifications (MDNs),
973 - MUST NOT be used on messages generated by automatic or periodic
974 processes, except for messages which are automatic responses to
977 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
978 to the default Received: header definition.
980 PH/12 Added log selector acl_warn_skipped (default on).
982 PH/13 After a successful wildlsearch lookup, discard the values of numeric
983 variables because (a) they are in the wrong storage pool and (b) even if
984 they were copied, it wouldn't work properly because of the caching.
986 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
987 checking when decoding. Apparently there are clients that generate
988 overlong encoded strings. Why am I not surprised?
990 PH/15 If the first argument of "${if match_address" was not empty, but did not
991 contain an "@" character, Exim crashed. Now it writes a panic log message
992 and treats the condition as false.
994 PH/16 In autoreply, treat an empty string for "once" the same as unset.
996 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
997 extension "envelope-auth". The code is finished and in agreement with
998 other implementations, but there is no documentation so far and in fact,
999 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1000 not changing the active code.
1002 Print executed "if" and "elsif" statements when debugging is used. This
1003 helps a great deal to understand what a filter does.
1005 Document more things not specified clearly in RFC3028. I had all this
1006 sorted out, when out of a sudden new issues came to my mind. Oops."
1008 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1011 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1012 canonical form (as documented). However, after a host name lookup from
1013 the IP address, check_host() was doing a simple string comparison with
1014 addresses acquired from the DNS when checking that the found name did
1015 have the original IP as one of its addresses. Since any found IPv6
1016 addresses are likely to be in abbreviated form, the comparison could
1017 fail. Luckily, there already exists a function for doing the comparison
1018 by converting both addresses to binary, so now that is used instead of
1019 the text comparison.
1021 PH/20 There was another similar case to PH/19, when a complete host name was
1022 given in a host list; looking up its IP address could give an abbreviated
1023 form, whereas the current host's name might or might not be abbreviated.
1024 The same fix has been applied.
1030 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1031 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1034 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1035 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1037 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1039 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1040 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1041 Also, for "Auto-Submitted", treat anything other than "no" as
1042 non-personal, in accordance with RFC 3834. (Previously it treated
1043 anything starting "auto-" as non-personal.)
1045 TF/01 The control=submission/name=... option had a problem with syntax
1046 errors if the name included a slash character. The /name= option
1047 now slurps the rest of the string, so it can include any characters
1048 but it must come last in the list of options (after /sender_retain
1051 PH/05 Some modifications to the interface to the fake nameserver for the new
1059 TK/01 Added the "success_on_redirect" address verification option. See
1060 NewStuff for rationale and an example.
1062 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1064 PH/02 Patch to exigrep to allow it to work on syslog lines.
1066 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1067 fread() to skip over the body file's header line, because in Cygwin the
1068 header line is locked and is inaccessible.
1070 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1071 co-exist for some time) to make it clear that it is the Exim ID that is
1072 referenced, not the Message-ID: header line.
1074 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1075 string_format() function, because snprintf() does not exist on all
1078 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1079 ${stat: expansion item.
1081 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1082 protocol synchronization error", to keep the pedants happy.
1084 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1085 well as for IRIX systems, when gcc is being used. See the host.c source
1088 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1090 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1093 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1094 transport and to the smtp transport in LMTP mode.
1096 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1098 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1099 run a filter in a subprocess. This could lead to confusion in subsequent
1100 lookups in the parent process. There should also be a search_tidyup() at
1101 the end of the subprocess.
1103 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1104 only if the host matched helo_try_verify_hosts, which caused the
1105 verification to occur when the EHLO/HELO command was issued. The ACL just
1106 tested the remembered result. Now, if a previous verification attempt has
1107 not happened, "verify = helo" does it there and then.
1109 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1111 TK/03 Fix log output including CR from clamd.
1113 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1114 exist provoked a memory error which could cause a segfault.
1116 PH/15 Installed PCRE 6.2
1118 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1120 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1121 of the problem. Specifically, suggested +O2 rather than +O1 for the
1124 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1126 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1127 record, the port was not correctly logged when the outgoing_port log
1128 selector was set (it logged the transort's default port).
1130 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1131 fallback_hosts, and "hosts" in the smtp transport.
1133 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1134 host errors such as "Connection refused".
1136 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1137 authentication with radiusclient 0.4.9:
1139 - Error returned from rc_read_config was caught wrongly
1140 - Username/password not passed on to radius server due to wrong length.
1142 The presumption is that some radiusclient API changes for 4.51/PH/17
1143 were not taken care of correctly. The code is still untested by me (my
1144 Linux distribution still has 0.3.2 of radiusclient), but it was
1145 contributed by a Radius user.
1147 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1148 expanding the "port" option of the smtp transport.
1150 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1151 while reading a MIME header. Thanks to Tom Hughes for a patch.
1153 PH/24 Include config.h inside local_scan.h so that configuration settings are
1156 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1157 that in an ACL for RCPT (for example), you can examine exactly what was
1160 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1161 commands, but it was not correctly comparing the address with the actual
1162 client host address. Thus, it would show the EHLO address in Received:
1163 header lines when this was not necessary.
1165 PH/27 Added the % operator to ${eval:}.
1167 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1168 it should be ignoring failures (because with -C, for example, it has lost
1169 privilege). It wasn't ignoring creation failures other than "already
1172 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1175 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1176 because an input line was too long, either on its own, or by virtue of
1177 too many continuations, the temporary file was not being removed, and the
1178 return code was incorrect.
1180 PH/31 Missing "BOOL" in function definition in filtertest.c.
1182 PH/32 Applied Sieve patches from the maintainer.
1184 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1186 PH/33 Added "verify = not_blind".
1188 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1189 Local/Makefile (with some defaults set). These are used in built scripts
1190 such as exicyclog, but they have never been used in the exim_install
1191 script (though there are many overriding facilities there). I have
1192 arranged that the exim_install script now takes note of these two
1195 PH/35 Installed configuration files for Dragonfly.
1197 PH/36 When a locally submitted message by a trusted user did not contain a
1198 From: header, and the sender address was obtained from -f or from an SMTP
1199 MAIL command, and the trusted user did not use -F to supply a sender
1200 name, $originator_name was incorrectly used when constructing a From:
1201 header. Furthermore, $originator_name was used for submission mode
1202 messages from external hosts without From: headers in a similar way,
1203 which is clearly wrong.
1205 PH/37 Added control=suppress_local_fixups.
1207 PH/38 When log_selector = +received_sender was set, and the addition of the
1208 sender made the log line's construction buffer exactly full, or one byte
1209 less than full, an overflow happened when the terminating "\n" was
1212 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1213 when the result of a list match is failure because a DNS lookup failed.
1215 PH/40 RM_COMMAND is now used in the building process.
1217 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1218 the "build-* directories that it finds.
1220 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1221 address in a domain literal was a prefix of an interface address.
1223 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1224 when verifying a sender address, unless rewrite_headers is false.
1226 PH/44 Wrote a long comment about why errors_to addresses are verified as
1227 recipients, not senders.
1229 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1230 the ratelimit ACL was added.
1232 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1234 PH/46 Added extra information about PostgreSQL errors to the error string.
1236 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1237 avoiding the need to install special zones in a real server. This is
1238 backwards compatible; if it can't find the fake resolver, it drops back.
1239 Thus, both old and new test suites can be run.
1241 TF/02 Added util/ratelimit.pl
1243 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1244 clock is set back in time.
1246 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1247 Candler <B.Candler@pobox.com>.
1249 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1250 OK for addresses that are the result of redirections.
1252 PH/48 A number of further additions for the benefit of the new test suite,
1253 including a fake gethostbyname() that interfaces to the fake DNS resolver
1256 TF/06 The fix for widen_domains has also been applied to qualify_single and
1257 search_parents which are the other dnslookup options that can cause
1260 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1263 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1265 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1266 service" if it encountered an MX record with an empty target hostname.
1267 The message is now "an MX or SRV record indicated no SMTP service".
1269 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1270 if the DNS of the sending site is misconfigured. This is quite a
1271 common situation. This change restores the behaviour of treating a
1272 helo verification defer as a failure.
1274 PH/51 If self=fail was set on a router, the bounce message did not include the
1275 actual error message.
1281 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1283 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1284 command itself ended in error, the underlying message about the transport
1285 filter timeout was being overwritten with the pipe command error. Now the
1286 underlying error message should be appended to the second error message.
1288 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1290 PH/02 Reduce the amount of output that "make" produces by default. Full output
1291 can still be requested.
1293 PH/03 The warning log line about a condition test deferring for a "warn" verb
1294 was being output only once per connection, rather than after each
1295 occurrence (because it was using the same function as for successful
1296 "warn" verbs). This seems wrong, so I have changed it.
1298 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1299 it should not have, which might have caused a crash in the right
1300 circumstances, but probably never did.
1302 PH/04 Installed a modified version of Tony Finch's patch to make submission
1303 mode fix the return path as well as the Sender: header line, and to
1304 add a /name= option so that you can make the user's friendly name appear
1307 TF/03 Added the control = fakedefer ACL modifier.
1309 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1310 Mark Lowes for thorough testing.
1312 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1314 TK/03 Merged latest SRS patch from Miles Wilton.
1316 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1317 with the definition in sysexits.h (which is #included earlier).
1318 Fortunately, Exim does not actually use EX_OK. The code used to try to
1319 preserve the sysexits.h value, by assumimg that macro definitions were
1320 scanned for macro replacements. I have been disabused of this notion,
1321 so now the code just undefines EX_OK before #including unistd.h.
1323 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1324 in the smtp transport. When a block could not be written in a single
1325 write() function, the timeout was being re-applied to each part-write.
1326 This seems wrong - if the receiver was accepting one byte at a time it
1327 would take for ever. The timeout is now adjusted when this happens. It
1328 doesn't have to be particularly precise.
1330 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1331 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1333 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1334 without a domain if the check to <postmaster@domain> fails.
1336 SC/01 Eximstats: added -xls and the ability to specify output files
1337 (patch written by Frank Heydlauf).
1339 SC/02 Eximstats: use FileHandles for outputing results.
1341 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1343 SC/04 Eximstats: fixed display of large numbers with -nvr option
1345 SC/05 Eximstats: fixed merging of reports with empty tables.
1347 SC/06 Eximstats: added the -include_original_destination flag
1349 SC/07 Eximstats: removed tabs and trailing whitespace.
1351 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1353 TK/06 MBOX spool code: Add real "From " MBOX separator line
1354 so the .eml file is really in mbox format (even though
1355 most programs do not really care). Patch from Alex Miller.
1357 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1358 The latter is generated from $received_to and is only set if the
1359 message has one envelope recipient. SA can use these headers,
1360 obviously out-of-the-box. Patch from Alex Miller.
1362 PH/08 The ${def test on a variable was returning false if the variable's
1363 value was "0", contrary to what the specification has always said!
1364 The result should be true unless the variable is empty.
1366 PH/09 The syntax error of a character other than { following "${if
1367 def:variable_name" (after optional whitespace) was not being diagnosed.
1368 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1369 accidental colon was present, for example, could give incorrect results.
1371 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1372 result is used (not including appendfile, where other changes are about
1375 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1376 This involved changing a lot of size variables from int to off_t. It
1377 should work with maildirs and everything.
1379 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1380 spamd dying while we are connected to it.
1382 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1385 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1386 function for BDB 4.3.
1388 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1390 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1391 the sources for PCRE have been re-arranged and more files are now
1394 PH/15 The code I had for printing potentially long long variables in PH/11
1395 above was not the best (it lost precision). The length of off_t variables
1396 is now inspected at build time, and an appropriate printing format (%ld
1397 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1398 to be "long long int" or "long int". This is needed for the internal
1399 formatting function string_vformat().
1401 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1402 the configuration file to be ":syslog", then the script "guesses" where
1403 the logs files are, rather than using the compiled in default. In our
1404 case the guess is not the same as the compiled default, so the script
1405 suddenly stopped working when I started to use syslog. The patch checks
1406 to see if log_file_path is "". If so, it attempts to read it from exim
1407 with no configuration file to get the compiled in version, before it
1408 falls back to the previous guessing code."
1410 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1411 implementing BATV in an Exim configuration. See NewStuff for the gory
1414 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1415 Makefile that are specific to HP-UX.
1417 PH/18 If the "use_postmaster" option was set for a recipient callout together
1418 with the "random" option, the postmaster address was used as the MAIL
1419 FROM address for the random test, but not for the subsequent recipient
1420 test. It is now used for both.
1422 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1423 patch removes a few documentation additions to RFC 3028, because the
1424 latest draft now contains them. It adds the new en;ascii-case comparator
1425 and a new error check for 8bit text in MIME parts. Comparator and
1426 require names are now matched exactly. I enabled the subaddress
1427 extension, but it is not well tested yet (read: it works for me)."
1429 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1430 rework some of the code of TK/09 above to avoid the hardwired use of
1431 "%lld" and "long long". Replaced the call to snprintf() with a call to
1434 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1435 records point to non-existent hosts", "retry timeout exceeded", and
1436 "retry time not reached for any host after a long failure period".
1438 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1439 experimental DomainKeys support:
1441 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1442 (2) On an error such as an illegally used "control", the wrong name for
1443 the control was given.
1445 These problems did NOT occur unless DomainKeys support was compiled.
1447 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1449 PH/24 Added ${if match_ip condition.
1451 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1452 databases so that it will be absolutely obvious if a crash occurs in the
1453 DB library. This is a regular occurrence (often caused by mis-matched
1456 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1457 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1458 picked up on a user's system that detects such things. There doesn't seem
1459 to be a gcc warning option for this - only an attribute that has to be
1460 put on the function's prototype. It seems that in Fedora Core 4 they have
1461 set this on a number of new functions. No doubt there will be more in due
1464 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1465 specify a transport. However, if an address that was verified by such a
1466 router was the subject of a callout, Exim crashed because it tried to
1467 read the rcpt_include_affixes from the non-existent transport. Now it
1468 just assumes that the setting of that option is false. This bug was
1469 introduced by 4.51/PH/31.
1471 PH/28 Changed -d+all to exclude +memory, because that information is very
1472 rarely of interest, but it makes the output a lot bigger. People tend to
1473 do -d+all out of habit.
1475 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1476 code in os-type was giving problems when libc.so lives in lib64, like on
1479 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1480 aren't the modern standard, and it seems that some systems' include files
1481 don't always have them. Exim was already checking for some of the newer
1482 ones like T_AAAA, and defining it itself. I've added checks for all the
1483 record types that Exim uses.
1485 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1486 not automatically generating a new one, as it is supposed to. This
1487 prevented TLS from working. If the file did exist, but contained invalid
1488 data, a new version was generated, as expected. It was only the case of a
1489 non-existent file that was broken.
1491 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1492 with a change in libdomainkeys > 0.64.
1494 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1495 from DNS. If the selector record carries the flag, it now has
1496 precedence over the domain-wide flag.
1498 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1500 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1501 the use of a port name that isn't defined in /etc/services) occurred, the
1502 message was deferred as in a normal delivery, and thus remained on the
1503 spool, instead of being failed because of the mua_wrapper setting. This
1504 is now fixed, and I tidied up some of the mua_wrapper messages at the
1507 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1508 the messages in a hash of arrays rather than using individual hashes.
1509 This is a bit cleaner and results in dramatic memory savings, albeit
1510 at a slight CPU cost.
1512 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1513 as requested by Marc Sherman.
1515 SC/10 Eximstats: added histograms for user specified patterns as requested
1518 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1520 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1521 fopen() in the content-scanning modules that did not already have it.
1527 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1528 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1530 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1532 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1534 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1535 to test Sieve filters that use "vacation".
1537 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1538 that changes the way the GnuTLS parameters are stored in the cache file.
1539 The new format can be generated externally. For backward compatibility,
1540 if the data in the cache doesn't make sense, Exim assumes it has read an
1541 old-format file, and it generates new data and writes a new file. This
1542 means that you can't go back to an older release without removing the
1545 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1546 work if there are any delivery delays because "one_time" forces the
1547 parent to be marked "delivered", so its unseen clone is never tried
1548 again. For this reason, Exim now forbids the simultaneous setting of
1551 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1552 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1553 entry if you want to know the details. The fix, however, neglected to
1554 consider the case when local delivery batching is involved. The test for
1555 "previously delivered" was not happening when checking to see if an
1556 address could be batched with a previous (undelivered) one; under
1557 certain circumstances this could lead to multiple deliveries to the same
1560 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1561 in its include files, and this causes problems building Exim.
1563 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1564 header_syntax) but Exim was just ignoring anything given after a slash.
1565 In particular, this caused confusion with an attempt to use "verify =
1566 reverse_host_lookup/defer_ok". An error is now given when options are
1567 supplied for verify items that do not have them. (Maybe reverse_host_
1568 lookup should have a defer_ok option, but that's a different point.)
1570 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1571 defined by RFC 821) to 2048, because there were problems with some AUTH
1572 commands, and RFC 1869 says the size should be increased for extended
1573 SMTP commands that take arguments.
1575 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1578 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1579 "unknown" error; now it says that the functionality isn't in the binary.
1581 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1582 an address' error message when a string expansion fails (syntax or
1583 whatever). Otherwise the password may appear in the log. Following change
1584 PH/42 below, there is no longer a chance of it appearing in a bounce
1587 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1589 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1590 of its addresses was ever tried. (Bugzilla bug #2.)
1592 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1593 the result incorrectly in the debug output. (It correctly added a newline
1594 to what was transported.)
1596 TF/01 Added $received_time.
1598 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1599 commented out examples of how to interface to a virus scanner and to
1600 SpamAssassin. Also added commented examples of av_scanner and
1601 spamd_address settings.
1603 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1604 and controls are allowed in which ACLs. There were a couple of minor
1605 errors. Some of the entries in the conditions table (which is a table of
1606 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1607 negation of where the condition IS allowed.
1609 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1611 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1612 header file does not have a version number, so I've had to invent a new
1613 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1614 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1615 radiusclient), but it was contributed by a Radius user.
1617 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1618 files or directories, for OpenSSL.
1620 PH/19 When an Exim process that is running as root has to create an Exim log
1621 file, it does so in a subprocess that runs as exim:exim so as to get the
1622 ownership right at creation (otherwise, other Exim processes might see
1623 the file with the wrong ownership). There was no test for failure of this
1624 fork() call, which would lead to the process getting stuck as it waited
1625 for a non-existent subprocess. Forks do occasionally fail when resources
1626 run out. I reviewed all the other calls to fork(); they all seem to check
1629 PH/20 When checking for unexpected SMTP input at connect time (before writing
1630 the banner), Exim was not dealing correctly with a non-positive return
1631 from the read() function. If the client had disconnected by this time,
1632 the result was a log entry for a synchronization error with an empty
1633 string after "input=" when read() returned zero. If read() returned -1
1634 (an event I could not check), uninitialized data bytes were printed.
1635 There were reports of junk text (parts of files, etc) appearing after
1638 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1640 PH/22 Added support for macro redefinition, and (re)definition in between
1641 driver and ACL definitions.
1643 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1644 forgetting to use the resulting value; it was using the unexpanded value.
1646 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1647 hadn't been configured. The fix is from Juergen Kreileder, who
1648 understands it better than I do:
1650 "Here's what I see happening with three configured cyrus_sasl
1651 authenticators configured (plain, login, cram-md5):
1653 On startup auth_cyrus_sasl_init() gets called for each of these.
1654 This means three calls to sasl_listmech() without a specified mech_list.
1655 => SASL tests which mechs of all available mechs actually work
1656 => three warnings about OTP not working
1657 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1659 With the patch, sasl_listmech() also gets called three times. But now
1660 SASL's mech_list option is set to the server_mech specified in the the
1661 authenticator. Or in other words, the answer from sasl_listmech()
1662 gets limited to just the mech you're testing for (which is different
1664 => the return list contains just 'plain' or 'login', 'cram-md5' or
1665 nothing depending on the value of ob->server_mech.
1667 I've just tested the patch: Authentication still works fine,
1668 unavailable mechs specified in the exim configuration are still
1669 caught, and the auth.log warnings about OTP are gone."
1671 PH/25 When debugging is enabled, the contents of the command line are added
1672 to the debugging output, even when log_selector=+arguments is not
1675 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1676 answer is "GNU", and only if the return is "GNU/something" is the answer
1679 PH/27 $acl_verify_message is now set immediately after the failure of a
1680 verification in an ACL, and so is available in subsequent modifiers. In
1681 particular, the message can be preserved by coding like this:
1683 warn !verify = sender
1684 set acl_m0 = $acl_verify_message
1686 Previously, $acl_verify_message was set only while expanding "message"
1687 and "log_message" when a very denied access.
1689 PH/28 Modified OS/os.c-Linux with
1691 -#ifndef OS_LOAD_AVERAGE
1692 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1694 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1695 nomenclature these days.)
1697 PH/29 Installed patch from the Sieve maintainer that adds the options
1698 sieve_useraddress and sieve_subaddress to the redirect router.
1700 PH/30 In these circumstances:
1701 . Two addresses routed to the same list of hosts;
1702 . First host does not offer TLS;
1703 . First host accepts first address;
1704 . First host gives temporary error to second address;
1705 . Second host offers TLS and a TLS session is established;
1706 . Second host accepts second address.
1707 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1708 and peerdn, if requested) that were in fact used only for the second
1711 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1712 attention to any local part prefix or suffix that was matched by the
1713 router that accepted the address. It now behaves in the same way as it
1714 does for delivery: the affixes are removed from the local part unless
1715 rcpt_include_affixes is set on the transport.
1717 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1718 timeout during the DATA phase of an incoming message.
1720 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1721 applied a patch sanctioned by the Sieve maintainer.
1723 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1724 the uid or gid is negative. A case of a negative gid caused this to be
1725 noticed. The fix allows for either to be negative.
1727 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1728 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1731 PH/36 The change PH/12 above was broken. Fixed it.
1733 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1734 the grounds that routing the same address twice would always produce the
1735 same answer. This might have been true once, but it is certainly no
1736 longer true now. Routing a child address may depend on the previous
1737 routing that produced that child. Some complicated redirection strategies
1738 went wrong when messages had multiple recipients, and made Exim's
1739 behaviour dependent on the order in which the addresses were given.
1741 I have moved the duplicate checking until after the routing is complete.
1742 Exim scans the addresses that are assigned to local and remote
1743 transports, and removes any duplicates. This means that more work will be
1744 done, as duplicates will always all be routed, but duplicates are
1745 presumably rare, so I don't expect this is of any significance.
1747 For deliveries to pipes, files, and autoreplies, the duplicate checking
1748 still happens during the routing process, since they are not going to be
1751 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1752 It corrects a timeout issue with spamd. This is Ian's comment: "The
1753 background is that sometimes spamd either never reads data from a
1754 connection it has accepted, or it never writes response data. The exiscan
1755 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1756 blindly assumes that writes won't block so it may never time out."
1758 PH/39 Allow G after quota size as well as K and M.
1760 PH/40 The value set for $authenticated_id in an authenticator may not contain
1761 binary zeroes or newlines because the value is written to log lines and
1762 to spool files. There was no check on this. Now the value is run through
1763 the string_printing() function so that such characters are converted to
1764 printable escape sequences.
1766 PH/41 $message_linecount is a new variable that contains the total number of
1767 lines in the message. Compare $body_linecount, which is the count for the
1770 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1771 bounce and delay warning messages, except in certain special cases, which
1774 (a) An SMTP error message from a remote host;
1775 (b) A message specified in a :fail: redirection;
1776 (c) A message specified in a "fail" command in a system filter;
1777 (d) A message specified in a FAIL return from the queryprogram router;
1778 (e) A message specified by the cannot_route_message router option.
1780 In these cases only, Exim does include the error details in bounce and
1781 warning messages. There are also a few cases where bland messages such
1782 as "unrouteable address" or "local delivery error" are given.
1784 PH/43 $value is now also set for the "else" part of a ${run expansion.
1786 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1787 being worked on, but at least Exim now implements the latest version to
1790 PH/45 In a pipe transport, although a timeout while waiting for the pipe
1791 process to complete was treated as a delivery failure, a timeout while
1792 writing the message to the pipe was logged, but erroneously treated as a
1793 successful delivery. Such timeouts include transport filter timeouts. For
1794 consistency with the overall process timeout, these timeouts are now
1795 treated as errors, giving rise to delivery failures by default. However,
1796 there is now a new Boolean option for the pipe transport called
1797 timeout_defer, which, if set TRUE, converts the failures into defers for
1798 both kinds of timeout. A transport filter timeout is now identified in
1801 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1802 systems where "make" and "gmake" are different, calling "gmake" at top
1803 level broke things. I've arranged for the value of $(MAKE) to be passed
1804 from the Makefile to this script so that it can call the same version of
1808 A note about Exim versions 4.44 and 4.50
1809 ----------------------------------------
1811 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1812 changes of various kinds. As a consequence, a big documentation update was
1813 needed. This delayed the release for rather longer than seemed good, especially
1814 in the light of a couple of (minor) security issues. Therefore, the changes
1815 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1816 release. So 4.44 and 4.50 are in effect two different branches that both start
1819 I have left the 4.50 change log unchanged; it contains all the changes since
1820 4.43. The change log for 4.44 is below; many of its items are identical to
1821 those for 4.50. This seems to be the most sensible way to preserve the
1822 historical information.
1828 1. Minor wording change to the doc/README.SIEVE file.
1830 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1831 computation of the current number of files was incorrect.
1833 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1834 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1837 4. Give more explanation in the error message when the command for a transport
1838 filter fails to execute.
1840 5. There are several places where Exim runs a non-Exim command in a
1841 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1842 was being done only for the command run by the queryprogram router. It is
1843 now done for all such subprocesses. The other cases are: ${run, transport
1844 filters, and the commands run by the lmtp and pipe transports.
1846 6. Added CONFIGURE_GROUP build-time option.
1848 7. Some older OS have a limit of 256 on the maximum number of file
1849 descriptors. Exim was using setrlimit() to set 1000 as a large value
1850 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1851 systems. I've change it so that if it can't get 1000, it tries for 256.
1853 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1854 was an oversight, and furthermore, ever since the addition of extra
1855 controls (e.g. 4.43/32), the checks on when to allow different forms of
1856 "control" were broken. There should now be diagnostics for all cases when a
1857 control that does not make sense is encountered.
1859 9. Added the /retain_sender option to "control=submission".
1861 10. $recipients is now available in the predata ACL (oversight).
1863 11. Tidy the search cache before the fork to do a delivery from a message
1864 received from the command line. Otherwise the child will trigger a lookup
1865 failure and thereby defer the delivery if it tries to use (for example) a
1866 cached ldap connection that the parent has called unbind on.
1868 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1869 of $address_data from the recipient verification was clobbered by the
1870 sender verification.
1872 13. The value of address_data from a sender verification is now available in
1873 $sender_address_data in subsequent conditions in the ACL statement.
1875 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1877 15. Added a new option "connect=<time>" to callout options, to set a different
1880 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1881 was its contents. (It was OK if the option was not defined at all.)
1883 17. A "Completed" log line is now written for messages that are removed from
1884 the spool by the -Mrm option.
1886 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1887 information about exactly what failed.
1889 19. Added -dd to debug only the daemon process.
1891 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1892 handles timeouts, both on the server side and network timeouts. Renamed the
1893 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1895 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1896 It is now set to "smtps".
1898 22. $host_address is now set to the target address during the checking of
1899 ignore_target_hosts.
1901 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1902 being passed; this would have caused $sender_host_name to have been used if
1903 matching the list had actually called for a host name (not very likely,
1904 since this list is usually IP addresses). A host name is now passed as
1907 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1908 code that specifies a non-restarting handler (typically sigaction() in
1909 modern systems) in an attempt to fix a rare and obscure crash bug.
1911 25. Narrowed the window for a race in the daemon that could cause it to ignore
1912 SIGCHLD signals. This is not a major problem, because they are used only to
1913 wake it up if nothing else does.
1915 26. A malformed maildirsize file could cause Exim to calculate negative values
1916 for the mailbox size or file count. Odd effects could occur as a result.
1917 The maildirsize information is now recalculated if the size or filecount
1920 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1921 support for a long time. Removed HAVE_SYS_VFS_H.
1923 28. Installed the latest version of exipick from John Jetmore.
1925 29. In an address list, if the pattern was not a regular expression, an empty
1926 subject address (from a bounce message) matched only if the pattern was an
1927 empty string. Non-empty patterns were not even tested. This was the wrong
1928 because it is perfectly reasonable to use an empty address as part of a
1929 database query. An empty address is now tested by patterns that are
1930 lookups. However, all the other forms of pattern expect the subject to
1931 contain a local part and a domain, and therefore, for them, an empty
1932 address still always fails if the pattern is not itself empty.
1934 30. Exim went into a mad DNS loop when attempting to do a callout where the
1935 host was specified on an smtp transport, and looking it up yielded more
1936 than one IP address.
1938 31. Re-factored the code for checking spool and log partition space into a
1939 function that finds that data and another that does the check. The former
1940 is then used to implement four new variables: $spool_space, $log_space,
1941 $spool_inodes, and $log_inodes.
1943 32. The RFC2047 encoding function was originally intended for short strings
1944 such as real names; it was not keeping to the 75-character limit for
1945 encoded words that the RFC imposes. It now respects the limit, and
1946 generates multiple encoded words if necessary. To be on the safe side, I
1947 have increased the buffer size for the ${rfc2047: expansion operator from
1950 33. It is now permitted to omit both strings after an "if" condition; if the
1951 condition is true, the result is "true". As before, when the second string
1952 is omitted, a false condition yields an empty string. This makes it less
1953 cumbersome to write custom ACL and router conditions.
1955 34. Failure to deliver a bounce message always caused it to be frozen, even if
1956 there was an errors_to setting on the router. The errors_to setting is now
1959 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1960 canonical form (fully expanded) before being placed in
1961 $sender_host_address.
1963 36. The table in the code that translates DNS record types into text (T_A to
1964 "A" for instance) was missing entries for NS and CNAME. It is just possible
1965 that this could have caused confusion if both these types were looked up
1966 for the same domain, because the text type is used as part of Exim's
1967 per-process caching. But the chance of anyone hitting this buglet seems
1970 37. The dnsdb lookup has been extended in a number of ways.
1972 (1) There is a new type, "zns", which walks up the domain tree until it
1973 finds some nameserver records. It should be used with care.
1975 (2) There is a new type, "mxh", which is like "mx" except that it returns
1976 just the host names, not the priorities.
1978 (3) It is now possible to give a list of domains (or IP addresses) to be
1979 looked up. The behaviour when one of the lookups defers can be
1980 controlled by a keyword.
1982 (4) It is now possible to specify the separator character for use when
1983 multiple records are returned.
1985 38. The dnslists ACL condition has been extended: it is now possible to supply
1986 a list of IP addresses and/or domains to be looked up in a particular DNS
1989 39. Added log_selector=+queue_time_overall.
1991 40. When running the queue in the test harness, wait just a tad after forking a
1992 delivery process, to get repeatability of debugging output.
1994 41. Include certificate and key file names in error message when GnuTLS fails
1995 to set them up, because the GnuTLS error message doesn't include the name
1996 of the failing file when there is a problem reading it.
1998 42. Allow both -bf and -bF in the same test run.
2000 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2002 44. The "Exiscan patch" is now merged into the mainline Exim source.
2004 45. Sometimes the final signoff response after QUIT could fail to get
2005 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2006 < 0 before doing a fflush(). This bug looks as though it goes back to the
2007 introduction of TLS in release 3.20, but "sometimes" must have been rare
2008 because the tests only now provoked it.
2010 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2011 (this can affect the format of dates).
2013 47. exim_tidydb, when checking for the continued existence of a message for
2014 which it has found a message-specific retry record, was not finding
2015 messages that were in split spool directories. Consequently, it was
2016 deleting retry records that should have stayed in existence.
2018 48. Steve fixed some bugs in eximstats.
2020 49. The SPA authentication driver was not abandoning authentication and moving
2021 on to the next authenticator when an expansion was forced to fail,
2022 contradicting the general specification for all authenticators. Instead it
2023 was generating a temporary error. It now behaves as specified.
2025 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2026 (the order specifies the preference for clients). The order is now AES256,
2027 AES128, 3DES, ARCFOUR128.
2029 51. Small patch to Sieve code - explicitly set From: when generating an
2032 52. Exim crashed if a remote delivery caused a very long error message to be
2033 recorded - for instance if somebody sent an entire SpamAssassin report back
2034 as a large number of 550 error lines. This bug was coincidentally fixed by
2035 increasing the size of one of Exim's internal buffers (big_buffer) that
2036 happened as part of the Exiscan merge. However, to be on the safe side, I
2037 have made the code more robust (and fixed the comments that describe what
2040 53. Now that there can be additional text after "Completed" in log lines (if
2041 the queue_time_overall log selector is set), a one-byte patch to exigrep
2042 was needed to allow it to recognize "Completed" as not the last thing in
2045 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2046 patch that reportedly fixes this has been added. I am not expert enough to
2047 create a test for it. This is what the patch creator wrote:
2049 "I found a little strange behaviour of ldap code when working with
2050 Windows 2003 AD Domain, where users was placed in more than one
2051 Organization Units. When I tried to give exim partial DN, the exit code
2052 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2053 But simultaneously result of request was absolutely normal ldap result,
2054 so I produce this patch..."
2056 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2057 REFERENCE, so I have modified the code to exclude the patch when that macro
2060 55. Some experimental protocols are using DNS PTR records for new purposes. The
2061 keys for these records are domain names, not reversed IP addresses. The
2062 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2063 leaves it alone. Component reversal etc. now happens only for IP addresses.
2066 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2068 57. Double the size of the debug message buffer (to 2048) so that more of very
2069 long debug lines gets shown.
2071 58. The exicyclog utility now does better if the number of log files to keep
2072 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2074 59. Two changes related to the smtp_active_hostname option:
2076 (1) $smtp_active_hostname is now available as a variable.
2077 (2) The default for smtp_banner uses $smtp_active_hostname instead
2078 of $primary_hostname.
2080 60. The host_aton() function is supposed to be passed a string that is known
2081 to be a valid IP address. However, in the case of IPv6 addresses, it was
2082 not checking this. This is a hostage to fortune. Exim now panics and dies
2083 if the condition is not met. A case was found where this could be provoked
2084 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2085 components; fortuitously, this particular loophole had already been fixed
2086 by change 4.50/55 above.
2088 If there are any other similar loopholes, the new check in host_aton()
2089 itself should stop them being exploited. The report I received stated that
2090 data on the command line could provoke the exploit when Exim was running as
2091 exim, but did not say which command line option was involved. All I could
2092 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2093 running as the user.
2096 61. There was a buffer overflow vulnerability in the SPA authentication code
2097 (which came originally from the Samba project). I have added a test to the
2098 spa_base64_to_bits() function which I hope fixes it.
2101 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2102 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2104 63. The daemon start-up calls getloadavg() while still root for those OS that
2105 need the first call to be done as root, but it missed one case: when
2106 deliver_queue_load_max is set with deliver_drop_privilege. This is
2107 necessary for the benefit of the queue runner, because there is no re-exec
2108 when deliver_drop_privilege is set.
2110 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2111 This has been fixed.
2113 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2114 was in use, was not putting the data itself into the right store pool;
2115 consequently, it could be overwritten for a subsequent message in the same
2116 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2119 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2121 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2122 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2123 the function were treating the return as a boolean value, which happened to
2124 work because 0=false and not-0=true, but is not correct code.
2126 68. The host_aton() function was not handling scoped IPv6 addresses (those
2127 with, for example, "%eth0" on the end) correctly.
2129 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2130 negated items (that is, ~something) in unsigned ints. Some compilers
2131 apparently mutter when there is no cast.
2133 70. If an address verification called from an ACL failed, and did not produce a
2134 user-specific message (i.e. there was only a "system" message), nothing was
2135 put in $acl_verify_message. In this situation, it now puts the system
2138 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2139 session; change 4.31/43 added the unwanted input to the log line - except
2140 that it did not do this in the start of session case. It now does.
2142 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2143 This is wrong and can cause the other end to generate a synchronization
2144 error if it is another Exim or anything else that does the synchronization
2145 check. A QUIT command is no longer sent after a timeout.
2147 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2148 during host lookups.
2150 74. The defer_ok option of callout verification was not working if it was used
2151 when verifying addresses in header lines, that is, for this case:
2153 verify = header_sender/callout=defer_ok
2155 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2156 those file descriptors could be used for SMTP connections. If anything
2157 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2158 could be sent to the SMTP client, causing chaos. The daemon now opens
2159 stdin, stdout, and stderr to /dev/null when it puts itself into the
2162 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2163 log by default. The user can override this with suitable Perl magic.
2165 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2166 the log message when discard triggers, was not working for the DATA ACL or
2167 for the non-SMTP ACL.
2169 78. Error message wording change in sieve.c.
2171 79. If smtp_accept_max_per_host was set, the number of connections could be
2172 restricted to fewer than expected, because the daemon was trying to set up
2173 a new connection before checking whether the processes handling previous
2174 connections had finished. The check for completed processes is now done
2175 earlier. On busy systems, this bug wouldn't be noticed because something
2176 else would have woken the daemon, and it would have reaped the completed
2179 80. If a message was submitted locally by a user whose login name contained one
2180 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2181 It caused a spool format error. I have fixed the spool reading code. A
2182 related problem was that the "from" clause in the Received: line became
2183 illegal because of the space(s). It is now covered by ${quote_local_part.
2185 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2188 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2190 83. Patch from Sieve maintainer to fix unterminated string problem in
2191 "vacation" handling.
2193 84. Some minor changes to the Linux configuration files to help with other
2194 OS variants using glibc.
2196 85. One more patch for Sieve to update vacation handling to latest spec.
2199 ----------------------------------------------------
2200 See the note above about the 4.44 and 4.50 releases.
2201 ----------------------------------------------------
2207 1. Change 4.43/35 introduced a bug that caused file counts to be
2208 incorrectly computed when quota_filecount was set in an appendfile
2211 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2212 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2215 3. Give more explanation in the error message when the command for a transport
2216 filter fails to execute.
2218 4. There are several places where Exim runs a non-Exim command in a
2219 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2220 was being done only for the command run by the queryprogram router. It is
2221 now done for all such subprocesses. The other cases are: ${run, transport
2222 filters, and the commands run by the lmtp and pipe transports.
2224 5. Some older OS have a limit of 256 on the maximum number of file
2225 descriptors. Exim was using setrlimit() to set 1000 as a large value
2226 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2227 systems. I've change it so that if it can't get 1000, it tries for 256.
2229 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2230 was an oversight, and furthermore, ever since the addition of extra
2231 controls (e.g. 4.43/32), the checks on when to allow different forms of
2232 "control" were broken. There should now be diagnostics for all cases when a
2233 control that does not make sense is encountered.
2235 7. $recipients is now available in the predata ACL (oversight).
2237 8. Tidy the search cache before the fork to do a delivery from a message
2238 received from the command line. Otherwise the child will trigger a lookup
2239 failure and thereby defer the delivery if it tries to use (for example) a
2240 cached ldap connection that the parent has called unbind on.
2242 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2243 of $address_data from the recipient verification was clobbered by the
2244 sender verification.
2246 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2247 was its contents. (It was OK if the option was not defined at all.)
2249 11. A "Completed" log line is now written for messages that are removed from
2250 the spool by the -Mrm option.
2252 12. $host_address is now set to the target address during the checking of
2253 ignore_target_hosts.
2255 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2256 being passed; this would have caused $sender_host_name to have been used if
2257 matching the list had actually called for a host name (not very likely,
2258 since this list is usually IP addresses). A host name is now passed as
2261 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2262 code that specifies a non-restarting handler (typically sigaction() in
2263 modern systems) in an attempt to fix a rare and obscure crash bug.
2265 15. Narrowed the window for a race in the daemon that could cause it to ignore
2266 SIGCHLD signals. This is not a major problem, because they are used only to
2267 wake it up if nothing else does.
2269 16. A malformed maildirsize file could cause Exim to calculate negative values
2270 for the mailbox size or file count. Odd effects could occur as a result.
2271 The maildirsize information is now recalculated if the size or filecount
2274 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2275 support for a long time. Removed HAVE_SYS_VFS_H.
2277 18. Updated exipick to current release from John Jetmore.
2279 19. Allow an empty sender to be matched against a lookup in an address list.
2280 Previously the only cases considered were a regular expression, or an
2283 20. Exim went into a mad DNS lookup loop when doing a callout where the
2284 host was specified on the transport, if the DNS lookup yielded more than
2287 21. The RFC2047 encoding function was originally intended for short strings
2288 such as real names; it was not keeping to the 75-character limit for
2289 encoded words that the RFC imposes. It now respects the limit, and
2290 generates multiple encoded words if necessary. To be on the safe side, I
2291 have increased the buffer size for the ${rfc2047: expansion operator from
2294 22. Failure to deliver a bounce message always caused it to be frozen, even if
2295 there was an errors_to setting on the router. The errors_to setting is now
2298 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2299 canonical form (fully expanded) before being placed in
2300 $sender_host_address.
2302 24. Updated eximstats to version 1.33
2304 25. Include certificate and key file names in error message when GnuTLS fails
2305 to set them up, because the GnuTLS error message doesn't include the name
2306 of the failing file when there is a problem reading it.
2308 26. Expand error message when OpenSSL has problems setting up cert/key files.
2311 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2312 (this can affect the format of dates).
2314 28. exim_tidydb, when checking for the continued existence of a message for
2315 which it has found a message-specific retry record, was not finding
2316 messages that were in split spool directories. Consequently, it was
2317 deleting retry records that should have stayed in existence.
2319 29. eximstats updated to version 1.35
2320 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2321 1.35 - bugfix such that pie charts by volume are generated correctly
2323 30. The SPA authentication driver was not abandoning authentication and moving
2324 on to the next authenticator when an expansion was forced to fail,
2325 contradicting the general specification for all authenticators. Instead it
2326 was generating a temporary error. It now behaves as specified.
2328 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2329 (the order specifies the preference for clients). The order is now AES256,
2330 AES128, 3DES, ARCFOUR128.
2332 31. Small patch to Sieve code - explicitly set From: when generating an
2335 32. Exim crashed if a remote delivery caused a very long error message to be
2336 recorded - for instance if somebody sent an entire SpamAssassin report back
2337 as a large number of 550 error lines. This bug was coincidentally fixed by
2338 increasing the size of one of Exim's internal buffers (big_buffer) that
2339 happened as part of the Exiscan merge. However, to be on the safe side, I
2340 have made the code more robust (and fixed the comments that describe what
2343 33. Some experimental protocols are using DNS PTR records for new purposes. The
2344 keys for these records are domain names, not reversed IP addresses. The
2345 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2346 leaves it alone. Component reversal etc. now happens only for IP addresses.
2349 34. The host_aton() function is supposed to be passed a string that is known
2350 to be a valid IP address. However, in the case of IPv6 addresses, it was
2351 not checking this. This is a hostage to fortune. Exim now panics and dies
2352 if the condition is not met. A case was found where this could be provoked
2353 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2354 components; fortuitously, this particular loophole had already been fixed
2355 by change 4.50/55 or 4.44/33 above.
2357 If there are any other similar loopholes, the new check in host_aton()
2358 itself should stop them being exploited. The report I received stated that
2359 data on the command line could provoke the exploit when Exim was running as
2360 exim, but did not say which command line option was involved. All I could
2361 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2362 running as the user.
2365 35. There was a buffer overflow vulnerability in the SPA authentication code
2366 (which came originally from the Samba project). I have added a test to the
2367 spa_base64_to_bits() function which I hope fixes it.
2370 36. The daemon start-up calls getloadavg() while still root for those OS that
2371 need the first call to be done as root, but it missed one case: when
2372 deliver_queue_load_max is set with deliver_drop_privilege. This is
2373 necessary for the benefit of the queue runner, because there is no re-exec
2374 when deliver_drop_privilege is set.
2376 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2377 was in use, was not putting the data itself into the right store pool;
2378 consequently, it could be overwritten for a subsequent message in the same
2379 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2382 38. Sometimes the final signoff response after QUIT could fail to get
2383 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2384 < 0 before doing a fflush(). This bug looks as though it goes back to the
2385 introduction of TLS in release 3.20, but "sometimes" must have been rare
2386 because the tests only now provoked it.
2392 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2393 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2394 Now it returns an integer. A number of calls were still expecting a T/F
2395 return. Fortuitously, in all cases, the tests worked in OK situations,
2396 which is the norm. However, things would have gone wrong on any write
2397 failures on the smtp file descriptor. This function is used when sending
2398 messages over SMTP and also when doing verify callouts.
2400 2. When Exim is called to do synchronous delivery of a locally submitted
2401 message (the -odf or -odi options), it no longer closes stderr before doing
2404 3. Implemented the mua_wrapper option.
2406 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2408 5. Implemented the functions header_remove(), header_testname(),
2409 header_add_at_position(), and receive_remove_recipient(), and exported them
2412 6. If an ACL "warn" statement specified the addition of headers, Exim already
2413 inserted X-ACL-Warn: at the start if there was no header name. However, it
2414 was not making this test for the second and subsequent header lines if
2415 there were newlines in the string. This meant that an invalid header could
2416 be inserted if Exim was badly configured.
2418 7. Allow an ACL "warn" statement to add header lines at the start or after all
2419 the Received: headers, as well as at the end.
2421 8. Added the rcpt_4xx retry error code.
2423 9. Added postmaster_mailfrom=xxx to callout verification option.
2425 10. Added mailfrom=xxxx to the callout verification option, for verify=
2428 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2429 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2431 12. Inserted some casts to stop certain compilers warning when using pointer
2432 differences as field lengths or precisions in printf-type calls (mostly
2433 affecting debugging statements).
2435 13. Added optional readline() support for -be (dynamically loaded).
2437 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2438 same clock tick as a message's arrival, so that its received time was the
2439 same as the "first fail" time on the retry record, and that message
2440 remained on the queue past the ultimate address timeout, every queue runner
2441 would try a delivery (because it was past the ultimate address timeout) but
2442 after another failure, the ultimate address timeout, which should have then
2443 bounced the address, did not kick in. This was a "< instead of <=" error;
2444 in most cases the first failure would have been in the next clock tick
2445 after the received time, and all would be well.
2447 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2448 being recognized when the domain list was tested by the match_domain
2449 condition in an expansion string.
2451 16. Added the ${str2b64: operator.
2453 17. Exim was always calling setrlimit() to set a large limit for the number of
2454 processes, without checking whether the existing limit was already
2455 adequate. (It did check for the limit on file descriptors.) Furthermore,
2456 errors from getrlimit() and setrlimit() were being ignored. Now they are
2457 logged to the main and panic logs, but Exim does carry on, to try to do its
2458 job under whatever limits there are.
2460 18. Imported PCRE 5.0.
2462 19. Trivial typo in log message " temporarily refused connection" (the leading
2465 20. If the log selector return_path_on_delivery was set and an address was
2466 redirected to /dev/null, the delivery process crashed because it assumed
2467 that a return path would always be set for a "successful" delivery. In this
2468 case, the whole delivery is bypassed as an optimization, and therefore no
2471 21. Internal re-arrangement: the function for sending a challenge and reading
2472 a response while authentication was assuming a zero-terminated challenge
2473 string. It's now changed to take a pointer and a length, to allow for
2474 binary data in such strings.
2476 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2478 23. Exim was not respecting finduser_retries when seeking the login of the
2479 uid under which it was called; it was always trying 10 times. (The default
2480 setting of finduser_retries is zero.) Also, it was sleeping after the final
2481 failure, which is pointless.
2483 24. Implemented tls_on_connect_ports.
2485 25. Implemented acl_smtp_predata.
2487 26. If the domain in control=submission is set empty, Exim assumes that the
2488 authenticated id is a complete email address when it generates From: or
2489 Sender: header lines.
2491 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2492 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2493 chown and chgrp in /bin and hostname in /usr/bin.
2495 28. Exim was keeping the "process log" file open after each use, just as it
2496 does for the main log. This opens the possibility of it remaining open for
2497 long periods when the USR1 signal hits a daemon. Occasional processlog
2498 errors were reported, that could have been caused by this. Anyway, it seems
2499 much more sensible not to leave this file open at all, so that is what now
2502 29. The long-running daemon process does not normally write to the log once it
2503 has entered its main loop, and it closes the log before doing so. This is
2504 so that log files can straightforwardly be renamed and moved. However,
2505 there are a couple of unusual error situations where the daemon does write
2506 log entries, and I had neglected to close the log afterwards.
2508 30. The text of an SMTP error response that was received during a remote
2509 delivery was being truncated at 512 bytes. This is too short for some of
2510 the long messages that one sometimes sees. I've increased the limit to
2513 31. It is now possible to make retry rules that apply only when a message has a
2514 specific sender, in particular, an empty sender.
2516 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2517 it possible to be selective about when SMTP synchronization is enforced.
2519 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2521 32. Implemented hosts_connection_nolog.
2523 33. Added an ACL for QUIT.
2525 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2528 35. Added mailbox_size and mailbox_filecount to appendfile.
2530 36. Added control = no_multiline_responses to ACLs.
2532 37. There was a bug in the logic of the code that waits for the clock to tick
2533 in the case where the clock went backwards by a substantial amount such
2534 that the microsecond fraction of "now" was more than the microsecond
2535 fraction of "then" (but the whole seconds number was less).
2537 38. Added support for the libradius Radius client library this is found on
2538 FreeBSD (previously only the radiusclient library was supported).
2544 1. When certain lookups returned multiple values in the form name=value, the
2545 quoting of the values was not always being done properly. Specifically:
2546 (a) If the value started with a double quote, but contained no whitespace,
2548 (b) If the value contained whitespace other than a space character (i.e.
2549 tabs or newlines or carriage returns) it was not quoted.
2550 This fix has been applied to the mysql and pgsql lookups by writing a
2551 separate quoting function and calling it from the lookup code. The fix
2552 should probably also be applied to nisplus, ibase and oracle lookups, but
2553 since I cannot test any of those, I have not disturbed their existing code.
2555 2. A hit in the callout cache for a specific address caused a log line with no
2556 reason for rejecting RCPT. Now it says "Previous (cached) callout
2557 verification failure".
2559 3. There was an off-by-one bug in the queryprogram router. An over-long
2560 return line was truncated at 256 instead of 255 characters, thereby
2561 overflowing its buffer with the terminating zero. As well as fixing this, I
2562 have increased the buffer size to 1024 (and made a note to document this).
2564 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2565 when Exim is waiting for an SMTP response from a remote server, Exim
2566 restarts its select() call on the socket, thereby resetting its timeout.
2567 This is not a problem when such interrupts are rare. Somebody set up a cron
2568 job to run exiwhat every 2 minutes, which is less than the normal select()
2569 timeout (5 or 10 minutes). This meant that the select() timeout never
2570 kicked in because it was always reset. I have fixed this by comparing the
2571 time when an interrupt arrives with the time at the start of the first call
2572 to select(). If more time than the timeout has elapsed, the interrupt is
2573 treated as a timeout.
2575 5. Some internal re-factoring in preparation for the addition of Sieve
2576 extensions (by MH). In particular, the "personal" test is moved to a
2577 separate function, and given an option for scanning Cc: and Bcc: (which is
2578 not set for Exim filters).
2580 6. When Exim created an email address using the login of the caller as the
2581 local part (e.g. when creating a From: or Sender: header line), it was not
2582 quoting the local part when it contained special characters such as @.
2584 7. Installed new OpenBSD configuration files.
2586 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2587 try to make them clearer.
2589 9. Callout options, other than the timeout value, were being ignored when
2590 verifying sender addresses in header lines. For example, when using
2592 verify = header_sender/callout=no_cache
2594 the cache was (incorrectly) being used.
2596 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2597 only the Cygwin environment.
2599 11. When return_path_on_delivery was set as a log selector, if different remote
2600 addresses in the same message used different return paths and parallel
2601 remote delivery occurred, the wrong values would sometimes be logged.
2602 (Whenever a remote delivery process finished, the return path value from
2603 the most recently started remote delivery process was logged.)
2605 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2606 lines when AUTH and/or TLS are in use. This is the "received protocol"
2607 field. Exim used to use "asmtp" for authenticated SMTP, without any
2608 indication (in the protocol name) for TLS use. Now it follows the RFC and
2609 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2610 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2611 names appear in log lines as well as in Received: header lines.
2613 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2614 extensions, and comparison tests, and to fix some bugs.
2616 14. Changes to the "personal" filter test:
2618 (1) The test was buggy in that it was just doing the equivalent of
2619 "contains" tests on header lines. For example, if a user's address was
2620 anne@some.where, the "personal" test would incorrectly be true for
2622 To: susanne@some.where
2624 This test is now done by extracting each address from the header in turn,
2625 and checking the entire address. Other tests that are part of "personal"
2626 are now done using regular expressions (for example, to check local parts
2627 of addresses in From: header lines).
2629 (2) The list of non-personal local parts in From: addresses has been
2630 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2631 taken from the Sieve specification recommendations.
2633 (3) If the message contains any header line starting with "List-" it is
2634 treated as non-personal.
2636 (4) The test for "circular" in the Subject: header line has been removed
2637 because it now seems ill-conceived.
2639 15. Minor typos in src/EDITME comments corrected.
2641 16. Installed latest exipick from John Jetmore.
2643 17. If headers_add on a router specified a text string that was too long for
2644 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2645 of string_sprintf() is now avoided.
2647 18. $message_body_size was not set (it was always zero) when running the DATA
2648 ACL and the local_scan() function.
2650 19. For the "mail" command in an Exim filter, no default was being set for
2651 the once_repeat time, causing a random time value to be used if "once" was
2652 specified. (If the value happened to be <= 0, no repeat happened.) The
2653 default is now 0s, meaning "never repeat". The "vacation" command was OK
2654 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2655 (I found it when inspecting the code).
2657 20. There is now an overall timeout for performing a callout verification. It
2658 defaults to 4 times the callout timeout, which applies to individual SMTP
2659 commands during the callout. The overall timeout applies when there is more
2660 than one host that can be tried. The timeout is checked before trying the
2661 next host. This prevents very long delays if there are a large number of
2662 hosts and all are timing out (e.g. when the network connections are timing
2663 out). The value of the overall timeout can be changed by specifying an
2664 additional sub-option for "callout", called "maxwait". For example:
2666 verify = sender/callout=5s,maxwait=20s
2668 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2669 to the end before writing, but this should make it even safer).
2671 22. Exim was forgetting that it had advertised PIPELINING for the second and
2672 subsequent messages on an SMTP connection. It was also not resetting its
2673 memory on STARTTLS and an internal HELO.
2675 23. When Exim logs an SMTP synchronization error within a session, it now
2676 records whether PIPELINING has been advertised or not.
2678 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2679 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2680 rather than long int.
2682 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2684 26. Added the never_mail option to autoreply.
2690 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2691 crash if the getsockname() call failed; this can happen if a connection is
2692 closed very soon after it is established. The problem was simply in the
2693 order in which certain operations were done, causing Exim to try to write
2694 to the SMTP stream before it had set up the file descriptor. The bug has
2695 been fixed by making things happen in the correct order.
2701 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2702 before the connection was closed, thus losing the rejection response.
2704 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2705 some early Solaris releases, but causes trouble in current releases where
2706 socklen_t is defined.
2708 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2711 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2714 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2715 that releasing the top of it at the end releases what was used for sub-
2716 expansions (unless the block got too big). However, discard this block if
2717 the first thing is a variable or header, so that we can use its block when
2718 it is dynamic (useful for very large $message_headers, for example).
2720 6. Lookups now cache *every* query, not just the most recent. A new, separate
2721 store pool is used for this. It can be recovered when all lookup caches are
2722 flushed. Lookups now release memory at the end of their result strings.
2723 This has involved some general refactoring of the lookup sources.
2725 7. Some code has been added to the store_xxx() functions to reduce the amount
2726 of flapping under certain conditions.
2728 8. log_incoming_interface used to affect only the <= reception log lines. Now
2729 it causes the local interface and port to be added to several more SMTP log
2730 lines, for example "SMTP connection from", and rejection lines.
2732 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2734 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2736 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2737 could be overwritten at the end of the current message (or the start of a
2738 new message if it was set in a HELO ACL). The value is now preserved for
2739 the duration of the SMTP connection.
2741 12. If a transport had a headers_rewrite setting, and a matching header line
2742 contained an unqualified address, that address was qualified, even if it
2743 did not match any rewriting rules. The underlying bug was that the values
2744 of the flags that permit the existence of unqualified sender and recipient
2745 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2746 non-local messages, and by -bnq for local messages) were not being
2747 preserved with the message after it was received.
2749 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2750 "next input=" as part of the text comprising the host identity instead of
2751 the correct text. The code was using the same buffer for two different
2752 strings. However, depending on which order the printing function evaluated
2753 its arguments, the bug did not always show up. Under Linux, for example, my
2754 test suite worked just fine.
2756 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2757 doesn't work with some older versions of Perl. It has been changed to "my",
2758 which in any case is probably the better facility to use.
2760 15. A really picky compiler found some instances of statements for creating
2761 error messages that either had too many or two few arguments for the format
2764 16. The size of the buffer for calls to the DNS resolver has been increased
2765 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2766 for addresses that have a lot of PTR records. This alleviates a problem; it
2767 does not fully solve it.
2769 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2770 buffer now truncates the list and logs the incident, which is the same
2771 action as happens when Exim is looking up a host name and its aliases.
2772 Previously in this situation something unpredictable would happen;
2773 sometimes it was "internal error: store_reset failed".
2775 18. If a server dropped the connection unexpectedly when an Exim client was
2776 using GnuTLS and trying to read a response, the client delivery process
2777 crashed while trying to generate an error log message.
2779 19. If a "warn" verb in an ACL added multiple headers to a message in a single
2780 string, for example:
2782 warn message = H1: something\nH2: something
2784 the text was added as a single header line from Exim's point of view
2785 though it ended up OK in the delivered message. However, searching for the
2786 second and subsequent header lines using $h_h2: did not work. This has been
2787 fixed. Similarly, if a system filter added multiple headers in this way,
2788 the routers could not see them.
2790 20. Expanded the error message when iplsearch is called with an invalid key to
2791 suggest using net-iplsearch in a host list.
2793 21. When running tests using -bh, any delays imposed by "delay" modifiers in
2794 ACLs are no longer actually imposed (and a message to that effect is
2797 22. If a "gecos" field in a passwd entry contained escaped characters, in
2798 particular, if it contained a \" sequence, Exim got it wrong when building
2799 a From: or a Sender: header from that name. A second bug also caused
2800 incorrect handling when an unquoted " was present following a character
2801 that needed quoting.
2803 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2804 was not being matched caselessly.
2806 24. Arranged for all hyphens in the exim.8 source to be escaped with
2809 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2810 properly. Recipient callout cache records were still being keyed to include
2811 the sender, even when use_sender was set false. This led to far more
2812 callouts that were necessary. The sender is no longer included in the key
2813 when use_sender is false.
2815 26. Added "control = submission" modifier to ACLs.
2817 27. Added the ${base62d: operator to decode base 62 numbers.
2819 28. dnsdb lookups can now access SRV records.
2821 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2822 the configuration file.
2824 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2825 (-v) mode. This makes the output for a verbose queue run more intelligible.
2827 31. Added a use_postmaster feature to recipient callouts.
2829 32. Added the $body_zerocount variable, containing the number of binary zero
2830 bytes in the message body.
2832 33. The time of last modification of the "new" subdirectory is now used as the
2833 "mailbox time last read" when there is a quota error for a maildir
2836 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2838 35. Added +ignore_unknown as a special item in host lists.
2840 36. Code for decoding IPv6 addresses in host lists is now included, even if
2841 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2842 address was recognized as an IP address, but was then not correctly decoded
2843 into binary, causing unexpected and incorrect effects when compared with
2850 1. Very minor rewording of debugging text in manualroute to say "list of
2851 hosts" instead of "hostlist".
2853 2. If verify=header_syntax was set, and a header line with an unqualified
2854 address (no domain) and a large number of spaces between the end of the
2855 name and the colon was received, the reception process suffered a buffer
2856 overflow, and (when I tested it) crashed. This was caused by some obsolete
2857 code that should have been removed. The fix is to remove it!
2859 3. When running in the test harness, delay a bit after writing a bounce
2860 message to get a bit more predictability in the log output.
2862 4. Added a call to search_tidyup() just before forking a reception process. In
2863 theory, someone could use a lookup in the expansion of smtp_accept_max_
2864 per_host which, without the tidyup, could leave open a database connection.
2866 5. Added the variables $recipient_data and $sender_data which get set from a
2867 lookup success in an ACL "recipients" or "senders" condition, or a router
2868 "senders" option, similar to $domain_data and $local_part_data.
2870 6. Moved the writing of debug_print from before to after the "senders" test
2873 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2874 problems for message scanning, either using a data ACL, or using
2875 local_scan() because the Received: header was not generated till after they
2876 were called (in order to set the time as the time of reception completion).
2877 I have revised the way this works. The header is now generated after the
2878 body is received, but before the ACL or local_scan() are called. After they
2879 are run, the timestamp in the header is updated.
2885 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2886 before starting a queue runner without re-exec. This happened only when
2887 deliver_drop_privilege was set or when the Exim user was set to root. The
2888 effect of the bug was that timeouts during subsequent deliveries caused
2889 crashes instead of being properly handled. The handler is now left at its
2890 default (and expected) setting.
2892 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2893 message, again when deliver_drop_privilege is set or Exim is run as root.
2894 The bug described in (1) was not present in this case, but the tidying up
2895 of the other signals was missing. I have made the two cases consistent.
2897 3. The ignore_target_hosts setting on a manualroute router was being ignored
2898 for hosts that were looked up using the /MX notation.
2900 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2903 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2904 operated on the sender address. After changing the $sender_address to <>
2905 for the sender address verify, Exim was re-instated it as the original
2906 (before rewriting) address, but remembering that it had rewritten it, so it
2907 wasn't rewriting it again. This bug also had the effect of breaking the
2908 sender address verification caching when the sender address was rewritten.
2910 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2911 This has been changed so that if the ip literal address matches
2912 ignore_target_hosts, the router declines.
2914 7. Added expansion conditions match_domain, match_address, and match_local_
2915 part (NOT match_host).
2917 8. The placeholder for the Received: header didn't have a length field set.
2919 9. Added code to Exim itself and to exim_lock to test for a specific race
2920 condition that could lead to file corruption when using MBX delivery. The
2921 issue is with the lockfile that is created in /tmp. If this file is removed
2922 after a process has opened it but before that process has acquired a lock,
2923 there is the potential for a second process to recreate the file and also
2924 acquire a lock. This could lead to two Exim processes writing to the file
2925 at the same time. The added code performs the same test as UW imapd; it
2926 checks after acquiring the lock that its file descriptor still refers to
2927 the same named file.
2929 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2930 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2933 11. Added the smtp_active_hostname option. If used, this will typically be made
2934 to depend on the incoming interface address. Because $interface_address is
2935 not set up until the daemon has forked a reception process, error responses
2936 that can happen earlier (such as "too many connections") no longer contain
2939 12. If an expansion in a condition on a "warn" statement fails because a lookup
2940 defers, the "warn" statement is abandoned, and the next ACL statement is
2941 processed. Previously this caused the whole ACL to be aborted.
2943 13. Added the iplsearch lookup type.
2945 14. Added ident_timeout as a log selector.
2947 15. Added tls_certificate_verified as a log selector.
2949 16. Added a global option tls_require_ciphers (compare the smtp transport
2950 option of the same name). This controls incoming TLS connections.
2952 17. I finally figured out how to make tls_require_ciphers do a similar thing
2953 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2954 before starting the TLS session.
2956 18. Tabs are now shown as \t in -bP output.
2958 19. If the log selector return_path_on_delivery was set, Exim crashed when
2959 bouncing a message because it had too many Received: header lines.
2961 20. If two routers both had headers_remove settings, and the first one included
2962 a superfluous trailing colon, the final name in the first list and the
2963 first name in the second list were incorrectly joined into one item (with a
2964 colon in the middle).
2970 1. Added -C and -D options to the exinext utility, mainly to make it easier
2971 to include in the automated testing, but these could be helpful when
2972 multiple configurations are in use.
2974 2. The exinext utility was not formatting the output nicely when there was
2975 an alternate port involved in the retry record key, nor when there was a
2976 message id as well (for retries that were specific to a specific message
2977 and a specific host). It was also confused by IPv6 addresses, because of
2978 the additional colons they contain. I have fixed the IPv4 problem, and
2979 patched it up to do a reasonable job for IPv6.
2981 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2982 delivery, the log line now contains "pipelined" if PIPELINING was used.
2984 4. An SMTP transport process used to panic and die if the bind() call to set
2985 an explicit outgoing interface failed. This has been changed; it is now
2986 treated in the same way as a connect() failure.
2988 5. A reference to $sender_host_name in the part of a conditional expansion
2989 that was being skipped was still causing a DNS lookup. This no longer
2992 6. The def: expansion condition was not recognizing references to header lines
2993 that used bh_ and bheader_.
2995 7. Added the _cache feature to named lists.
2997 8. The code for checking quota_filecount in the appendfile transport was
2998 allowing one more file than it should have been.
3000 9. For compatibility with Sendmail, the command line option
3008 and sets the incoming protocol and host name (for trusted callers). The
3009 host name and its colon can be omitted when only the protocol is to be set.
3010 Note the Exim already has two private options, -pd and -ps, that refer to
3011 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3012 "s", but I don't think that's a major issue.
3014 10. A number of refactoring changes to the code, none of which should affect
3017 (a) The number of logging options was getting close to filling up the
3018 32-bit word that was used as a bit map. I have split them into two classes:
3019 those that are passed in the argument to log_write(), and those that are
3020 only ever tested independently outside of that function. These are now in
3021 separate 32-bit words, so there is plenty of room for expansion again.
3022 There is no change in the user interface or the logging behaviour.
3024 (b) When building, for example, log lines, the code previously used a
3025 macro that called string_cat() twice, in order to add two strings. This is
3026 not really sufficiently general. Furthermore, there was one instance where
3027 it was actually wrong because one of the argument was used twice, and in
3028 one call a function was used. (As it happened, calling the function twice
3029 did not affect the overall behaviour.) The macro has been replaced by a
3030 function that can join an arbitrary number of extra strings onto a growing
3033 (c) The code for expansion conditions now uses a table and a binary chop
3034 instead of a serial search (which was left over from when there were very
3035 few conditions). Also, it now recognizes conditions like "pam" even when
3036 the relevant support is not compiled in: a suitably worded error message is
3037 given if an attempt is made to use such a condition.
3039 11. Added ${time_interval:xxxxx}.
3041 12. A bug was causing one of the ddress fields not to be passed back correctly
3042 from remote delivery subprocesses. The field in question was not being
3043 subsequently used, so this caused to problems in practice.
3045 13. Added new log selectors queue_time and deliver_time.
3047 14. Might have fixed a bug in maildirsizefile handling that threw up
3048 "unexpected character" debug warnings, and recalculated the data
3049 unnecessarily. In any case, I expanded the warning message to give more
3052 15. Added the message "Restricted characters in address" to the statements in
3053 the default ACL that block characters like @ and % in local parts.
3055 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3056 Three changes have been made:
3058 (a) There was a serious bug; a negative response to MAIL caused the whole
3059 recipient domain to be cached as invalid, thereby blocking all messages
3060 to all local parts at the same domain, from all senders. This bug has
3061 been fixed. The domain is no longer cached after a negative response to
3062 MAIL if the sender used is not empty.
3064 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3067 (c) A new callout option, "use_sender" has been added for people who want
3068 the modified behaviour.
3074 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3077 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3078 indeed breaks things for older releases.
3080 3. Added additional logging to the case where there is a problem reading data
3081 from a filter that is running in a subprocess using a pipe, in order to
3082 try to track down a specific problem.
3084 4. Testing facility fudge: when running in the test harness and attempting
3085 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3086 getting "No route to host". Convert this to a timeout.
3088 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3091 6. Some OS don't have socklen_t but use size_t instead. This affects the
3092 fifth argument of getsockopt() amongst other things. This is now
3093 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3094 can be set for individual OS. I have set it for SunOS5, OSF1, and
3095 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3096 some earlier ones do not.
3098 7. Change 4.30/15 was not doing the test caselessly.
3100 8. The standard form for an IPv6 address literal was being rejected by address
3101 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3102 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3103 this, as well as the form without the "IPv6" on the front (but only when
3104 address literals are enabled, of course).
3106 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3108 10. Exim crashed if a message with an empty sender address specified by -f
3109 encountered a router with an errors_to setting. This could be provoked only
3110 by a command such as
3114 where an empty string was supplied; "<>" did not hit this bug.
3116 11. Installed PCRE release 4.5.
3118 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3119 remained set. It is now erased.
3121 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3122 times from message ids (which are base 36 rather than the normal 62).
3124 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3125 were being counted as actual protocol errors, and logged if the log
3126 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3127 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3128 and DATA following a set of rejected RCPTs do not count as protocol errors.
3129 In other words, Exim assumes they were pipelined, though this may not
3130 actually be the case. Of course, in all cases the client gets an
3131 appropriate error code.
3133 15. If a lookup fails in an ACL condition, a message about the failure may
3134 be available; it is used if testing the ACL cannot continue, because most
3135 such messages specify what the cause of the deferral is. However, some
3136 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3137 that caused an old message to be retained and used if a later statement
3138 caused a defer, replacing the real cause of the deferral.
3140 16. If an IP address had so many PTR records that the DNS lookup buffer
3141 was not large enough to hold them, Exim could crash while trying to process
3142 the truncated data. It now detects and logs this case.
3144 17. Further to 4.21/58, another change has been made: if (and only if) the
3145 first line of a message (the first header line) ends with CRLF, a bare LF
3146 in a subsequent header line has a space inserted after it, so as not to
3147 terminate the header.
3149 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3150 unnecessarily, used atoi() instead of strtol(), and didn't check the
3151 termination when getting file sizes from file names by regex.
3153 19. Completely re-implemented the support for maildirsize files, in the light
3154 of a number of problems with the previous contributed implementation
3155 (4.30/29). In particular:
3157 . If the quota is zero, the maildirsize file is maintained, but no quota is
3160 . If the maildir directory does not exist, it is created before any attempt
3161 to write a maildirsize file.
3163 . The quota value in the file is just a cache; if the quota is changed in
3164 the transport, the new value overrides.
3166 . A regular expression is available for excluding directories from the
3169 20. The autoreply transport checks the characters in options that define the
3170 message's headers; it allows continued headers, but it was checking with
3171 isspace() after an embedded newline instead of explicitly looking for a
3174 21. If all the "regular" hosts to which an address was routed had passed their
3175 expiry times, and had not reached their retry times, the address was
3176 bounced, even if fallback hosts were defined. Now Exim should go on to try
3179 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3180 equivalent code in the SMTP transport. Some hosts send humungous responses
3181 to HELO/EHLO, more than 1024 it seems.
3183 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3184 gives compiler warnings in some environments. I've now done it "properly",
3187 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3188 (because of problems with the built-in one) was declared to return uschar *
3189 instead of char *, causing compiler failure.
3191 25. Fixed a file descriptor leak when processing alias/forward files.
3193 26. Fixed a minor format string issue in dbfn.c.
3195 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3197 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3198 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3201 29. When a sender address is verified, it is cached, to save repeating the test
3202 when there is more than one recipient in a message. However, when the
3203 verification involves a callout, it is possible for different callout
3204 options to be set for different recipients. It is too complicated to keep
3205 track of this in the cache, so now Exim always runs a verification when a
3206 callout is required, relying on the callout cache for the optimization.
3207 The overhead is duplication of the address routing, but this should not be
3210 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3211 to be verified with callout=postmaster, and the main callout worked but the
3212 postmaster check failed, the verification correctly failed. However, if a
3213 subsequent RCPT command asked for sender verification *without* the
3214 postmaster check, incorrect caching caused this verification also to fail,
3217 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3218 it was not caching the DNS options (qualify_single, search_parents) that
3219 were used when the lookup failed. A subsequent lookup with different
3220 options therefore always gave the same answer, though there were cases
3221 where it should not have. (Example: a "domains = !$mx_any" option on a
3222 dnslookup router: the "domains" option is always processed without any
3223 widening, but the router might have qualify_single set.) Now Exim uses the
3224 cached value only when the same options are set.
3226 32. Added John Jetmore's "exipick" utility to the distribution.
3228 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3229 than a timeout (e.g. a certificate is required, and is not provided), an
3230 Exim server now closes the connection immediately. Previously it waited for
3231 the client to close - but if the client is SSL, it seems that they each
3232 wait for each other, leading to a delay before one of them times out.
3234 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3235 maintained 0.8.x compatibility because I don't think many are using it, and
3236 it is clearly obsolete.
3238 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3241 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3242 client certificate was expired. A simple patch fixes this, though I don't
3243 understand the full logic of why the verify callback is called multiple
3246 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3247 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3248 which causes problems with some clients (such as the Certicom SSL Plus
3249 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3250 disables the coutermeasure allowing Eudora to connect."
3252 38. Exim was not checking that a write() to a log file succeeded. This could
3253 lead to Bad Things if a log got too big, in particular if it hit a file
3254 size limit. Exim now panics and dies if it cannot write to a log file, just
3255 as it does if it cannot open a log file.
3257 39. Modified OS/Makefile-Linux so that it now contains
3259 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3261 The two -D definitions ensure that Exim is compiled with large file
3262 support, which makes it possible to handle log files that are bigger than
3265 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3266 instance) a domain was checked against a named list that involved a lookup,
3267 causing $domain_data to be set, then another domain was checked against the
3268 same list, then the first domain was re-checked, the value of $domain_data
3269 after the final check could be wrong. In particular, if the second check
3270 failed, it could be set empty. This bug probably also applied to
3273 41. The strip_trailing_dot option was not being applied to the address given
3274 with the -f command-line option.
3276 42. The code for reading a message's header from the spool was incrementing
3277 $received_count, but never initializing it. This meant that the value was
3278 incorrect (doubled) while delivering a message in the same process in which
3279 it was received. In the most common configuration of Exim, this never
3280 happens - a fresh exec is done - but it can happen when
3281 deliver_drop_privilege is set.
3283 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3284 it now includes up to 150 characters of the unexpected data in the log
3287 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3288 and building data strings. The size of both of these buffers was 10 000
3289 bytes - far larger than anybody would *ever* want, thought I. Needless to
3290 say, somebody hit the limit. I have increased the maximum line length to
3291 20 000 and the maximum data length of concatenated lines to 100 000. I have
3292 also fixed two bugs, because there was no checking on these buffers. Tsk,
3293 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3294 buffer is too small.
3296 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3297 lsearch lookups. Now it does.
3299 46. When parsing a route_list item in a manualroute router, a fixed-length
3300 buffer was used for the list of hosts. I made this 1024 bytes long,
3301 thinking that nobody would ever have a list of hosts that long. Wrong.
3302 Somebody had a whole pile of complicated expansion conditions, and the
3303 string was silently truncated, leading to an expansion error. It turns out
3304 that it is easier to change to an unlimited length (owing to other changes
3305 that have happened since this code was originally written) than to build
3306 structure for giving a limitation error. The length of the item that
3307 expands into the list of hosts is now unlimited.
3309 47. The lsearch lookup could not handle data where the length of text line was
3310 more than 4095 characters. Such lines were truncated, leading to shortened
3311 data being returned. It should now handle lines of any length.
3313 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3314 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3317 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3318 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3319 no longer happens in comment lines. A list of replacements is now placed
3320 at the head of all of the source files, except those whose only change is
3321 to replace PERL_COMMAND in the very first #! line.
3323 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3324 messages on the queue, with a bottom-up merge sort, using code contributed
3325 by Michael Haardt. This should make operations like -bp somewhat faster on
3326 large queues. It won't affect queue runners, except when queue_run_in_order
3329 51. Installed eximstats 1.31 in the distribution.
3331 52. Added support for SRV lookups to the dnslookup router.
3333 53. If an ACL referred to $message_body or $message_body_end, the value was not
3334 reset for any messages that followed in the same SMTP session.
3336 54. The store-handling optimization for building very long strings was not
3337 differentiating between the different store pools. I don't think this
3338 actually made any difference in practice, but I've tidied it.
3340 55. While running the routers to verify a sender address, $sender_address
3341 was still set to the sender address. This is wrong, because when routing to
3342 send a bounce to the sender, it would be empty. Therefore, I have changed
3343 it so that, while verifying a sender address, $sender_address is set to <>.
3344 (There is no change to what happens when verifying a recipient address.)
3346 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3347 target A or AAAA records (if not already returned) without resetting the
3348 qualify_single or search_parents options of the DNS resolver. These are
3349 inappropriate in this case because the targets of MX and SRV records must
3350 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3351 target that, when qualified, matched something in the local domain. These
3352 two options are now turned off when doing these lookups.
3354 57. It seems that at least some releases of Reiserfs (which does not have the
3355 concept of a fixed number of inodes) returns zero and not -1 for the
3356 number of available inodes. This interacted badly with check_spool_inodes,
3357 which assumed that -1 was the "no such thing" setting. What I have done is
3358 to check that the total number of inodes is greater than zero before doing
3359 the test of how many are available.
3361 58. When a "warn" ACL statement has a log_message modifier, the message is
3362 remembered, and not repeated. This is to avoid a lot of repetition when a
3363 message has many recipients that cause the same warning to be written.
3364 Howewer, Exim was preserving the list of already written lines for an
3365 entire SMTP session, which doesn't seem right. The memory is now reset if a
3366 new message is started.
3368 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3369 debugging output unless log_rewrite was also set.
3371 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3372 of (int)(handle) when we know that handle contains (void *)(-1).
3374 61. The Exim daemon panic-logs an error return when it closes the incoming
3375 connection. However "connection reset by peer" seems to be common, and
3376 isn't really an error worthy of noting specially, so that particular error
3379 62. When Exim is trying to find all the local interfaces, it used to panic and
3380 die if the ioctl to get the interface flags failed. However, it seems that
3381 on at least one OS (Solaris 9) it is possible to have an interface that is
3382 included in the list of interfaces, but for which you get a failure error
3383 for this call. This happens when the interface is not "plumbed" into a
3384 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3385 failure of the "get flags" call assumes that the interface is down.
3387 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3388 makes life easier for people who are doing arithmetic on fields extracted
3389 from dates, where you often get leading zeros that should not be
3390 interpreted as octal.
3392 64. Added qualify_domain to the redirect router, to override the global
3395 65. If a pathologically long header line contained very many addresses (the
3396 report of this problem mentioned 10 000) and each of them was rewritten,
3397 Exim could use up a very large amount of memory. (It kept on making new
3398 copies of the header line as it rewrote, and never released the old ones.)
3399 At the expense of a bit more processing, the header rewriting function has
3400 been changed so that it no longer eats memory in this way.
3402 66. The generation of the Received: header has been moved from the time that a
3403 message starts to be received, to the time that it finishes. The timestamp
3404 in the Received: header should now be very close to that of the <= log
3405 line. There are two side-effects of this change:
3407 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3408 logged header lines no longer include the local Received: line, because
3409 it has not yet been created. The same applies to a copy of the message
3410 that is returned to a non-SMTP sender when a message is rejected.
3412 (b) When a filter file is tested using -bf, no additional Received: header
3413 is added to the test message. After some thought, I decided that this
3416 This change does not affect the value of $received_for. It is still set
3417 after address rewriting, but before local_scan() is called.
3419 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3421 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3422 gave an unhelpful panic error message, and a defer error. I have managed to
3423 change this behaviour so that it now rejects any supplied certificate,
3424 which seems right, as the list of acceptable certificates is empty.
3426 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3427 gave an unhelpful defer error. I have not managed to make this reject any
3428 supplied certificates, but the error message it gives is "no certificate
3429 supplied", which is not helpful.
3431 70. exigrep's output now also includes lines that are not associated with any
3432 message, but which match the given pattern. Implemented by a patch from
3433 Martin Sluka, which also tidied up the Perl a bit.
3435 71. Recipient callout verification, like sender verification, was using <> in
3436 the MAIL FROM command. This isn't really the right thing, since the actual
3437 sender may affect whether the remote host accepts the recipient or not. I
3438 have changed it to use the actual sender in the callout; this means that
3439 the cache record is now keyed on a recipient/sender pair, not just the
3440 recipient address. There doesn't seem to be a real danger of callout loops,
3441 since a callout by the remote host to check the sender would use <>.
3442 [SEE ABOVE: changed after hitting problems.]
3444 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3445 temporary errors. However, in the case of such a code being given after
3446 the end of a data transmission (i.e. after ".") Exim was failing to write
3447 a retry record for the message. (Yes, there was some broken host that was
3448 actually sending 8xx at this point.)
3450 73. An unknown lookup type in a host list could cause Exim to panic-die when
3451 the list was checked. (An example that provoked this was putting <; in the
3452 middle of a list instead of at the start.) If this happened during a DATA
3453 ACL check, a -D file could be left lying around. This kind of configuration
3454 error no longer causes Exim to die; instead it causes a defer errror. The
3455 incident is still logged to the main and panic logs.
3457 74. Buglet left over from Exim 3 conversion. The message "too many messages
3458 in one connection" was written to the rejectlog but not the mainlog, except
3459 when address rewriting (yes!) was being logged.
3461 75. Added write_rejectlog option.
3463 76. When a system filter was run not as root (that is, when system_filter_user
3464 was set), the values of the $n variables were not being returned to the
3465 main process; thus, they were not subsequently available in the $sn
3468 77. Added +return_path_on_delivery log selector.
3470 78. A connection timeout was being treated differently from recipients deferred
3471 when testing hosts_max_try with a message that was older than the host's
3472 retry timeout. (The host should not be counted, thus allowing all hosts to
3473 be tried at least once before bouncing.) This may have been the cause of an
3474 occasionally reported bug whereby a message would remain on the queue
3475 longer than the retry timeout, but would be bounced if a delivery was
3476 forced. I say "may" because I never totally pinned down the problem;
3477 setting up timeout/retry tests is difficult. See also the next item.
3479 79. The ultimate address timeout was not being applied to errors that involved
3480 a combination of host plus message (for example, a timeout on a MAIL
3481 command). When an address resolved to a number of possible hosts, and they
3482 were not all tried for each delivery (e.g. because of hosts_max_try), a
3483 message could remain on the queue longer than the retry timeout.
3485 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3488 81. Fixed an obscure SMTP outgoing bug which required at least the following
3489 conditions: (a) there was another message waiting for the same server;
3490 (b) the server returned 5xx to all RCPT commands in the first message so
3491 that the message was not completed; (c) the server dropped the connection
3492 or gave a negative response to the RSET that Exim sends to abort the
3493 transaction. The observed case was a dropped connection after DATA that had
3494 been sent in pipelining mode. That is, the server had advertised PIPELINING
3495 but was not implementing it correctly. The effect of the bug was incorrect
3496 behaviour, such as trying another host, and this could lead to a crash.
3502 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3503 and daemon.c were passed as pointers to ints; they should have been
3504 pointers to socklen_t variables (which are typically unsigned ints).
3506 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3509 3. Fixed a really odd bug that affected only the testing scheme; patching a
3510 certain fixed string in the binary changed the value of another string that
3511 happened to be identical to the end of the original first string.
3513 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3514 name", it returns that address as the IP address. On some operating
3515 systems (e.g. Solaris), it also passes back the IP address string as the
3516 "host name". However, on others (e.g. Linux), it passes back an empty
3517 string. Exim wasn't checking for this, and was changing the host name to an
3518 empty string, assuming it had been canonicized.
3520 5. Although rare, it is permitted to have more than one PTR record for a given
3521 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3522 all the names associated with an address, because they do in Solaris.
3523 However, it seems that they do not in Linux for data that comes from the
3524 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3525 I found this out when I moved to a new Linux workstation and tried to run
3526 the Exim test suite.
3528 To get round this problem I have changed the code so that it now does its
3529 own call to the DNS to look up PTR records when searching for a host name.
3530 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3531 addresses that are only in /etc/hosts are still found.
3533 This behaviour is, however, controlled by an option called host_lookup_
3534 order, which defaults to "bydns:byaddr". If people want to use the other
3535 order, or indeed, just use one or the other means of lookup, they can
3536 specify it in this variable.
3538 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3539 some operating systems, this comes back from gethostbyaddr() as an empty
3540 string, and this is what Exim used to test for. However, it seems that in
3541 other systems, "." is yielded. Exim now tests for this case too.
3543 7. The values of check_spool_space and check_log_space are now held internally
3544 as a number of kilobytes instead of an absolute number of bytes. If a
3545 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3546 kilobyte. This means that much larger values can be stored.
3548 8. Exim monitor: an attempt to get the action menu when not actually pointing
3549 at a message produces an empty menu entitled "No message selected". This
3550 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3551 no entries in it ("Shell widget menu has zero width and/or height"). So I
3552 have added a single, blank menu entry in this case.
3554 9. Added ${quote_local_part.
3556 10. MIME decoding is now applied to the contents of Subject: header lines when
3559 11. Now that a reference to $sender_host_address automatically causes a reverse
3560 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3561 host lookup before query-style lookups in lists that might use this
3562 variable. This has therefore been abolished, and the "net-" prefix is no
3563 longer necessary for query-style lookups.
3565 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3566 have been a typo for LFLAGS, so it has been changed.
3568 13. The install script calls Exim with "-C /dev/null" in order to find the
3569 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3570 to be output. Howeve, since Exim outputs its version number before the
3571 error, it didn't break the script. It just looked ugly. I fixed this by
3572 always allowing "-C /dev/null" if the caller is root.
3574 14. Ignore overlarge ACL variable number when reading spool file - insurance
3575 against a later release with more variables having written the file.
3577 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3578 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3579 this, as well as the form without the "IPv6" on the front.
3581 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3582 OS/Makefile-Darwin file.
3584 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3585 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3587 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3588 variable. However, some OS use ints for inodes. I've added cast to long int
3589 to get rid of the compiler warning.
3591 19. I had forgotten to lock out "/../" in configuration file names when
3592 ALT_CONFIG_PREFIX was set.
3594 20. Routers used for verification do not need to specify transports. However,
3595 if such a router generated a host list, and callout was configured, Exim
3596 crashed, because it could not find a port number from the (non-existent)
3597 transport. It now assumes port 25 in this circumstance.
3599 21. Added the -t option to exigrep.
3601 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3602 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3603 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3604 them is set, LOOKUP_LSEARCH is forced.
3606 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3608 24. Added sender and host information to the "rejected by local_scan()" log
3609 line; previously there was no indication of these.
3611 25. Added .include_if_exists.
3613 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3614 Linux. It turns out that not all file systems support this. Apparently some
3615 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3616 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3617 not supported on the file descriptor, is now ignored when Exim is trying to
3618 sync a directory. This applies only to Linux.
3620 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3622 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3623 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3626 29. Added support for maildirsize files from supplied patch (modified a bit).
3628 30. The use of :fail: followed by an empty string could lead Exim to respond to
3629 sender verification failures with (e.g.):
3631 550 Verification failed for <xxx>
3632 550 Sender verify failed
3634 where the first response line was missing the '-' that indicates it is not
3635 the final line of the response.
3637 31. The loop for finding the name of the user that called Exim had a hardwired
3638 limit of 10; it now uses the value of finduser_retries, which is used for
3639 all other user lookups.
3641 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3644 33. Exim was neglecting to zero errno before one call of strtol() when
3645 expanding a string and expecting an integer value. On some systems this
3646 resulted in spurious "integer overflow" errors. Also, it was casting the
3647 result into an int without checking.
3649 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3650 did not work. The code looks as if it has *never* worked, though it appears
3651 to have been documented since at least releast 1.62. I have made it work.
3653 35. The "timeout_DNS" error in retry rules, also documented since at least
3654 1.62, also never worked. As it isn't clear exactly what this means, and
3655 clearly it isn't a major issue, I have abolished the feature by treating it
3656 as "timeout", and writing a warning to the main and panic logs.
3658 36. The display of retry rules for -brt wasn't always showing the error code
3661 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3662 timeout_connect_A, timeout_connect_MX.
3664 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3665 to the empty sender.
3667 39. The daemon was not analysing the content of -oX till after it had closed
3668 stderr and disconnected from the controlling terminal. This meant that any
3669 syntax errors were only noted on the panic log, and the return code from
3670 the command was 0. By re-arranging the code a little, I've made the
3671 decoding happen first, so such errors now appear on stderr, and the return
3672 code is 1. However, the actual setting up of the sockets still happens in
3673 the disconnected process, so errors there are still only recorded on the
3676 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3677 connections (as happens on some IP stacks) was logged at start up time as
3678 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3679 it from "IPv6 and IPv4", which means that two separate sockets are being
3682 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3683 says whether AF_INET or AF_INET6 was passed as an argument.
3685 42. Exiwhat output was messed up when time zones were included in log
3688 43. Exiwhat now gives more information about the daemon's listening ports,
3689 and whether -tls-on-connect was used.
3691 44. The "port" option of the smtp transport is now expanded.
3693 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3694 silently ignored. Now an error message is written to the main and panic
3697 46. There's a new ACL modifier called "logwrite" which writes to a log file
3698 as soon as it is encountered.
3700 47. Added $local_user_uid and $local_user_gid at routing time.
3702 48. Exim crashed when trying to verify a sender address that was being
3705 49. Exim was recognizing only a space character after ".include". It now also
3706 recognizes a tab character.
3708 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3709 extracting the relevant information from the specification. The man page no
3710 longer contains scrambled data for the -d option, and I've added a section
3711 at the front about calling Exim under different names.
3713 51. Added "extra_headers" argument to the "mail" command in filter files.
3715 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3718 53. Installed eximstats 1.29.
3720 54. Added transport_filter_timeout as a generic transport option.
3722 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3723 Cc: header lines. This was required by RFC 822, but it not required by RFC
3726 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3727 incoming messages that did not have them. Now it does so only if the
3728 message originates locally, that is, if there is no associated remote host
3729 address. When Resent- header lines are present, this applies to the Resent-
3730 lines rather than the non-Resent- lines.
3732 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3733 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3735 58. Messages for configuration errors now include the name of the main
3736 configuration file - useful now that there may be more than one file in a
3737 list (.included file names were always shown).
3739 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3740 for those rare installations that do not start the daemon as root or run it
3741 setuid root. I've cut out the call to initgroups() if the daemon is not
3744 60. The Exim user and group can now be bound into the binary as text strings
3745 that are looked up at the start of Exim's processing.
3747 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3749 62. Added $mailstore_basename variable.
3751 63. Installed patch to sieve.c from Michael Haardt.
3753 64. When Exim failed to open the panic log after failing to open the main log,
3754 the original message it was trying to log was written to stderr and debug
3755 output, but if they were not available (the usual case in production), it
3756 was lost. Now it is written to syslog before the two lines that record the
3757 failures to open the logs.
3759 65. Users' Exim filters run in subprocesses under the user's uid. It is
3760 possible for a "deliver" command or an alias in a "personal" command to
3761 provoke an address rewrite. If logging of address rewriting is configured,
3762 this fails because the process is not running as root or exim. There may be
3763 a better way of dealing with this, but for the moment (because 4.30 needs
3764 to be released), I have disabled address rewrite logging when running a
3765 filter in a non-root, non-exim process.
3771 1. The buildconfig auxiliary program wasn't quoting the value set for
3772 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3773 not defined. This bug was masked in 4.22 by the effect that was fixed in
3776 2. Some messages that were rejected after a message id was allocated were
3777 shown as "incomplete" by exigrep. It no longer does this for messages that
3778 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3780 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3781 have allow_domain_literals set, the ID did not get logged in the <= line.
3782 Domain literals are now always recognized in Message-ID: header lines.
3784 4. The first argument for a ${extract expansion item is the key name or field
3785 number. Leading and trailing spaces in this item were not being ignored,
3786 causing some misleading effects.
3788 5. When deliver_drop_privilege was set, single queue runner processes started
3789 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3790 same command in the process it spins off) were not dropping privilege.
3792 6. When the daemon running as "exim" started a queue runner, it always
3793 re-executed Exim in the spun-off process. This is a waste of effort when
3794 deliver_drop_privilege is set. The new process now just calls the
3795 queue-runner function directly.
3801 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3804 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3805 ignored. Though the use of -oP was forcing the writing of a pid file, it
3806 was always written to the default place.
3808 3. If the message "no IP address found for host xxxx" is generated during
3809 incoming verification, it is now followed by identification of the incoming
3810 connection (so you can more easily find what provoked it).
3812 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3814 5. Added some features to "harden" Exim a bit more against certain attacks:
3816 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3817 be put in Local/Makefile. This is like the never_users runtime option,
3818 but it cannot be overridden. The default setting is "root".
3820 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3821 prefix string with which any file named in a -C command line option
3824 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3825 is retained for -C and -D only if the caller of Exim is root. Without
3826 it, the exim user may also use -C and -D and retain privilege.
3828 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3829 command line option is disabled.
3831 6. Macro names set by the -D option must start with an upper case letter, just
3832 like macro names defined in the configuration file.
3834 7. Added "dereference=" facility to LDAP.
3836 8. Two instances of the typo "uknown" in the source files are fixed.
3838 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3839 the Configure-Makefile script screwed up while processing it.
3841 10. Incorporated PCRE 4.4.
3843 11. The SMTP synchronization check was not operating right at the start of an
3844 SMTP session. For example, it could not catch a HELO sent before the client
3845 waited for the greeting. There is now a check for outstanding input at the
3846 point when the greeting is written. Because of the duplex, asynchronous
3847 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3848 way, but not yet received, when the check is performed.
3850 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3851 on TCP/IP sockets, because this apparently causes some broken clients to
3854 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3855 unchanged) from the Cygwin maintainer.
3857 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3858 was supported instead of testing for mbx. Effectively a typo.
3860 15. The spa authenticator server code was not checking that the input it
3861 received was valid base64.
3863 16. The debug output line for the "set" modifier in ACLs was not showing the
3864 name of the variable that was being set.
3866 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3868 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3869 lookup on its own. Something else was needed to trigger the lookup. For
3870 example, a match in host_lookup or the need for a host name in a host list.
3871 Now, if $sender_host_name is referenced and the host name has not yet been
3872 looked up, a lookup is performed. If the lookup fails, the variable remains
3873 empty, and $host_lookup_failed is set to "1".
3875 19. Added "eqi" as a case-independent comparison operator.
3877 20. The saslauthd authentication condition could segfault if neither service
3878 nor realm was specified.
3880 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3881 error message that was logged was misleading, and incoming SMTP
3882 connections were dropped. The message is now more accurate, and temporary
3883 errors are given to SMTP connections.
3885 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3886 (except RSET) with a 421 error, until QUIT is received. However, it was
3887 failing to send a response to QUIT.
3889 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3890 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3891 ACL is now run after the helo_try_verify_hosts code.
3893 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3894 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3895 case-independent, so other case variants are also recognized). Apparently
3896 some systems use these upper case variants.
3898 25. If more than two messages were waiting for the same host, and a transport
3899 filter was specified for the transport, Exim sent two messages over the
3900 same TCP/IP connection, and then failed with "socket operation on non-
3901 socket" when it tried to send the third.
3903 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3905 27. The extern definition of crypt16() in expand.c was not being excluded when
3906 the OS had its own crypt16() function.
3908 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3909 as a preferred synonym for return_size_limit, both as an option and as an
3912 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3914 30. Changed the default configuration ACL to relax the local part checking rule
3915 for addresses that are not in any local domains. For these addresses,
3916 slashes and pipe symbols are allowed within local parts, but the sequence
3917 /../ is explicitly forbidden.
3919 31. SPA server authentication was not clearing the challenge buffer before
3922 32. log_message in a "warn" ACL statement was writing to the reject log as
3923 well as to the main log, which contradicts the documentation and doesn't
3924 seem right (because no rejection is happening). So I have stopped it.
3926 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3927 However, I am unable to do any testing of this.
3929 34. Fixed an infelicity in the appendfile transport. When checking directories
3930 for a mailbox, to see if any needed to be created, it was accidentally
3931 using path names with one or more superfluous leading slashes; tracing
3932 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3934 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3935 contents are added to the log line that is written for every discarded
3936 recipient. (Previously a log_message setting was ignored.)
3938 36. The ${quote: operator now quotes the string if it is empty.
3940 37. The install script runs exim in order to find its version number. If for
3941 some reason other than non-existence or emptiness, which it checks, it
3942 could not run './exim', it was installing it with an empty version number,
3943 i.e. as "exim-". This error state is now caught, and the installation is
3946 38. An argument was missing from the function that creates an error message
3947 when Exim fails to connect to the socket for saslauthd authentication.
3948 This could cause Exim to crash, or give a corrupted message.
3950 39. Added isip, isip4, and isip6 to ${if conditions.
3952 40. The ACL variables $acl_xx are now saved with the message, and can be
3953 accessed later in routers, transports, and filters.
3955 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3956 strings in the file are not string-expanded.
3958 42. If a MAIL command specified a SIZE value that was too large to fit into an
3959 int variable, the check against message_size_limit failed. Such values are
3960 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3961 day this will have to be increased, but I don't think I want to be around
3962 when emails are that large.
3969 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3970 iconv() is not standard in FreeBSD.
3972 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3973 IPv6 enabled. The observed symptom was a segmentation fault on return from
3974 the function os_common_find_running_interfaces() in src/os.c.
3976 3. In the check_special_case() function in daemon.c I had used "errno" as an
3977 argument name, which causes warnings on some systems. This was basically a
3978 typo, since it was named "eno" in the comments!
3980 4. The code that waits for the clock to tick (at a resolution of some fraction
3981 of a second) so as to ensure message-id uniqueness was always waiting for
3982 at least one whole tick, when it could have waited for less. [This is
3983 almost certainly not relevant at current processor speeds, where it is
3984 unlikely to ever wait at all. But we try to future-proof.]
3986 5. The function that sleeps for a time interval that includes fractions of a
3987 second contained a race. It did not block SIGALRM between setting the
3988 timer, and suspending (a couple of lines later). If the interval was short
3989 and the sigsuspend() was delayed until after it had expired, the suspension
3990 never ended. On busy systems this could lead to processes getting stuck for
3993 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3994 process, before it forks any delivery processes. The open lookup caching
3995 mechanism meant that the open file or database connection was passed into
3996 the delivery process. The problem was that delivery processes always tidy
3997 up cached lookup data. This could cause a problem for the next delivery
3998 process started by the queue runner, because the external queue runner
3999 process does not know about the closure. So the next delivery process
4000 still has data in the lookup cache. In the case of a file lookup, there was
4001 no problem because closing a file descriptor in a subprocess doesn't affect
4002 the parent. However, if the lookup was caching a connection to a database,
4003 the connection was closed, and the second delivery process was likely to
4004 see errors such as "PGSQL: query failed: server closed the connection
4005 unexpectedly". The problem has been fixed by closing all cached lookups
4006 in a queue runner before running a delivery process.
4008 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4009 seem to have the "const" qualifier which it has on other OS. I've
4012 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4013 *of the same type* (client or server) with the same public name that an
4014 error should be diagnosed.
4016 9. When Exim looked up a host name for an IP address, but failed to find the
4017 original IP address when looking up the host name (a safety check), it
4018 output the message "<ip address> does not match any IP for NULL", which was
4019 confusing, to say the least. The bug was that the host name should have
4020 appeared instead of "NULL".
4022 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4023 user that is built into the binary, and the -C or -D options is used, root
4024 privilege is dropped before the configuration file is read. In addition,
4025 logging is switched to stderr instead of the normal log files. If the
4026 configuration then re-defines the Exim user, the unprivileged environment
4027 is probably not what is expected, so Exim logs a panic warning message (but
4030 However, if deliver_drop_privilege is set, the unprivileged state may well
4031 be exactly what is intended, so the warning has been cut out in that case,
4032 and Exim is allowed to try to write to its normal log files.
4038 1. smtp_return_error_details was not giving details for temporary sender
4039 or receiver verification errors.
4041 2. Diagnose a configuration error if two authenticators have the same public
4044 3. Exim used not to create the message log file for a message until the first
4045 delivery attempt. This could be confusing when incoming messages were held
4046 for policy or load reasons. The message log file is now created at the time
4047 the message is received, and an initial "Received" line is written to it.
4049 4. The automatically generated man page for command line options had a minor
4050 bug that caused no ill effects; however, a more serious problem was that
4051 the procedure for building the man page automatically didn't always
4052 operate. Consequently, release 4.20 contains an out-of-date version. This
4053 shouldn't happen again.
4055 5. When building Exim with embedded Perl support, the script that builds the
4056 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4057 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4059 6. The freeze_tell option was not being used for messages that were frozen on
4060 arrival, either by an ACL or by local_scan().
4062 7. Added the smtp_incomplete_transaction log selector.
4064 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4065 was accepting AUTH without a new EHLO.
4067 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4068 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4071 10. Logging of TCP/IP connections (when configured) now happens in the main
4072 daemon process instead of the child process, so that the TCP/IP connection
4073 count is more accurate (but it can never be perfect).
4075 11. The use of "drop" in a nested ACL was not being handled correctly in the
4076 outer ACL. Now, if condition failure induced by the nested "drop" causes
4077 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4078 or "require"), the connection is dropped.
4080 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4081 that yield "discard" can now be used with an "accept" or a "discard" verb,
4082 but an error is generated for any others (because I can't see a useful way
4083 to define what should happen).
4085 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4086 are now processed in the same way as lines in the Exim configuration file.
4087 In particular, continuation lines are supported.
4089 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4091 15. Added -ti meaning -t -i.
4093 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4094 domains, and warn by logging if others are found.
4096 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4097 returned from GIFCONF when Exim is trying to find the list of interfaces on
4098 a host. The code in os.c has been modified to copy each ifreq to an aligned
4099 structure in all cases.
4101 Also, in some cases, the returned ifreq's were being copied to a 'struct
4102 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4103 means the last couple of bytes of an IPv6 address could be chopped if the
4104 ifreq contained only a normal sockaddr (14 bytes storage).
4106 18. Named domain lists were not supported in the hosts_treat_as_local option.
4107 An entry such as +xxxx was not recognized, and was treated as a literal
4110 19. Ensure that header lines added by a DATA ACL are included in the reject log
4111 if the ACL subsequently rejects the message.
4113 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4114 MD5 (which is deprecated).
4116 21. When testing a filter file using -bf, Exim was writing a message when it
4117 took the sender from a "From " line in the message, but it was not doing so
4118 when it took $return_path from a Return-Path: header line. It now does.
4120 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4121 with a valid header line field name (a series of printing characters
4122 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4124 23. Changed "disc" in the source to "disk" to conform to the documentation and
4125 the book and for uniformity.
4127 24. Ignore Sendmail's -Ooption=value command line item.
4129 25. When execve() failed while trying to run a command in a pipe transport,
4130 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4131 could be confused with a return value of 69 from the command itself. This
4132 has been changed to 127, the value the shell returns if it is asked to run
4133 a non-existent command. The wording for the related log line suggests a
4134 non-existent command as the problem.
4136 26. If received_header_text expands to an empty string, do not add a Received:
4137 header line to the message. (Well, it adds a token one on the spool, but
4138 marks it "old" so that it doesn't get used or transmitted.)
4140 27. Installed eximstats 1.28 (addition of -nt option).
4142 28. There was no check for failure on the call to getsockname() in the daemon
4143 code. This can fail if there is a shortage of resources on the system, with
4144 ENOMEM, for example. A temporary error is now given on failure.
4146 29. Contrary to the C standard, it seems that in some environments, the
4147 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4148 program. Exim now does this explicitly; it affects the formatting of
4149 timestamps using strftime().
4151 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4152 complaints. I've now initialized all the variables, to avoid this.
4154 32. Header lines added by a system filter were not being "seen" during
4155 transport-time rewrites.
4157 33. The info_callback() function passed to OpenSSL is set up with type void
4158 (*)(SSL *, int, int), as described somewhere. However, when calling the
4159 function (actually a macro) that sets it up, the type void(*)() is
4160 expected. I've put in a cast to prevent warnings from picky compilers.
4162 34. If a DNS black list lookup found a CNAME record, but there were no A
4163 records associated with the domain it pointed at, Exim crashed.
4165 35. If a DNS black list lookup returned more than one A record, Exim ignored
4166 all but the first. It now scans all returned addresses if a particular IP
4167 value is being sought. In this situation, the contents of the
4168 $dnslist_value variable are a list of all the addresses, separated by a
4171 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4172 to accept a host name and all its aliases if the forward lookup for any of
4173 them yielded the IP address of the incoming connection. Now it accepts only
4174 those names whose forward lookup yields the correct IP address. Any other
4175 names are discarded. This closes a loophole whereby a rogue DNS
4176 administrator could create reverse DNS records to break through a
4177 wildcarded host restriction in an ACL.
4179 37. If a user filter or a system filter that ran in a subprocess used any of
4180 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4181 the wrong values were passed to the pipe command ($thisaddress had the
4182 value of $0, $0 had the value of $1, etc). This bug was introduced by
4183 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4185 38. Improved the line breaking for long SMTP error messages from ACLs.
4186 Previously, if there was no break point between 40 and 75 characters, Exim
4187 left the rest of the message alone. Two changes have been made: (a) I've
4188 reduced the minimum length to 35 characters; (b) if it can't find a break
4189 point between 35 and 75 characters, it looks ahead and uses the first one
4190 that it finds. This may give the occasional overlong line, but at least the
4191 remaining text gets split now.
4193 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4194 file descriptors might be low, and that setting 1000 would always raise it.
4195 It turns out that in some environments, the limit is already over 1000 and
4196 that lowering it causes trouble. So now Exim takes care not to decrease it.
4198 40. When delivering a message, the value of $return_path is set to $sender_
4199 address at the start of routing (routers may change the value). By an
4200 oversight, this default was not being set up when an address was tested by
4201 -bt or -bv, which affected the outcome if any router or filter referred to
4204 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4205 log only when "message" or "log_message" are set. However, if one of the
4206 conditions was an address verification, or a call to a nested ACL, the
4207 messages generated by the underlying test were being passed through. This
4208 no longer happens. The underlying message is available in $acl_verify_
4209 message for both "message" and "log_message" expansions, so it can be
4210 passed through if needed.
4212 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4213 new expansion $bh_ to give the encoded byte string without charset
4214 translation. Translation happens only if iconv() is available; HAVE_ICONV
4215 indicates this at build time. HEADERS_CHARSET gives the charset to
4216 translate to; headers_charset can change it in the configuration, and
4217 "headers charset" can change it in an individual filter file.
4219 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4220 that creates RFC 2047 encoded "words" labels them as that charset instead
4221 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4222 expansion operator; (ii) when Exim creates a From: line for a local
4223 message; (iii) when a header line is rewritten to include a "phrase" part.
4225 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4226 buggy, causing it to skip the first lines of messages whose message ID
4227 ended in 'D'. This would not have bitten before Exim release 4.14, because
4228 message IDs were unlikely to end in 'D' before then. The effect was to have
4229 incorrect size information for certain domains.
4231 45. #include "config.h" was missing at the start of the crypt16.c module. This
4232 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4235 46. If there was a timeout during a "random" callout check, Exim treated it as
4236 a failure of the random address, and carried on sending RSET and the real
4237 address. If the delay was just some slowness somewhere, the response to the
4238 original RCPT would be taken as a response to RSET and so on, causing
4239 mayhem of various kinds.
4241 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4242 when I implemented it. It didn't allow for the fact that some option values
4243 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4244 the right test for positive values.
4246 48. Domain names in DNS records are case-independent. Exim always looks them up
4247 in lower case. Some resolvers return domain names in exactly the case they
4248 appear in the zone file, that is, they may contain uppercase letters. Not
4249 all resolvers do this - some return always lower case. Exim was treating a
4250 change of case by a resolver as a change of domain, similar to a widening
4251 of a domain abbreviation. This triggered its re-routing code and so it was
4252 trying to route what was effectively the same domain again. This normally
4253 caused routing to fail (because the router wouldn't handle the domain
4254 twice). Now Exim checks for this case specially, and just changes the
4255 casing of the domain that it ultimately uses when it transmits the message
4258 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4261 50. If a filter generated a file delivery with a non-absolute name (possible if
4262 no home directory exists for the router), the forbid_file option was not
4265 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4266 the existing equality matching.
4268 52. Exim was using ints instead of ino_t variables in some places where it was
4269 dealing with inode numbers.
4271 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4272 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4273 called TMPDIR, and if it finds it is different, it changes its value.
4275 54. The smtp_printf() function is now made available to local_scan() so
4276 additional output lines can be written before returning. There is also an
4277 smtp_fflush() function to enable the detection of a dropped connection.
4278 The variables smtp_input and smtp_batched_input are exported to
4281 55. Changed the default runtime configuration: the message "Unknown user"
4282 has been removed from the ACL, and instead placed on the localuser router,
4283 using the cannot_route_message feature. This means that any verification
4284 failures that generate their own messages won't get overridden. Similarly,
4285 the "Unrouteable address" message that was in the ACL for unverifiable
4286 relay addresses has also been removed.
4288 56. Added hosts_avoid_esmtp to the smtp transport.
4290 57. The exicyclog script was not checking for the esoteric option
4291 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4292 will work only if exicyclog is run under the appropriate euid.
4294 58. Following a discussion on the list, the rules by which Exim recognises line
4295 endings on incoming messages have been changed. The -dropcr and drop_cr
4296 options are now no-ops, retained only for backwards compatibility. The
4297 following line terminators are recognized: LF CRLF CR. However, special
4298 processing applies to CR:
4300 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4301 nor a local message in the state where . is a terminator.
4303 (ii) If a bare CR is encountered in a header line, an extra space is added
4304 after the line terminator so as not to end the header. The reasoning
4305 behind this is that bare CRs in header lines are most likely either
4306 to be mistakes, or people trying to play silly games.
4308 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4309 was being incorrectly given as 18 bytes larger than it should have been.
4310 This is a VOB (very old bug).
4312 60. This may never have affected anything current, but just in case it has:
4313 When the local host is found other than at the start of a list of hosts,
4314 the local host, those with the same MX, and any that follow, are discarded.
4315 When the list in question was part of a longer list of hosts, the following
4316 hosts (not currently being processed) were also being discarded. This no
4317 longer happens. I'm not sure if this situation could ever has previously
4320 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4323 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4324 header. This is something that is recommended in a new Internet Draft, and
4325 is something that is documented as being done by Sendmail. There are two
4326 possible values. For messages generated by the autoreply transport, Exim
4329 Auto-Submitted: auto-replied
4331 whereas for all other generated messages (e.g. bounces) it adds
4333 Auto-Submitted: auto-generated
4335 63. The "personal" condition in filters now includes a test for the
4336 Auto-Submitted: header. If it contains the string "auto-" the message it
4337 not considered personal.
4339 64. Added rcpt_include_affixes as a generic transport option.
4341 65. Added queue_only_override (default true).
4343 66. Added the syslog_duplication option.
4345 67. If what should have been the first header line of a message consisted of
4346 a space followed by a colon, Exim was mis-interpreting it as a header line.
4347 It isn't of course - it is syntactically invalid and should therefore be
4348 treated as the start of the message body. The misbehaviour could have
4349 caused a number of strange effects, including loss of data in subsequent
4350 header lines, and spool format errors.
4352 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4353 client host had authenticated. This control can now be exercised by an ACL
4354 for more flexibility.
4356 69. By default, callouts do not happen when testing with -bh. There is now a
4357 variant, -bhc, which does actually run the callout code, including
4358 consulting and updating the callout cache.
4360 70. Added support for saslauthd authentication, courtesy of Alexander
4363 71. If statvfs() failed on the spool or log directories while checking their
4364 size for availability, Exim confusingly gave the error "space shortage".
4365 Furthermore, in debugging mode it crashed with a floating point exception.
4366 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4367 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4368 serious problem, Exim now writes to the main and panic logs when this
4369 happens, with details of the failure. It then refuses to accept the
4370 incoming message, giving the message "spool directory problem" or "log
4371 directory problem" with a 421 code for SMTP messages.
4373 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4374 0, 1, and 2 exist, because some OS complain for execs without them (see
4375 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4376 descriptors. However, the code omitted to check that the open succeeded,
4377 causing mysterious errors if for some reason the permissions on /dev/null
4378 got screwed. Now Exim writes a message to the main and panic logs, and
4379 bombs out if it can't open /dev/null.
4381 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4382 interact so that it is all more flexible. It is supposed to remain
4383 backwards compatible. Also added extra_local_interfaces.
4385 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4386 to bomb out with an assertion failure - to the client this appears as a
4387 connection drop. This problem occurs in the part of the code that was taken
4388 from the Samba project. Fortunately, the assertion is in a very simple
4389 function, so I have fixed this by reproducing the function inline in the
4390 one place where it is called, and arranging for authentication to fail
4391 instead of killing the process with assert().
4393 75. The SPA client code was not working when the server requested OEM rather
4394 than Unicode encoding.
4396 76. Added code to make require_files with a specific uid setting more usable in
4397 the case where statting the file as root fails - usually a non-root-mounted
4398 NFS file system. When this happens and the failure is EACCES, Exim now
4399 forks a subprocess and does the per-uid checking as the relevant uid.
4401 77. Added process_log_path.
4403 78. If log_file_path was not explicitly set, a setting of check_log_space or
4404 check_log_inodes was ignored.
4406 79. If a space check for the spool or log partitions fails, the incident is now
4407 logged. Of course, in the latter case the data may get lost...
4409 80. Added the %p formatting code to string_format() so that it can be used to
4410 print addresses in debug_print(). Adjusted all the address printing in the
4411 debugging in store.c to use %p rather than %d.
4413 81. There was a concern that a line of code in smtp_in.c could overflow a
4414 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4415 initially expressed, the concern was not well-founded, because trailing
4416 spaces are removed early. However, if the trailing spaces were followed by
4417 a NULL, they did not get removed, so the overflow was possible. Two fixes
4420 (a) I re-wrote the offending code in a cleaner fashion.
4421 (b) If an incoming SMTP command contains a NULL character, it is rejected
4424 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4425 runs initgroups(), so that if the Exim user is in any additional groups,
4426 they will be used during message reception.
4432 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob