1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.433 2006/11/13 12:29:30 ph10 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
10 leftover -K file (the existence of which was triggered by #402).
11 While we were at it, introduced process PID as part of the -K
12 filename. This should rule out race conditions when creating
15 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
16 processing considerably. Previous code took too long for large mails,
17 triggering a timeout which in turn triggers #401.
19 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
20 in the DK code in transports.c. sendfile() is not really portable,
21 hence the _LINUX specificness.
23 TF/01 In the add_headers option to the mail command in an Exim filter,
24 there was a bug that Exim would claim a syntax error in any
25 header after the first one which had an odd number of characters
28 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
29 callout verification, Exim cached a "reject" for the entire domain. This
30 is correct for most verifications, but it is not correct for a recipient
31 verification with use_sender or use_postmaster set, because in that case
32 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
33 case of MAIL FROM:<> rejection from other early rejections (e.g.
34 rejection of HELO). When verifying a recipient using a non-null MAIL
35 address, the cache is ignored if it shows MAIL FROM:<> rejection.
36 Whatever the result of the callout, the value of the domain cache is
37 left unchanged (for any other kind of callout, getting as far as trying
38 RCPT means that the domain itself is ok).
40 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
43 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
44 manifest itself as EPIPE rather than ECONNECT. When tidying away a
45 session, the daemon ignores ECONNECT errors and logs others; it now
46 ignores EPIPE as well.
48 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
49 (quoted-printable decoding).
51 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
52 later the small subsequent patch to fix an introduced bug.
54 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
56 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
58 PH/08 An error is now given if message_size_limit is specified negative.
60 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
61 to be given (somewhat) arbitrary names.
63 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
66 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
67 miscellaneous code fixes
69 PH/10 Added the log_reject_target ACL modifier to specify where to log
72 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
73 hostname. This is wrong, because it relates to the incoming message (and
74 probably the interface on which it is arriving) and not to the outgoing
75 callout (which could be using a different interface). This has been
76 changed to use the value of the helo_data option from the smtp transport
77 instead - this is what is used when a message is actually being sent. If
78 there is no remote transport (possible with a router that sets up host
79 addresses), $smtp_active_hostname is used.
81 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
82 tweaks were necessary in order to get it to work (see also 21 below):
83 (a) The code assumed that strncpy() returns a negative number on buffer
84 overflow, which isn't the case. Replaced with Exim's string_format()
86 (b) There were several signed/unsigned issues. I just did the minimum
87 hacking in of casts. There is scope for a larger refactoring.
88 (c) The code used strcasecmp() which is not a standard C function.
89 Replaced with Exim's strcmpic() function.
90 (d) The code set only $1; it now sets $auth1 as well.
91 (e) A simple test gave the error "authentication client didn't specify
92 service in request". It would seem that Dovecot has changed its
93 interface. Fortunately there's a specification; I followed it and
94 changed what the client sends and it appears to be working now.
96 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
99 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
100 address A is aliased to B and C, where B exists and C does not. Without
101 -v the output is "A verified" because verification stops after a
102 successful redirection if more than one address is generated. However,
103 with -v the child addresses are also verified. Exim was outputting "A
104 failed to verify" and then showing the successful verification for C,
105 with its parentage. It now outputs "B failed to verify", showing B's
106 parentage before showing the successful verification of C.
108 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
109 look up a TXT record in a specific list after matching in a combined
112 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
113 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
114 they consult the DNS. I had assumed they would set it the way they
115 wanted; and indeed my experiments on Linux seem to show that in some
116 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
117 To be on the safe side, however, I have now made the interface to
118 host_find_byname() similar to host_find_bydns(), with an argument
119 containing the DNS resolver options. The host_find_byname() function now
120 sets these options at its start, just as host_find_bydns() does. The smtp
121 transport options dns_qualify_single and dns_search_parents are passed to
122 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
123 of host_find_byname() use the default settings of RES_DEFNAMES
124 (qualify_single) but not RES_DNSRCH (search_parents).
126 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
127 spool_read_header() do less string testing, by means of a preliminary
128 switch on the second character of optional "-foo" lines. (This is
129 overdue, caused by the large number of possibilities that now exist.
130 Originally there were few.) While I was there, I also converted the
131 str(n)cmp tests so they don't re-test the leading "-" and the first
132 character, in the hope this might squeeze out yet more improvement.
134 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
135 flag allowing group syntax was set by the header_syntax check but not
136 turned off, possible causing trouble later; (2) The flag was not being
137 set at all for the header_verify test, causing "group"-style headers to
138 be rejected. I have now set it in this case, and also caused header_
139 verify to ignore an empty address taken from a group. While doing this, I
140 came across some other cases where the code for allowing group syntax
141 while scanning a header line wasn't quite right (mostly, not resetting
142 the flag correctly in the right place). These bugs could have caused
143 trouble for malformed header lines. I hope it is now all correct.
145 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
146 with the "reply" argument non-NULL. The code, however (which originally
147 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
148 but it didn't always do it. This confused somebody who was copying the
149 code for some other use. I have removed all the tests.
151 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
152 feature that was used to support insecure browsers during the U.S. crypto
153 embargo. It requires special client support, and Exim is probably the
154 only MTA that supported it -- and would never use it because real RSA is
155 always available. This code has been removed, because it had the bad
156 effect of slowing Exim down by computing (never used) parameters for the
157 RSA_EXPORT functionality.
159 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
160 authenticator to fail if there's a tab character in the incoming data
161 (there should never be unless someone is messing about, as it's supposed
162 to be base64-encoded). Also added, on Timo's advice, the "secured" option
163 if the connection is using TLS or if the remote IP is the same as the
164 local IP, and the "valid-client-cert option" if a client certificate has
167 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
168 authenticators. This can be used for authorization after authentication
169 succeeds. (In the case of plaintext, it servers for both authentication
172 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
173 if any retry times were supplied.
175 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
176 connection, where there is no client IP address to check. In this
177 situation, the verify now always succeeds.
179 PH/25 Applied John Jetmore's -Mset patch.
181 PH/26 Added -bem to be like -Mset, but loading a message from a file.
183 PH/27 In a string expansion for a processed (not raw) header when multiple
184 headers of the same name were present, leading whitespace was being
185 removed from all of them, but trailing whitespace was being removed only
186 from the last one. Now trailing whitespace is removed from each header
187 before concatenation. Completely empty headers in a concatenation (as
190 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
191 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
193 PH/29 [Removed. This was a change that I later backed out, and forgot to
194 correct the ChangeLog entry (that I had efficiently created) before
195 committing the later change.]
197 PH/30 Exim was sometimes attempting to deliver messages that had suffered
198 address errors (4xx response to RCPT) over the same connection as other
199 messages routed to the same hosts. Such deliveries are always "forced",
200 so retry times are not inspected. This resulted in far too many retries
201 for the affected addresses. The effect occurred only when there were more
202 hosts than the hosts_max_try setting in the smtp transport when it had
203 the 4xx errors. Those hosts that it had tried were not added to the list
204 of hosts for which the message was waiting, so if all were tried, there
205 was no problem. Two fixes have been applied:
207 (i) If there are any address or message errors in an SMTP delivery, none
208 of the hosts (tried or untried) are now added to the list of hosts
209 for which the message is waiting, so the message should not be a
210 candidate for sending over the same connection that was used for a
211 successful delivery of some other message. This seems entirely
212 reasonable: after all the message is NOT "waiting for some host".
213 This is so "obvious" that I'm not sure why it wasn't done
214 previously. Hope I haven't missed anything, but it can't do any
215 harm, as the worst effect is to miss an optimization.
217 (ii) If, despite (i), such a delivery is accidentally attempted, the
218 routing retry time is respected, so at least it doesn't keep
219 hammering the server.
221 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
222 in ${readsocket because some servers need this prod.
224 PH/32 Added some extra debug output when updating a wait-xxx database.
226 PH/33 The hint "could be header name not terminated by colon", which has been
227 given for certain expansion errors for a long time, was not being given
228 for the ${if def:h_colon_omitted{... case.
230 PH/34 The spec says: "With one important exception, whenever a domain list is
231 being scanned, $domain contains the subject domain." There was at least
232 one case where this was not true.
234 PH/35 The error "getsockname() failed: connection reset by peer" was being
235 written to the panic log as well as the main log, but it isn't really
236 panic-worthy as it just means the connection died rather early on. I have
237 removed the panic log writing for the ECONNRESET error when getsockname()
240 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
241 runs only) independently of the message's sender address. This meant
242 that, if the 4xx error was in fact related to the sender, a different
243 message to the same recipient with a different sender could confuse
244 things. In particualar, this can happen when sending to a greylisting
245 server, but other circumstances could also provoke similar problems.
246 I have changed the default so that the retry time for these errors is now
247 based a combination of the sender and recipient addresses. This change
248 can be overridden by setting address_retry_include_sender=false in the
251 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
252 remote server are returned as part of bounce messages. This was not
253 happening for LMTP over a pipe (the lmtp transport), but now it is the
254 same for both kinds of LMTP.
256 PH/38 Despite being documented as not happening, Exim was rewriting addresses
257 in header lines that were in fact CNAMEs. This is no longer the case.
259 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
260 and queue runs started by the daemon processed all messages. This has
261 been fixed so that -R and -S can now usefully be given with -q<time>.
263 PH/40 Import PCRE release 6.7 (fixes some bugs).
265 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
267 PH/42 Give an error if -q is specified more than once.
269 PH/43 Renamed the variables $interface_address and $interface_port as
270 $received_ip_address and $received_port, to make it clear that these
271 values apply to message reception, and not to the outgoing interface when
272 a message is delivered. (The old names remain recognized, of course.)
274 PH/44 There was no timeout on the connect() call when using a Unix domain
275 socket in the ${readsocket expansion. There now is.
281 SC/01 Use a glob alias rather than an array ref in eximstats generated
282 parser. This improves both readability and performance.
284 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
285 Don't display local sender or destination tables in eximstats unless
286 there is data to show.
287 Added average volumes into the eximstats top table text output.
289 SC/03 Collect data on the number of addresses (recipients) as well
290 as the number of messages in eximstats.
292 TF/01 Correct an error in the documentation for the redirect router. Exim
293 does (usually) call initgroups() when daemonizing.
295 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
296 with consistent privilege compared to when running as a daemon.
298 TF/03 Note in the spec that $authenticated_id is not set for local
299 submissions from trusted users.
301 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
302 Thanks to Dean Brooks <dean@iglou.com> for the patch.
304 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
305 by adding some example configuration directives to the default
306 configuration file. A little bit of work is required to uncomment the
307 directives and define how usernames and passwords are checked, but
308 there is now a framework to start from.
310 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
311 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
312 without this. I don't know how relevant this is to other LDAP libraries.
314 PH/02 Add the verb name to the "unknown ACL verb" error.
316 PH/03 Magnus Holmgren's patch for filter_prepend_home.
318 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
320 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
321 directory not expanded when it should be if an expanded home directory
322 was set for the address (which is overridden by the transport).
324 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
327 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
328 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
329 because it is too late at that time, and has no effect.
331 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
332 security issue with \' (bugzilla #107). I could not use the
333 PQescapeStringConn() function, because it needs a PGconn value as one of
336 PH/08 When testing addresses using -bt, indicate those final addresses that
337 are duplicates that would not cause an additional delivery. At least one
338 person was confused, thinking that -bt output corresponded to deliveries.
339 (Suppressing duplicates isn't a good idea as you lose the information
340 about possibly different redirections that led to the duplicates.)
342 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
343 systems where poll() doesn't work, in particular OS X.
345 PH/10 Added more information to debugging output for retry time not reached.
347 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
348 operations in malware.c.
350 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
353 PH/13 If write_rejectlog was set false when logging was sent to syslog with
354 syslog_duplication set false, log lines that would normally be written
355 both the the main log and to the reject log were not written to syslog at
358 PH/14 In the default configuration, change the use of "message" in ACL warn
359 statements to "add_header".
361 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
362 not followed by a command (e.g. "seen endif").
364 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
365 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
368 PH/17 Added extra conditions to the default value of delay_warning_condition
372 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
373 { match{$h_precedence:}{(?i)bulk|list|junk} } \
374 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
377 The Auto-Submitted: and various List- headers are standardised, whereas I
378 don't think Precedence: ever was.
380 PH/18 Refactored debugging code in route_finduser() to show more information,
381 in particular, the error code if getpwnam() issues one.
383 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
384 This is apparently needed in addition to the PH/07 change above to avoid
385 any possible encoding problems.
387 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
388 but not after initializing Perl.
390 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
391 output them only if debugging. By default they are written stderr,
392 apparently, which is not desirable.
394 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
397 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
400 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
402 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
403 authenticated or an ident call has been made. Suppress the default
404 values for $authenticated_id and $authenticated_sender (but permit -oMai
405 and -oMas) when testing with -bh.
407 PH/24 Re-jigged the order of the tests in the default configuration so that the
408 tests for valid domains and recipients precede the DNS black list and CSA
409 tests, on the grounds that those ones are more expensive.
411 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
412 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
413 This bug exists in every version of Exim that I still have, right back to
416 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
417 However, an attempt to turn on case-sensitivity in a regex key by
418 including (?-i) didn't work because the subject string was already
419 lowercased, and the effects were non-intuitive. It turns out that a
420 one-line patch can be used to allow (?-i) to work as expected.
426 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
427 other effects) broke the use of negated acl sub-conditions.
429 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
432 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
433 "Deny" causes Exim to reject the incoming connection with a 554 error.
434 Unfortunately, if there is a major crisis, such as a disk failure,
435 tcp-wrappers gives "deny", whereas what one would like would be some
436 kind of temporary error. A kludge has been added to help with this.
437 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
438 554 error is used if errno is still zero or contains ENOENT (which occurs
439 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
442 PH/03 Add -lutil to the default FreeBSD LIBS setting.
444 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
445 errors. Otherwise a message that provokes a temporary error (when other
446 messages do not) can cause a whole host to time out.
448 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
449 addresses were routed directly to files or pipes from a redirect router.
450 File deliveries just didn't batch; pipe deliveries might have suffered
453 PH/06 A failure to get a lock for a hints database would erroneously always say
454 "Failed to get write lock", even when it was really a read lock.
456 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
457 of 0600. This has been changed to use the value of the lockfile_mode
458 option (which defaults to 0600).
460 PH/08 Applied small patch from the Sieve maintainer.
462 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
463 folder from quota calculations, a direct delivery into this folder messed
464 up the contents of the maildirsize file. This was because the regex was
465 used only to exclude .Trash (or whatever) when the size of the mailbox
466 was calculated. There was no check that a delivery was happening into an
467 excluded directory. This bug has been fixed by ignoring all quota
468 processing for deliveries into excluded directories.
470 PH/10 Added the maildirfolder_create_regex option to appendfile.
476 PH/01 The code for finding all the local interface addresses on a FreeBSD
477 system running IPv6 was broken. This may well have applied to all BSD
478 systems, as well as to others that have similar system calls. The broken
479 code found IPv4 interfaces correctly, but gave incorrect values for the
480 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
481 that it would not match correctly against @[] and not recognize the IPv6
484 PH/02 The ipliteral router was not recognizing addresses of the form user@
485 [ipv6:....] because it didn't know about the "ipv6:" prefix.
487 PH/03 Added disable_ipv6.
489 PH/04 Changed $reply_address to use the raw form of the headers instead of the
490 decoded form, because it is most often used to construct To: headers
491 lines in autoreplies, and the decoded form may well be syntactically
492 invalid. However, $reply_address has leading white space removed, and all
493 newlines turned into spaces so that the autoreply transport does not
496 PH/05 If group was specified without a user on a router, and no group or user
497 was specified on a transport, the group from the router was ignored.
499 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
500 for visible compile-time settings that can be used to change these
501 numbers, for those that want even more. Backwards compatibility with old
502 spool files has been maintained. However, going back to a previous Exim
503 release will lost any variables that are in spool files.
505 PH/07 Two small changes when running in the test harness: increase delay when
506 passing a TCP/IP connection to a new process, in case the original
507 process has to generate a bounce, and remove special handling of
508 127.0.0.2 (sic), which is no longer necessary.
510 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
511 be the same on different OS.
513 PH/09 Moved a debug statement in filter processing to avoid a race problem when
516 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
517 whether --show-vars was specified or not
519 JJ/02 exipick: Added support for new ACL variable spool format introduced
522 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
523 syntactically invalid From: or Reply-to: line, and a filter used this to
524 generate an autoreply, and therefore failed to obtain an address for the
525 autoreply, Exim could try to deliver to a non-existent relative file
526 name, causing unrelated and misleading errors. What now happens is that
527 it logs this as a hard delivery error, but does not attempt to create a
530 PH/11 The exinext utility has a -C option for testing purposes, but although
531 the given file was scanned by exinext itself; it wasn't being passed on
532 when Exim was called.
534 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
535 an end-of-file indication when reading a command response.
537 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
538 compiled. In many other places in Exim, IPv6 addresses are always
539 recognized, so I have changed this. It also means that IPv4 domain
540 literals of the form [IPV4:n.n.n.n] are now always recognized.
542 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
543 used if the router is not running as root, for example, when verifying at
544 ACL time, or when using -bh. The debugging output from this situation was
545 non-existent - all you got was a failure to exec. I have made two
548 (a) Failures to set uid/gid, the current directory, or a process leader
549 in a subprocess such as that created by queryprogram now generate
550 suitable debugging ouput when -d is set.
552 (b) The queryprogram router detects when it is not running as root,
553 outputs suitable debugging information if -d is set, and then runs
554 the subprocess without attempting to change uid/gid.
556 PH/15 Minor change to Makefile for building test_host (undocumented testing
559 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
560 additional section of a DNS packet that returns MX or SRV records.
561 Instead, it always explicitly searches for A/AAAA records. This avoids
562 major problems that occur when a DNS server includes only records of one
563 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
564 fixed another bug: if SRV records were looked up and the corresponding
565 address records were *not* found in the additional section, the port
566 values from the SRV records were lost.
568 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
569 using the correct key (the original address) when searching the retry
570 rules in order to find which one to use for generating the retry hint.
572 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
573 adding the default one. Similarly, if it contains a Reply-To: header, the
574 errors_reply_to option, if set, is not used.
576 PH/19 When calculating a retry time, Exim used to measure the "time since
577 failure" by looking at the "first failed" field in the retry record. Now
578 it does not use this if it is later than than the arrival time of the
579 message. Instead it uses the arrival time. This makes for better
580 behaviour in cases where some deliveries succeed, thus re-setting the
581 "first failed" field. An example is a quota failure for a huge message
582 when small messages continue to be delivered. Without this change, the
583 "time since failure" will always be short, possible causing more frequent
584 delivery attempts for the huge message than are intended.
585 [Note: This change was subsequently modified - see PH/04 for 4.62.]
587 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
588 $1, $2, $3) because the numerical variables can be reset during some
589 expansion items (e.g. "match"), thereby losing the authentication data.
591 PH/21 Make -bV show the size of off_t variables so that the test suite can
592 decide whether to run tests for quotas > 2G.
594 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
595 mailbox_size, and mailbox_filecount in the appendfile transport. If a
596 filecount value is greater than 2G or if a quota value is greater than 2G
597 on a system where the size of off_t is not greater than 4, a panic error
600 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
601 never match. The debug and -bh output now contains an explicit error
602 message indicating a malformed IPv4 address or mask.
604 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
605 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
608 PH/25 Do not write to syslog when running in the test harness. The only
609 occasion when this arises is a failure to open the main or panic logs
610 (for which there is an explicit test).
612 PH/26 Added the /no_tell option to "control=freeze".
614 PH/27 If a host name lookup failed very early in a connection, for example, if
615 the IP address matched host_lookup and the reverse lookup yielded a name
616 that did not have a forward lookup, an error message of the form "no IP
617 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
618 could be logged. Now it outputs the IP address instead of "NULL".
620 PH/28 An enabling patch from MH: add new function child_open_exim2() which
621 allows the sender and the authenticated sender to be set when
622 submitting a message from within Exim. Since child_open_exim() is
623 documented for local_scan(), the new function should be too.
625 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
626 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
627 results in an empty string is now treated as unset.
629 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
631 PH/31 Added sender_verify_fail logging option.
633 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
634 needed by RFC 822 but not by RFC 2822 was commented out. I have now
635 tidied the source and removed it altogether.
637 PH/33 When a queue run was abandoned because the load average was too high, a
638 log line was always written; now it is written only if the queue_run log
639 selector is set. In addition, the log line for abandonment now contains
640 information about the queue run such as the pid. This is always present
641 in "start" and "stop" lines but was omitted from the "abandon" line.
643 PH/34 Omit spaces between a header name and the colon in the error message that
644 is given when verify = headers_syntax fails (if there are lots of them,
645 the message gets confusing).
647 PH/35 Change the default for dns_check_names_pattern to allow slashes within
648 names, as there are now some PTR records that contain slashes. This check
649 is only to protect against broken name servers that fall over on strange
650 characters, so the fact that it applies to all lookups doesn't matter.
652 PH/36 Now that the new test suite is complete, we can remove some of the
653 special code in Exim that was needed for the old test suite. For example,
654 sorting DNS records because real resolvers return them in an arbitrary
655 order. The new test suite's fake resolver always returns records in the
658 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
659 bounces) except when queue_only is set, to avoid logging races between
660 the different processes.
662 PH/38 Panic-die if .include specifies a non-absolute path.
664 PH/39 A tweak to the "H" retry rule from its user.
666 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
667 a label. They prevented compilation on older perls.
669 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
670 a warning to be raised on newish perls.
672 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
673 on queue. Changes to match documented behaviour of showing count of
674 messages matching specified criteria.
676 PH/40 Changed the default ident timeout from 30s to 5s.
678 PH/41 Added support for the use of login_cap features, on those BSD systems
679 that have them, for controlling the resources used by pipe deliveries.
681 PH/42 The content-scanning code uses fopen() to create files in which to put
682 message data. Previously it was not paying any attention to the mode of
683 the files. Exim runs with umask(0) because the rest of the code creates
684 files with open(), and sets the required mode explicitly. Thus, these
685 files were ending up world-writeable. This was not a big issue, because,
686 being within the spool directory, they were not world-accessible. I have
687 created a function called modefopen, which takes an additional mode
688 argument. It sets umask(777), creates the file, chmods it to the required
689 mode, then resets the umask. All the relevant calls to fopen() in the
690 content scanning code have been changed to use this function.
692 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
693 to 24 hours. This avoids potential overflow problems when processing G
694 and H retry rules. I suspect nobody ever tinkers with this value.
696 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
698 PH/45 When the plaintext authenticator is running as a client, the server's
699 challenges are checked to ensure they are valid base64 strings. By
700 default, the authentication attempt is cancelled if an invalid string is
701 received. Setting client_ignore_invalid_base64 true ignores these errors.
702 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
703 they are received. Thus, the responses can be made to depend on the
704 challenges. If an invalid string is ignored, an empty string is placed in
707 PH/46 Messages that are created by the autoreply transport now contains a
708 References: header, in accordance with RFCs 2822 and 3834.
710 PH/47 Added authenticated_sender_force to the smtp transport.
712 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
714 PH/49 Installed latest patch from the Sieve maintainer.
716 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
717 was also set, the appendfile transport was unnecessarily scanning a
718 directory of message files (e.g. for maildir delivery) to find the count
719 of files (along with the size), even though it did not need this
720 information. It now does the scan only if it needs to find either the
721 size of the count of files.
723 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
725 PH/52 Two bugs concerned with error handling when the smtp transport is
728 (i) Exim was not creating retry information for temporary errors given
729 for individual recipients after the DATA command when the smtp transport
730 was used in LMTP mode. This meant that they could be retried too
731 frequently, and not timed out correctly.
733 (ii) Exim was setting the flag that allows error details to be returned
734 for LMTP errors on RCPT commands, but not for LMTP errors for individual
735 recipients that were returned after the DATA command.
737 PH/53 This is related to PH/52, but is more general: for any failing address,
738 when detailed error information was permitted to be returned to the
739 sender, but the error was temporary, then after the final timeout, only
740 "retry timeout exceeded" was returned. Now it returns the full error as
741 well as "retry timeout exceeded".
743 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
744 do this, and (what is worse) MTAs that accept it.
746 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
747 will now be deprecated.
749 PH/56 New os.c-cygwin from the Cygwin maintainer.
751 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
752 formats (previously only available in exim formats via -bpr, -bpru,
753 and -bpra. Now also available in native and exiqgrep formats)
755 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
756 with very large, slow to parse queues
758 JJ/08 exipick: added ! as generic prefix to negate any criteria format
760 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
762 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
763 responses to authentication challenges, though it was showing the
764 challenges; (ii) I've removed the CR characters from the debug output for
767 PH/58 Allow for the insertion of a newline as well as a space when a string
768 is turned into more than one encoded-word during RFC 2047 encoding. The
769 Sieve code now uses this.
771 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
772 data_4xx, lost_connection, tls_required.
774 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
775 message was being sent as an SMTP response.
777 PH/61 Add -l and -k options to exicyclog.
779 PH/62 When verifying, if an address was redirected to one new address, so that
780 verification continued, and the new address failed or deferred after
781 having set something in $address_data, the value of $address_data was not
782 passed back to the ACL. This was different to the case when no
783 redirection occurred. The value is now passed back in both cases.
785 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
786 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
787 use login_cap.h, so on its own it isn't the distinguishing feature. The
788 new name refers directly to the setclassresources() function.
790 PH/65 Added configuration files for NetBSD3.
792 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
794 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
795 is preferred over IPv4.
797 PH/68 The bounce_return_message and bounce_return_body options were not being
798 honoured for bounces generated during the reception of non-SMTP messages.
799 In particular, this applied to messages rejected by the ACL. This bug has
800 been fixed. However, if bounce_return_message is true and bounce_return_
801 body is false, the headers that are returned for a non-SMTP message
802 include only those that have been read before the error was detected.
803 (In the case of an ACL rejection, they have all been read.)
805 PH/69 The HTML version of the specification is now built in a directory called
806 spec_html instead of spec.html, because the latter looks like a path with
807 a MIME-type, and this confuses some software.
809 PH/70 Catch two compiler warnings in sieve.c.
811 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
812 function verify_get_ident() calls ip_connect() to connect a socket, but
813 if the "connect()" function timed out, ip_connect() used to close the
814 socket. However, verify_get_ident() also closes the socket later, and in
815 between Exim writes to the log, which may get opened at this point. When
816 the socket was closed in ip_connect(), the log could get the same file
817 descriptor number as the socket. This naturally causes chaos. The fix is
818 not to close the socket in ip_connect(); the socket should be closed by
819 the function that creates it. There was only one place in the code where
820 this was missing, in the iplookup router, which I don't think anybody now
821 uses, but I've fixed it anyway.
823 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
824 well as to direct DNS lookups. Otherwise the handling of names in host
825 lists is inconsistent and therefore confusing.
831 PH/01 Two changes to the default runtime configuration:
833 (1) Move the checks for relay_from_hosts and authenticated clients from
834 after to before the (commented out) DNS black list checks.
836 (2) Add control=submission to the relay_from_hosts and authenticated
837 clients checks, on the grounds that messages accepted by these
838 statements are most likely to be submissions.
840 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
842 (1) Generate an error if the third argument for the ${prvs expansion is
845 (2) Treat a missing third argument of ${prvscheck as if it were an empty
848 (3) Reset the variables that are obtained from the first argument of
849 ${prvscheck and used in the second argument before leaving the code,
850 because their memory is reclaimed, so using them afterwards may do
853 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
854 one (it's much easier than Tom thought :-).
856 (5) Because of (4), we can now allow for the use of $prvscheck_result
857 inside the third argument.
859 PH/03 For some reason, the default setting of PATH when running a command from
860 a pipe transport was just "/usr/bin". I have changed it to
863 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
864 anything to be listed in the output from -bV.
866 PH/05 When a filter generated an autoreply, the entire To: header line was
867 quoted in the delivery log line, like this:
869 => >A.N.Other <ano@some.domain> <original@ddress> ...
871 This has been changed so that it extracts the operative address. There
872 may be more than one such address. If so, they are comma-separated, like
875 => >ano@some.domain,ona@other.domain <original@ddress> ...
877 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
878 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
879 not being looked up in the rDNS to get a host name, Exim was showing the
880 IP address twice in Received: lines, even though the IP addresses were
881 identical. For example:
883 Received: from [1.2.3.4] (helo=[1.2.3.4])
885 However, if the real host name was known, it was omitting the HELO data
886 if it matched the actual IP address. This has been tidied up so that it
887 doesn't show the same IP address twice.
889 PH/07 When both +timestamp and +memory debugging was on, the value given by
890 $tod_xxx expansions could be wrong, because the tod_stamp() function was
891 called by the debug printing, thereby overwriting the timestamp buffer.
892 Debugging no longer uses the tod_stamp() function when +timestamp is set.
894 PH/08 When the original message was included in an autoreply transport, it
895 always said "this is a copy of the message, including all the headers",
896 even if body_only or headers_only was set. It now gives an appropriate
899 PH/09 Applied a patch from the Sieve maintainer which:
901 o fixes some comments
902 o adds the (disabled) notify extension core
903 o adds some debug output for the result of if/elsif tests
904 o points to the current vacation draft in the documentation
905 and documents the missing references header update
909 o fixes a bug in processing the envelope test (when testing
910 multiple envelope elements, the last element determinted the
913 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
914 Electronic Mail") by including:
916 Auto-submitted: auto-generated
918 in the messages that it generates (bounce messages and others, such as
919 warnings). In the case of bounce messages for non-SMTP mesages, there was
920 also a typo: it was using "Auto_submitted" (underscore instead of
921 hyphen). Since every message generated by Exim is necessarily in response
922 to another message, thes have all been changed to:
924 Auto-Submitted: auto-replied
926 in accordance with these statements in the RFC:
928 The auto-replied keyword:
930 - SHOULD be used on messages sent in direct response to another
931 message by an automatic process,
933 - MUST NOT be used on manually-generated messages,
935 - MAY be used on Delivery Status Notifications (DSNs) and Message
936 Disposition Notifications (MDNs),
938 - MUST NOT be used on messages generated by automatic or periodic
939 processes, except for messages which are automatic responses to
942 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
943 to the default Received: header definition.
945 PH/12 Added log selector acl_warn_skipped (default on).
947 PH/13 After a successful wildlsearch lookup, discard the values of numeric
948 variables because (a) they are in the wrong storage pool and (b) even if
949 they were copied, it wouldn't work properly because of the caching.
951 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
952 checking when decoding. Apparently there are clients that generate
953 overlong encoded strings. Why am I not surprised?
955 PH/15 If the first argument of "${if match_address" was not empty, but did not
956 contain an "@" character, Exim crashed. Now it writes a panic log message
957 and treats the condition as false.
959 PH/16 In autoreply, treat an empty string for "once" the same as unset.
961 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
962 extension "envelope-auth". The code is finished and in agreement with
963 other implementations, but there is no documentation so far and in fact,
964 nobody wrote the draft yet. This extension is currently #undef'ed, thus
965 not changing the active code.
967 Print executed "if" and "elsif" statements when debugging is used. This
968 helps a great deal to understand what a filter does.
970 Document more things not specified clearly in RFC3028. I had all this
971 sorted out, when out of a sudden new issues came to my mind. Oops."
973 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
976 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
977 canonical form (as documented). However, after a host name lookup from
978 the IP address, check_host() was doing a simple string comparison with
979 addresses acquired from the DNS when checking that the found name did
980 have the original IP as one of its addresses. Since any found IPv6
981 addresses are likely to be in abbreviated form, the comparison could
982 fail. Luckily, there already exists a function for doing the comparison
983 by converting both addresses to binary, so now that is used instead of
986 PH/20 There was another similar case to PH/19, when a complete host name was
987 given in a host list; looking up its IP address could give an abbreviated
988 form, whereas the current host's name might or might not be abbreviated.
989 The same fix has been applied.
995 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
996 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
999 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1000 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1002 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1004 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1005 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1006 Also, for "Auto-Submitted", treat anything other than "no" as
1007 non-personal, in accordance with RFC 3834. (Previously it treated
1008 anything starting "auto-" as non-personal.)
1010 TF/01 The control=submission/name=... option had a problem with syntax
1011 errors if the name included a slash character. The /name= option
1012 now slurps the rest of the string, so it can include any characters
1013 but it must come last in the list of options (after /sender_retain
1016 PH/05 Some modifications to the interface to the fake nameserver for the new
1024 TK/01 Added the "success_on_redirect" address verification option. See
1025 NewStuff for rationale and an example.
1027 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1029 PH/02 Patch to exigrep to allow it to work on syslog lines.
1031 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1032 fread() to skip over the body file's header line, because in Cygwin the
1033 header line is locked and is inaccessible.
1035 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1036 co-exist for some time) to make it clear that it is the Exim ID that is
1037 referenced, not the Message-ID: header line.
1039 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1040 string_format() function, because snprintf() does not exist on all
1043 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1044 ${stat: expansion item.
1046 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1047 protocol synchronization error", to keep the pedants happy.
1049 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1050 well as for IRIX systems, when gcc is being used. See the host.c source
1053 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1055 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1058 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1059 transport and to the smtp transport in LMTP mode.
1061 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1063 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1064 run a filter in a subprocess. This could lead to confusion in subsequent
1065 lookups in the parent process. There should also be a search_tidyup() at
1066 the end of the subprocess.
1068 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1069 only if the host matched helo_try_verify_hosts, which caused the
1070 verification to occur when the EHLO/HELO command was issued. The ACL just
1071 tested the remembered result. Now, if a previous verification attempt has
1072 not happened, "verify = helo" does it there and then.
1074 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1076 TK/03 Fix log output including CR from clamd.
1078 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1079 exist provoked a memory error which could cause a segfault.
1081 PH/15 Installed PCRE 6.2
1083 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1085 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1086 of the problem. Specifically, suggested +O2 rather than +O1 for the
1089 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1091 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1092 record, the port was not correctly logged when the outgoing_port log
1093 selector was set (it logged the transort's default port).
1095 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1096 fallback_hosts, and "hosts" in the smtp transport.
1098 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1099 host errors such as "Connection refused".
1101 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1102 authentication with radiusclient 0.4.9:
1104 - Error returned from rc_read_config was caught wrongly
1105 - Username/password not passed on to radius server due to wrong length.
1107 The presumption is that some radiusclient API changes for 4.51/PH/17
1108 were not taken care of correctly. The code is still untested by me (my
1109 Linux distribution still has 0.3.2 of radiusclient), but it was
1110 contributed by a Radius user.
1112 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1113 expanding the "port" option of the smtp transport.
1115 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1116 while reading a MIME header. Thanks to Tom Hughes for a patch.
1118 PH/24 Include config.h inside local_scan.h so that configuration settings are
1121 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1122 that in an ACL for RCPT (for example), you can examine exactly what was
1125 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1126 commands, but it was not correctly comparing the address with the actual
1127 client host address. Thus, it would show the EHLO address in Received:
1128 header lines when this was not necessary.
1130 PH/27 Added the % operator to ${eval:}.
1132 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1133 it should be ignoring failures (because with -C, for example, it has lost
1134 privilege). It wasn't ignoring creation failures other than "already
1137 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1140 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1141 because an input line was too long, either on its own, or by virtue of
1142 too many continuations, the temporary file was not being removed, and the
1143 return code was incorrect.
1145 PH/31 Missing "BOOL" in function definition in filtertest.c.
1147 PH/32 Applied Sieve patches from the maintainer.
1149 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1151 PH/33 Added "verify = not_blind".
1153 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1154 Local/Makefile (with some defaults set). These are used in built scripts
1155 such as exicyclog, but they have never been used in the exim_install
1156 script (though there are many overriding facilities there). I have
1157 arranged that the exim_install script now takes note of these two
1160 PH/35 Installed configuration files for Dragonfly.
1162 PH/36 When a locally submitted message by a trusted user did not contain a
1163 From: header, and the sender address was obtained from -f or from an SMTP
1164 MAIL command, and the trusted user did not use -F to supply a sender
1165 name, $originator_name was incorrectly used when constructing a From:
1166 header. Furthermore, $originator_name was used for submission mode
1167 messages from external hosts without From: headers in a similar way,
1168 which is clearly wrong.
1170 PH/37 Added control=suppress_local_fixups.
1172 PH/38 When log_selector = +received_sender was set, and the addition of the
1173 sender made the log line's construction buffer exactly full, or one byte
1174 less than full, an overflow happened when the terminating "\n" was
1177 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1178 when the result of a list match is failure because a DNS lookup failed.
1180 PH/40 RM_COMMAND is now used in the building process.
1182 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1183 the "build-* directories that it finds.
1185 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1186 address in a domain literal was a prefix of an interface address.
1188 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1189 when verifying a sender address, unless rewrite_headers is false.
1191 PH/44 Wrote a long comment about why errors_to addresses are verified as
1192 recipients, not senders.
1194 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1195 the ratelimit ACL was added.
1197 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1199 PH/46 Added extra information about PostgreSQL errors to the error string.
1201 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1202 avoiding the need to install special zones in a real server. This is
1203 backwards compatible; if it can't find the fake resolver, it drops back.
1204 Thus, both old and new test suites can be run.
1206 TF/02 Added util/ratelimit.pl
1208 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1209 clock is set back in time.
1211 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1212 Candler <B.Candler@pobox.com>.
1214 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1215 OK for addresses that are the result of redirections.
1217 PH/48 A number of further additions for the benefit of the new test suite,
1218 including a fake gethostbyname() that interfaces to the fake DNS resolver
1221 TF/06 The fix for widen_domains has also been applied to qualify_single and
1222 search_parents which are the other dnslookup options that can cause
1225 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1228 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1230 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1231 service" if it encountered an MX record with an empty target hostname.
1232 The message is now "an MX or SRV record indicated no SMTP service".
1234 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1235 if the DNS of the sending site is misconfigured. This is quite a
1236 common situation. This change restores the behaviour of treating a
1237 helo verification defer as a failure.
1239 PH/51 If self=fail was set on a router, the bounce message did not include the
1240 actual error message.
1246 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1248 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1249 command itself ended in error, the underlying message about the transport
1250 filter timeout was being overwritten with the pipe command error. Now the
1251 underlying error message should be appended to the second error message.
1253 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1255 PH/02 Reduce the amount of output that "make" produces by default. Full output
1256 can still be requested.
1258 PH/03 The warning log line about a condition test deferring for a "warn" verb
1259 was being output only once per connection, rather than after each
1260 occurrence (because it was using the same function as for successful
1261 "warn" verbs). This seems wrong, so I have changed it.
1263 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1264 it should not have, which might have caused a crash in the right
1265 circumstances, but probably never did.
1267 PH/04 Installed a modified version of Tony Finch's patch to make submission
1268 mode fix the return path as well as the Sender: header line, and to
1269 add a /name= option so that you can make the user's friendly name appear
1272 TF/03 Added the control = fakedefer ACL modifier.
1274 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1275 Mark Lowes for thorough testing.
1277 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1279 TK/03 Merged latest SRS patch from Miles Wilton.
1281 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1282 with the definition in sysexits.h (which is #included earlier).
1283 Fortunately, Exim does not actually use EX_OK. The code used to try to
1284 preserve the sysexits.h value, by assumimg that macro definitions were
1285 scanned for macro replacements. I have been disabused of this notion,
1286 so now the code just undefines EX_OK before #including unistd.h.
1288 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1289 in the smtp transport. When a block could not be written in a single
1290 write() function, the timeout was being re-applied to each part-write.
1291 This seems wrong - if the receiver was accepting one byte at a time it
1292 would take for ever. The timeout is now adjusted when this happens. It
1293 doesn't have to be particularly precise.
1295 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1296 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1298 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1299 without a domain if the check to <postmaster@domain> fails.
1301 SC/01 Eximstats: added -xls and the ability to specify output files
1302 (patch written by Frank Heydlauf).
1304 SC/02 Eximstats: use FileHandles for outputing results.
1306 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1308 SC/04 Eximstats: fixed display of large numbers with -nvr option
1310 SC/05 Eximstats: fixed merging of reports with empty tables.
1312 SC/06 Eximstats: added the -include_original_destination flag
1314 SC/07 Eximstats: removed tabs and trailing whitespace.
1316 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1318 TK/06 MBOX spool code: Add real "From " MBOX separator line
1319 so the .eml file is really in mbox format (even though
1320 most programs do not really care). Patch from Alex Miller.
1322 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1323 The latter is generated from $received_to and is only set if the
1324 message has one envelope recipient. SA can use these headers,
1325 obviously out-of-the-box. Patch from Alex Miller.
1327 PH/08 The ${def test on a variable was returning false if the variable's
1328 value was "0", contrary to what the specification has always said!
1329 The result should be true unless the variable is empty.
1331 PH/09 The syntax error of a character other than { following "${if
1332 def:variable_name" (after optional whitespace) was not being diagnosed.
1333 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1334 accidental colon was present, for example, could give incorrect results.
1336 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1337 result is used (not including appendfile, where other changes are about
1340 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1341 This involved changing a lot of size variables from int to off_t. It
1342 should work with maildirs and everything.
1344 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1345 spamd dying while we are connected to it.
1347 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1350 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1351 function for BDB 4.3.
1353 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1355 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1356 the sources for PCRE have been re-arranged and more files are now
1359 PH/15 The code I had for printing potentially long long variables in PH/11
1360 above was not the best (it lost precision). The length of off_t variables
1361 is now inspected at build time, and an appropriate printing format (%ld
1362 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1363 to be "long long int" or "long int". This is needed for the internal
1364 formatting function string_vformat().
1366 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1367 the configuration file to be ":syslog", then the script "guesses" where
1368 the logs files are, rather than using the compiled in default. In our
1369 case the guess is not the same as the compiled default, so the script
1370 suddenly stopped working when I started to use syslog. The patch checks
1371 to see if log_file_path is "". If so, it attempts to read it from exim
1372 with no configuration file to get the compiled in version, before it
1373 falls back to the previous guessing code."
1375 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1376 implementing BATV in an Exim configuration. See NewStuff for the gory
1379 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1380 Makefile that are specific to HP-UX.
1382 PH/18 If the "use_postmaster" option was set for a recipient callout together
1383 with the "random" option, the postmaster address was used as the MAIL
1384 FROM address for the random test, but not for the subsequent recipient
1385 test. It is now used for both.
1387 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1388 patch removes a few documentation additions to RFC 3028, because the
1389 latest draft now contains them. It adds the new en;ascii-case comparator
1390 and a new error check for 8bit text in MIME parts. Comparator and
1391 require names are now matched exactly. I enabled the subaddress
1392 extension, but it is not well tested yet (read: it works for me)."
1394 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1395 rework some of the code of TK/09 above to avoid the hardwired use of
1396 "%lld" and "long long". Replaced the call to snprintf() with a call to
1399 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1400 records point to non-existent hosts", "retry timeout exceeded", and
1401 "retry time not reached for any host after a long failure period".
1403 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1404 experimental DomainKeys support:
1406 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1407 (2) On an error such as an illegally used "control", the wrong name for
1408 the control was given.
1410 These problems did NOT occur unless DomainKeys support was compiled.
1412 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1414 PH/24 Added ${if match_ip condition.
1416 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1417 databases so that it will be absolutely obvious if a crash occurs in the
1418 DB library. This is a regular occurrence (often caused by mis-matched
1421 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1422 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1423 picked up on a user's system that detects such things. There doesn't seem
1424 to be a gcc warning option for this - only an attribute that has to be
1425 put on the function's prototype. It seems that in Fedora Core 4 they have
1426 set this on a number of new functions. No doubt there will be more in due
1429 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1430 specify a transport. However, if an address that was verified by such a
1431 router was the subject of a callout, Exim crashed because it tried to
1432 read the rcpt_include_affixes from the non-existent transport. Now it
1433 just assumes that the setting of that option is false. This bug was
1434 introduced by 4.51/PH/31.
1436 PH/28 Changed -d+all to exclude +memory, because that information is very
1437 rarely of interest, but it makes the output a lot bigger. People tend to
1438 do -d+all out of habit.
1440 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1441 code in os-type was giving problems when libc.so lives in lib64, like on
1444 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1445 aren't the modern standard, and it seems that some systems' include files
1446 don't always have them. Exim was already checking for some of the newer
1447 ones like T_AAAA, and defining it itself. I've added checks for all the
1448 record types that Exim uses.
1450 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1451 not automatically generating a new one, as it is supposed to. This
1452 prevented TLS from working. If the file did exist, but contained invalid
1453 data, a new version was generated, as expected. It was only the case of a
1454 non-existent file that was broken.
1456 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1457 with a change in libdomainkeys > 0.64.
1459 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1460 from DNS. If the selector record carries the flag, it now has
1461 precedence over the domain-wide flag.
1463 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1465 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1466 the use of a port name that isn't defined in /etc/services) occurred, the
1467 message was deferred as in a normal delivery, and thus remained on the
1468 spool, instead of being failed because of the mua_wrapper setting. This
1469 is now fixed, and I tidied up some of the mua_wrapper messages at the
1472 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1473 the messages in a hash of arrays rather than using individual hashes.
1474 This is a bit cleaner and results in dramatic memory savings, albeit
1475 at a slight CPU cost.
1477 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1478 as requested by Marc Sherman.
1480 SC/10 Eximstats: added histograms for user specified patterns as requested
1483 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1485 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1486 fopen() in the content-scanning modules that did not already have it.
1492 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1493 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1495 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1497 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1499 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1500 to test Sieve filters that use "vacation".
1502 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1503 that changes the way the GnuTLS parameters are stored in the cache file.
1504 The new format can be generated externally. For backward compatibility,
1505 if the data in the cache doesn't make sense, Exim assumes it has read an
1506 old-format file, and it generates new data and writes a new file. This
1507 means that you can't go back to an older release without removing the
1510 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1511 work if there are any delivery delays because "one_time" forces the
1512 parent to be marked "delivered", so its unseen clone is never tried
1513 again. For this reason, Exim now forbids the simultaneous setting of
1516 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1517 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1518 entry if you want to know the details. The fix, however, neglected to
1519 consider the case when local delivery batching is involved. The test for
1520 "previously delivered" was not happening when checking to see if an
1521 address could be batched with a previous (undelivered) one; under
1522 certain circumstances this could lead to multiple deliveries to the same
1525 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1526 in its include files, and this causes problems building Exim.
1528 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1529 header_syntax) but Exim was just ignoring anything given after a slash.
1530 In particular, this caused confusion with an attempt to use "verify =
1531 reverse_host_lookup/defer_ok". An error is now given when options are
1532 supplied for verify items that do not have them. (Maybe reverse_host_
1533 lookup should have a defer_ok option, but that's a different point.)
1535 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1536 defined by RFC 821) to 2048, because there were problems with some AUTH
1537 commands, and RFC 1869 says the size should be increased for extended
1538 SMTP commands that take arguments.
1540 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1543 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1544 "unknown" error; now it says that the functionality isn't in the binary.
1546 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1547 an address' error message when a string expansion fails (syntax or
1548 whatever). Otherwise the password may appear in the log. Following change
1549 PH/42 below, there is no longer a chance of it appearing in a bounce
1552 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1554 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1555 of its addresses was ever tried. (Bugzilla bug #2.)
1557 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1558 the result incorrectly in the debug output. (It correctly added a newline
1559 to what was transported.)
1561 TF/01 Added $received_time.
1563 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1564 commented out examples of how to interface to a virus scanner and to
1565 SpamAssassin. Also added commented examples of av_scanner and
1566 spamd_address settings.
1568 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1569 and controls are allowed in which ACLs. There were a couple of minor
1570 errors. Some of the entries in the conditions table (which is a table of
1571 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1572 negation of where the condition IS allowed.
1574 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1576 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1577 header file does not have a version number, so I've had to invent a new
1578 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1579 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1580 radiusclient), but it was contributed by a Radius user.
1582 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1583 files or directories, for OpenSSL.
1585 PH/19 When an Exim process that is running as root has to create an Exim log
1586 file, it does so in a subprocess that runs as exim:exim so as to get the
1587 ownership right at creation (otherwise, other Exim processes might see
1588 the file with the wrong ownership). There was no test for failure of this
1589 fork() call, which would lead to the process getting stuck as it waited
1590 for a non-existent subprocess. Forks do occasionally fail when resources
1591 run out. I reviewed all the other calls to fork(); they all seem to check
1594 PH/20 When checking for unexpected SMTP input at connect time (before writing
1595 the banner), Exim was not dealing correctly with a non-positive return
1596 from the read() function. If the client had disconnected by this time,
1597 the result was a log entry for a synchronization error with an empty
1598 string after "input=" when read() returned zero. If read() returned -1
1599 (an event I could not check), uninitialized data bytes were printed.
1600 There were reports of junk text (parts of files, etc) appearing after
1603 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1605 PH/22 Added support for macro redefinition, and (re)definition in between
1606 driver and ACL definitions.
1608 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1609 forgetting to use the resulting value; it was using the unexpanded value.
1611 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1612 hadn't been configured. The fix is from Juergen Kreileder, who
1613 understands it better than I do:
1615 "Here's what I see happening with three configured cyrus_sasl
1616 authenticators configured (plain, login, cram-md5):
1618 On startup auth_cyrus_sasl_init() gets called for each of these.
1619 This means three calls to sasl_listmech() without a specified mech_list.
1620 => SASL tests which mechs of all available mechs actually work
1621 => three warnings about OTP not working
1622 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1624 With the patch, sasl_listmech() also gets called three times. But now
1625 SASL's mech_list option is set to the server_mech specified in the the
1626 authenticator. Or in other words, the answer from sasl_listmech()
1627 gets limited to just the mech you're testing for (which is different
1629 => the return list contains just 'plain' or 'login', 'cram-md5' or
1630 nothing depending on the value of ob->server_mech.
1632 I've just tested the patch: Authentication still works fine,
1633 unavailable mechs specified in the exim configuration are still
1634 caught, and the auth.log warnings about OTP are gone."
1636 PH/25 When debugging is enabled, the contents of the command line are added
1637 to the debugging output, even when log_selector=+arguments is not
1640 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1641 answer is "GNU", and only if the return is "GNU/something" is the answer
1644 PH/27 $acl_verify_message is now set immediately after the failure of a
1645 verification in an ACL, and so is available in subsequent modifiers. In
1646 particular, the message can be preserved by coding like this:
1648 warn !verify = sender
1649 set acl_m0 = $acl_verify_message
1651 Previously, $acl_verify_message was set only while expanding "message"
1652 and "log_message" when a very denied access.
1654 PH/28 Modified OS/os.c-Linux with
1656 -#ifndef OS_LOAD_AVERAGE
1657 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1659 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1660 nomenclature these days.)
1662 PH/29 Installed patch from the Sieve maintainer that adds the options
1663 sieve_useraddress and sieve_subaddress to the redirect router.
1665 PH/30 In these circumstances:
1666 . Two addresses routed to the same list of hosts;
1667 . First host does not offer TLS;
1668 . First host accepts first address;
1669 . First host gives temporary error to second address;
1670 . Second host offers TLS and a TLS session is established;
1671 . Second host accepts second address.
1672 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1673 and peerdn, if requested) that were in fact used only for the second
1676 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1677 attention to any local part prefix or suffix that was matched by the
1678 router that accepted the address. It now behaves in the same way as it
1679 does for delivery: the affixes are removed from the local part unless
1680 rcpt_include_affixes is set on the transport.
1682 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1683 timeout during the DATA phase of an incoming message.
1685 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1686 applied a patch sanctioned by the Sieve maintainer.
1688 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1689 the uid or gid is negative. A case of a negative gid caused this to be
1690 noticed. The fix allows for either to be negative.
1692 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1693 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1696 PH/36 The change PH/12 above was broken. Fixed it.
1698 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1699 the grounds that routing the same address twice would always produce the
1700 same answer. This might have been true once, but it is certainly no
1701 longer true now. Routing a child address may depend on the previous
1702 routing that produced that child. Some complicated redirection strategies
1703 went wrong when messages had multiple recipients, and made Exim's
1704 behaviour dependent on the order in which the addresses were given.
1706 I have moved the duplicate checking until after the routing is complete.
1707 Exim scans the addresses that are assigned to local and remote
1708 transports, and removes any duplicates. This means that more work will be
1709 done, as duplicates will always all be routed, but duplicates are
1710 presumably rare, so I don't expect this is of any significance.
1712 For deliveries to pipes, files, and autoreplies, the duplicate checking
1713 still happens during the routing process, since they are not going to be
1716 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1717 It corrects a timeout issue with spamd. This is Ian's comment: "The
1718 background is that sometimes spamd either never reads data from a
1719 connection it has accepted, or it never writes response data. The exiscan
1720 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1721 blindly assumes that writes won't block so it may never time out."
1723 PH/39 Allow G after quota size as well as K and M.
1725 PH/40 The value set for $authenticated_id in an authenticator may not contain
1726 binary zeroes or newlines because the value is written to log lines and
1727 to spool files. There was no check on this. Now the value is run through
1728 the string_printing() function so that such characters are converted to
1729 printable escape sequences.
1731 PH/41 $message_linecount is a new variable that contains the total number of
1732 lines in the message. Compare $body_linecount, which is the count for the
1735 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1736 bounce and delay warning messages, except in certain special cases, which
1739 (a) An SMTP error message from a remote host;
1740 (b) A message specified in a :fail: redirection;
1741 (c) A message specified in a "fail" command in a system filter;
1742 (d) A message specified in a FAIL return from the queryprogram router;
1743 (e) A message specified by the cannot_route_message router option.
1745 In these cases only, Exim does include the error details in bounce and
1746 warning messages. There are also a few cases where bland messages such
1747 as "unrouteable address" or "local delivery error" are given.
1749 PH/43 $value is now also set for the "else" part of a ${run expansion.
1751 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1752 being worked on, but at least Exim now implements the latest version to
1755 PH/45 In a pipe transport, although a timeout while waiting for the pipe
1756 process to complete was treated as a delivery failure, a timeout while
1757 writing the message to the pipe was logged, but erroneously treated as a
1758 successful delivery. Such timeouts include transport filter timeouts. For
1759 consistency with the overall process timeout, these timeouts are now
1760 treated as errors, giving rise to delivery failures by default. However,
1761 there is now a new Boolean option for the pipe transport called
1762 timeout_defer, which, if set TRUE, converts the failures into defers for
1763 both kinds of timeout. A transport filter timeout is now identified in
1766 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1767 systems where "make" and "gmake" are different, calling "gmake" at top
1768 level broke things. I've arranged for the value of $(MAKE) to be passed
1769 from the Makefile to this script so that it can call the same version of
1773 A note about Exim versions 4.44 and 4.50
1774 ----------------------------------------
1776 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1777 changes of various kinds. As a consequence, a big documentation update was
1778 needed. This delayed the release for rather longer than seemed good, especially
1779 in the light of a couple of (minor) security issues. Therefore, the changes
1780 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1781 release. So 4.44 and 4.50 are in effect two different branches that both start
1784 I have left the 4.50 change log unchanged; it contains all the changes since
1785 4.43. The change log for 4.44 is below; many of its items are identical to
1786 those for 4.50. This seems to be the most sensible way to preserve the
1787 historical information.
1793 1. Minor wording change to the doc/README.SIEVE file.
1795 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1796 computation of the current number of files was incorrect.
1798 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1799 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1802 4. Give more explanation in the error message when the command for a transport
1803 filter fails to execute.
1805 5. There are several places where Exim runs a non-Exim command in a
1806 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1807 was being done only for the command run by the queryprogram router. It is
1808 now done for all such subprocesses. The other cases are: ${run, transport
1809 filters, and the commands run by the lmtp and pipe transports.
1811 6. Added CONFIGURE_GROUP build-time option.
1813 7. Some older OS have a limit of 256 on the maximum number of file
1814 descriptors. Exim was using setrlimit() to set 1000 as a large value
1815 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1816 systems. I've change it so that if it can't get 1000, it tries for 256.
1818 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1819 was an oversight, and furthermore, ever since the addition of extra
1820 controls (e.g. 4.43/32), the checks on when to allow different forms of
1821 "control" were broken. There should now be diagnostics for all cases when a
1822 control that does not make sense is encountered.
1824 9. Added the /retain_sender option to "control=submission".
1826 10. $recipients is now available in the predata ACL (oversight).
1828 11. Tidy the search cache before the fork to do a delivery from a message
1829 received from the command line. Otherwise the child will trigger a lookup
1830 failure and thereby defer the delivery if it tries to use (for example) a
1831 cached ldap connection that the parent has called unbind on.
1833 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1834 of $address_data from the recipient verification was clobbered by the
1835 sender verification.
1837 13. The value of address_data from a sender verification is now available in
1838 $sender_address_data in subsequent conditions in the ACL statement.
1840 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1842 15. Added a new option "connect=<time>" to callout options, to set a different
1845 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1846 was its contents. (It was OK if the option was not defined at all.)
1848 17. A "Completed" log line is now written for messages that are removed from
1849 the spool by the -Mrm option.
1851 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1852 information about exactly what failed.
1854 19. Added -dd to debug only the daemon process.
1856 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1857 handles timeouts, both on the server side and network timeouts. Renamed the
1858 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1860 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1861 It is now set to "smtps".
1863 22. $host_address is now set to the target address during the checking of
1864 ignore_target_hosts.
1866 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1867 being passed; this would have caused $sender_host_name to have been used if
1868 matching the list had actually called for a host name (not very likely,
1869 since this list is usually IP addresses). A host name is now passed as
1872 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1873 code that specifies a non-restarting handler (typically sigaction() in
1874 modern systems) in an attempt to fix a rare and obscure crash bug.
1876 25. Narrowed the window for a race in the daemon that could cause it to ignore
1877 SIGCHLD signals. This is not a major problem, because they are used only to
1878 wake it up if nothing else does.
1880 26. A malformed maildirsize file could cause Exim to calculate negative values
1881 for the mailbox size or file count. Odd effects could occur as a result.
1882 The maildirsize information is now recalculated if the size or filecount
1885 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1886 support for a long time. Removed HAVE_SYS_VFS_H.
1888 28. Installed the latest version of exipick from John Jetmore.
1890 29. In an address list, if the pattern was not a regular expression, an empty
1891 subject address (from a bounce message) matched only if the pattern was an
1892 empty string. Non-empty patterns were not even tested. This was the wrong
1893 because it is perfectly reasonable to use an empty address as part of a
1894 database query. An empty address is now tested by patterns that are
1895 lookups. However, all the other forms of pattern expect the subject to
1896 contain a local part and a domain, and therefore, for them, an empty
1897 address still always fails if the pattern is not itself empty.
1899 30. Exim went into a mad DNS loop when attempting to do a callout where the
1900 host was specified on an smtp transport, and looking it up yielded more
1901 than one IP address.
1903 31. Re-factored the code for checking spool and log partition space into a
1904 function that finds that data and another that does the check. The former
1905 is then used to implement four new variables: $spool_space, $log_space,
1906 $spool_inodes, and $log_inodes.
1908 32. The RFC2047 encoding function was originally intended for short strings
1909 such as real names; it was not keeping to the 75-character limit for
1910 encoded words that the RFC imposes. It now respects the limit, and
1911 generates multiple encoded words if necessary. To be on the safe side, I
1912 have increased the buffer size for the ${rfc2047: expansion operator from
1915 33. It is now permitted to omit both strings after an "if" condition; if the
1916 condition is true, the result is "true". As before, when the second string
1917 is omitted, a false condition yields an empty string. This makes it less
1918 cumbersome to write custom ACL and router conditions.
1920 34. Failure to deliver a bounce message always caused it to be frozen, even if
1921 there was an errors_to setting on the router. The errors_to setting is now
1924 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1925 canonical form (fully expanded) before being placed in
1926 $sender_host_address.
1928 36. The table in the code that translates DNS record types into text (T_A to
1929 "A" for instance) was missing entries for NS and CNAME. It is just possible
1930 that this could have caused confusion if both these types were looked up
1931 for the same domain, because the text type is used as part of Exim's
1932 per-process caching. But the chance of anyone hitting this buglet seems
1935 37. The dnsdb lookup has been extended in a number of ways.
1937 (1) There is a new type, "zns", which walks up the domain tree until it
1938 finds some nameserver records. It should be used with care.
1940 (2) There is a new type, "mxh", which is like "mx" except that it returns
1941 just the host names, not the priorities.
1943 (3) It is now possible to give a list of domains (or IP addresses) to be
1944 looked up. The behaviour when one of the lookups defers can be
1945 controlled by a keyword.
1947 (4) It is now possible to specify the separator character for use when
1948 multiple records are returned.
1950 38. The dnslists ACL condition has been extended: it is now possible to supply
1951 a list of IP addresses and/or domains to be looked up in a particular DNS
1954 39. Added log_selector=+queue_time_overall.
1956 40. When running the queue in the test harness, wait just a tad after forking a
1957 delivery process, to get repeatability of debugging output.
1959 41. Include certificate and key file names in error message when GnuTLS fails
1960 to set them up, because the GnuTLS error message doesn't include the name
1961 of the failing file when there is a problem reading it.
1963 42. Allow both -bf and -bF in the same test run.
1965 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1967 44. The "Exiscan patch" is now merged into the mainline Exim source.
1969 45. Sometimes the final signoff response after QUIT could fail to get
1970 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1971 < 0 before doing a fflush(). This bug looks as though it goes back to the
1972 introduction of TLS in release 3.20, but "sometimes" must have been rare
1973 because the tests only now provoked it.
1975 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1976 (this can affect the format of dates).
1978 47. exim_tidydb, when checking for the continued existence of a message for
1979 which it has found a message-specific retry record, was not finding
1980 messages that were in split spool directories. Consequently, it was
1981 deleting retry records that should have stayed in existence.
1983 48. Steve fixed some bugs in eximstats.
1985 49. The SPA authentication driver was not abandoning authentication and moving
1986 on to the next authenticator when an expansion was forced to fail,
1987 contradicting the general specification for all authenticators. Instead it
1988 was generating a temporary error. It now behaves as specified.
1990 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1991 (the order specifies the preference for clients). The order is now AES256,
1992 AES128, 3DES, ARCFOUR128.
1994 51. Small patch to Sieve code - explicitly set From: when generating an
1997 52. Exim crashed if a remote delivery caused a very long error message to be
1998 recorded - for instance if somebody sent an entire SpamAssassin report back
1999 as a large number of 550 error lines. This bug was coincidentally fixed by
2000 increasing the size of one of Exim's internal buffers (big_buffer) that
2001 happened as part of the Exiscan merge. However, to be on the safe side, I
2002 have made the code more robust (and fixed the comments that describe what
2005 53. Now that there can be additional text after "Completed" in log lines (if
2006 the queue_time_overall log selector is set), a one-byte patch to exigrep
2007 was needed to allow it to recognize "Completed" as not the last thing in
2010 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2011 patch that reportedly fixes this has been added. I am not expert enough to
2012 create a test for it. This is what the patch creator wrote:
2014 "I found a little strange behaviour of ldap code when working with
2015 Windows 2003 AD Domain, where users was placed in more than one
2016 Organization Units. When I tried to give exim partial DN, the exit code
2017 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2018 But simultaneously result of request was absolutely normal ldap result,
2019 so I produce this patch..."
2021 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2022 REFERENCE, so I have modified the code to exclude the patch when that macro
2025 55. Some experimental protocols are using DNS PTR records for new purposes. The
2026 keys for these records are domain names, not reversed IP addresses. The
2027 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2028 leaves it alone. Component reversal etc. now happens only for IP addresses.
2031 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2033 57. Double the size of the debug message buffer (to 2048) so that more of very
2034 long debug lines gets shown.
2036 58. The exicyclog utility now does better if the number of log files to keep
2037 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2039 59. Two changes related to the smtp_active_hostname option:
2041 (1) $smtp_active_hostname is now available as a variable.
2042 (2) The default for smtp_banner uses $smtp_active_hostname instead
2043 of $primary_hostname.
2045 60. The host_aton() function is supposed to be passed a string that is known
2046 to be a valid IP address. However, in the case of IPv6 addresses, it was
2047 not checking this. This is a hostage to fortune. Exim now panics and dies
2048 if the condition is not met. A case was found where this could be provoked
2049 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2050 components; fortuitously, this particular loophole had already been fixed
2051 by change 4.50/55 above.
2053 If there are any other similar loopholes, the new check in host_aton()
2054 itself should stop them being exploited. The report I received stated that
2055 data on the command line could provoke the exploit when Exim was running as
2056 exim, but did not say which command line option was involved. All I could
2057 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2058 running as the user.
2061 61. There was a buffer overflow vulnerability in the SPA authentication code
2062 (which came originally from the Samba project). I have added a test to the
2063 spa_base64_to_bits() function which I hope fixes it.
2066 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2067 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2069 63. The daemon start-up calls getloadavg() while still root for those OS that
2070 need the first call to be done as root, but it missed one case: when
2071 deliver_queue_load_max is set with deliver_drop_privilege. This is
2072 necessary for the benefit of the queue runner, because there is no re-exec
2073 when deliver_drop_privilege is set.
2075 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2076 This has been fixed.
2078 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2079 was in use, was not putting the data itself into the right store pool;
2080 consequently, it could be overwritten for a subsequent message in the same
2081 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2084 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2086 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2087 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2088 the function were treating the return as a boolean value, which happened to
2089 work because 0=false and not-0=true, but is not correct code.
2091 68. The host_aton() function was not handling scoped IPv6 addresses (those
2092 with, for example, "%eth0" on the end) correctly.
2094 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2095 negated items (that is, ~something) in unsigned ints. Some compilers
2096 apparently mutter when there is no cast.
2098 70. If an address verification called from an ACL failed, and did not produce a
2099 user-specific message (i.e. there was only a "system" message), nothing was
2100 put in $acl_verify_message. In this situation, it now puts the system
2103 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2104 session; change 4.31/43 added the unwanted input to the log line - except
2105 that it did not do this in the start of session case. It now does.
2107 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2108 This is wrong and can cause the other end to generate a synchronization
2109 error if it is another Exim or anything else that does the synchronization
2110 check. A QUIT command is no longer sent after a timeout.
2112 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2113 during host lookups.
2115 74. The defer_ok option of callout verification was not working if it was used
2116 when verifying addresses in header lines, that is, for this case:
2118 verify = header_sender/callout=defer_ok
2120 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2121 those file descriptors could be used for SMTP connections. If anything
2122 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2123 could be sent to the SMTP client, causing chaos. The daemon now opens
2124 stdin, stdout, and stderr to /dev/null when it puts itself into the
2127 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2128 log by default. The user can override this with suitable Perl magic.
2130 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2131 the log message when discard triggers, was not working for the DATA ACL or
2132 for the non-SMTP ACL.
2134 78. Error message wording change in sieve.c.
2136 79. If smtp_accept_max_per_host was set, the number of connections could be
2137 restricted to fewer than expected, because the daemon was trying to set up
2138 a new connection before checking whether the processes handling previous
2139 connections had finished. The check for completed processes is now done
2140 earlier. On busy systems, this bug wouldn't be noticed because something
2141 else would have woken the daemon, and it would have reaped the completed
2144 80. If a message was submitted locally by a user whose login name contained one
2145 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2146 It caused a spool format error. I have fixed the spool reading code. A
2147 related problem was that the "from" clause in the Received: line became
2148 illegal because of the space(s). It is now covered by ${quote_local_part.
2150 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2153 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2155 83. Patch from Sieve maintainer to fix unterminated string problem in
2156 "vacation" handling.
2158 84. Some minor changes to the Linux configuration files to help with other
2159 OS variants using glibc.
2161 85. One more patch for Sieve to update vacation handling to latest spec.
2164 ----------------------------------------------------
2165 See the note above about the 4.44 and 4.50 releases.
2166 ----------------------------------------------------
2172 1. Change 4.43/35 introduced a bug that caused file counts to be
2173 incorrectly computed when quota_filecount was set in an appendfile
2176 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2177 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2180 3. Give more explanation in the error message when the command for a transport
2181 filter fails to execute.
2183 4. There are several places where Exim runs a non-Exim command in a
2184 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2185 was being done only for the command run by the queryprogram router. It is
2186 now done for all such subprocesses. The other cases are: ${run, transport
2187 filters, and the commands run by the lmtp and pipe transports.
2189 5. Some older OS have a limit of 256 on the maximum number of file
2190 descriptors. Exim was using setrlimit() to set 1000 as a large value
2191 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2192 systems. I've change it so that if it can't get 1000, it tries for 256.
2194 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2195 was an oversight, and furthermore, ever since the addition of extra
2196 controls (e.g. 4.43/32), the checks on when to allow different forms of
2197 "control" were broken. There should now be diagnostics for all cases when a
2198 control that does not make sense is encountered.
2200 7. $recipients is now available in the predata ACL (oversight).
2202 8. Tidy the search cache before the fork to do a delivery from a message
2203 received from the command line. Otherwise the child will trigger a lookup
2204 failure and thereby defer the delivery if it tries to use (for example) a
2205 cached ldap connection that the parent has called unbind on.
2207 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2208 of $address_data from the recipient verification was clobbered by the
2209 sender verification.
2211 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2212 was its contents. (It was OK if the option was not defined at all.)
2214 11. A "Completed" log line is now written for messages that are removed from
2215 the spool by the -Mrm option.
2217 12. $host_address is now set to the target address during the checking of
2218 ignore_target_hosts.
2220 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2221 being passed; this would have caused $sender_host_name to have been used if
2222 matching the list had actually called for a host name (not very likely,
2223 since this list is usually IP addresses). A host name is now passed as
2226 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2227 code that specifies a non-restarting handler (typically sigaction() in
2228 modern systems) in an attempt to fix a rare and obscure crash bug.
2230 15. Narrowed the window for a race in the daemon that could cause it to ignore
2231 SIGCHLD signals. This is not a major problem, because they are used only to
2232 wake it up if nothing else does.
2234 16. A malformed maildirsize file could cause Exim to calculate negative values
2235 for the mailbox size or file count. Odd effects could occur as a result.
2236 The maildirsize information is now recalculated if the size or filecount
2239 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2240 support for a long time. Removed HAVE_SYS_VFS_H.
2242 18. Updated exipick to current release from John Jetmore.
2244 19. Allow an empty sender to be matched against a lookup in an address list.
2245 Previously the only cases considered were a regular expression, or an
2248 20. Exim went into a mad DNS lookup loop when doing a callout where the
2249 host was specified on the transport, if the DNS lookup yielded more than
2252 21. The RFC2047 encoding function was originally intended for short strings
2253 such as real names; it was not keeping to the 75-character limit for
2254 encoded words that the RFC imposes. It now respects the limit, and
2255 generates multiple encoded words if necessary. To be on the safe side, I
2256 have increased the buffer size for the ${rfc2047: expansion operator from
2259 22. Failure to deliver a bounce message always caused it to be frozen, even if
2260 there was an errors_to setting on the router. The errors_to setting is now
2263 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2264 canonical form (fully expanded) before being placed in
2265 $sender_host_address.
2267 24. Updated eximstats to version 1.33
2269 25. Include certificate and key file names in error message when GnuTLS fails
2270 to set them up, because the GnuTLS error message doesn't include the name
2271 of the failing file when there is a problem reading it.
2273 26. Expand error message when OpenSSL has problems setting up cert/key files.
2276 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2277 (this can affect the format of dates).
2279 28. exim_tidydb, when checking for the continued existence of a message for
2280 which it has found a message-specific retry record, was not finding
2281 messages that were in split spool directories. Consequently, it was
2282 deleting retry records that should have stayed in existence.
2284 29. eximstats updated to version 1.35
2285 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2286 1.35 - bugfix such that pie charts by volume are generated correctly
2288 30. The SPA authentication driver was not abandoning authentication and moving
2289 on to the next authenticator when an expansion was forced to fail,
2290 contradicting the general specification for all authenticators. Instead it
2291 was generating a temporary error. It now behaves as specified.
2293 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2294 (the order specifies the preference for clients). The order is now AES256,
2295 AES128, 3DES, ARCFOUR128.
2297 31. Small patch to Sieve code - explicitly set From: when generating an
2300 32. Exim crashed if a remote delivery caused a very long error message to be
2301 recorded - for instance if somebody sent an entire SpamAssassin report back
2302 as a large number of 550 error lines. This bug was coincidentally fixed by
2303 increasing the size of one of Exim's internal buffers (big_buffer) that
2304 happened as part of the Exiscan merge. However, to be on the safe side, I
2305 have made the code more robust (and fixed the comments that describe what
2308 33. Some experimental protocols are using DNS PTR records for new purposes. The
2309 keys for these records are domain names, not reversed IP addresses. The
2310 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2311 leaves it alone. Component reversal etc. now happens only for IP addresses.
2314 34. The host_aton() function is supposed to be passed a string that is known
2315 to be a valid IP address. However, in the case of IPv6 addresses, it was
2316 not checking this. This is a hostage to fortune. Exim now panics and dies
2317 if the condition is not met. A case was found where this could be provoked
2318 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2319 components; fortuitously, this particular loophole had already been fixed
2320 by change 4.50/55 or 4.44/33 above.
2322 If there are any other similar loopholes, the new check in host_aton()
2323 itself should stop them being exploited. The report I received stated that
2324 data on the command line could provoke the exploit when Exim was running as
2325 exim, but did not say which command line option was involved. All I could
2326 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2327 running as the user.
2330 35. There was a buffer overflow vulnerability in the SPA authentication code
2331 (which came originally from the Samba project). I have added a test to the
2332 spa_base64_to_bits() function which I hope fixes it.
2335 36. The daemon start-up calls getloadavg() while still root for those OS that
2336 need the first call to be done as root, but it missed one case: when
2337 deliver_queue_load_max is set with deliver_drop_privilege. This is
2338 necessary for the benefit of the queue runner, because there is no re-exec
2339 when deliver_drop_privilege is set.
2341 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2342 was in use, was not putting the data itself into the right store pool;
2343 consequently, it could be overwritten for a subsequent message in the same
2344 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2347 38. Sometimes the final signoff response after QUIT could fail to get
2348 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2349 < 0 before doing a fflush(). This bug looks as though it goes back to the
2350 introduction of TLS in release 3.20, but "sometimes" must have been rare
2351 because the tests only now provoked it.
2357 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2358 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2359 Now it returns an integer. A number of calls were still expecting a T/F
2360 return. Fortuitously, in all cases, the tests worked in OK situations,
2361 which is the norm. However, things would have gone wrong on any write
2362 failures on the smtp file descriptor. This function is used when sending
2363 messages over SMTP and also when doing verify callouts.
2365 2. When Exim is called to do synchronous delivery of a locally submitted
2366 message (the -odf or -odi options), it no longer closes stderr before doing
2369 3. Implemented the mua_wrapper option.
2371 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2373 5. Implemented the functions header_remove(), header_testname(),
2374 header_add_at_position(), and receive_remove_recipient(), and exported them
2377 6. If an ACL "warn" statement specified the addition of headers, Exim already
2378 inserted X-ACL-Warn: at the start if there was no header name. However, it
2379 was not making this test for the second and subsequent header lines if
2380 there were newlines in the string. This meant that an invalid header could
2381 be inserted if Exim was badly configured.
2383 7. Allow an ACL "warn" statement to add header lines at the start or after all
2384 the Received: headers, as well as at the end.
2386 8. Added the rcpt_4xx retry error code.
2388 9. Added postmaster_mailfrom=xxx to callout verification option.
2390 10. Added mailfrom=xxxx to the callout verification option, for verify=
2393 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2394 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2396 12. Inserted some casts to stop certain compilers warning when using pointer
2397 differences as field lengths or precisions in printf-type calls (mostly
2398 affecting debugging statements).
2400 13. Added optional readline() support for -be (dynamically loaded).
2402 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2403 same clock tick as a message's arrival, so that its received time was the
2404 same as the "first fail" time on the retry record, and that message
2405 remained on the queue past the ultimate address timeout, every queue runner
2406 would try a delivery (because it was past the ultimate address timeout) but
2407 after another failure, the ultimate address timeout, which should have then
2408 bounced the address, did not kick in. This was a "< instead of <=" error;
2409 in most cases the first failure would have been in the next clock tick
2410 after the received time, and all would be well.
2412 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2413 being recognized when the domain list was tested by the match_domain
2414 condition in an expansion string.
2416 16. Added the ${str2b64: operator.
2418 17. Exim was always calling setrlimit() to set a large limit for the number of
2419 processes, without checking whether the existing limit was already
2420 adequate. (It did check for the limit on file descriptors.) Furthermore,
2421 errors from getrlimit() and setrlimit() were being ignored. Now they are
2422 logged to the main and panic logs, but Exim does carry on, to try to do its
2423 job under whatever limits there are.
2425 18. Imported PCRE 5.0.
2427 19. Trivial typo in log message " temporarily refused connection" (the leading
2430 20. If the log selector return_path_on_delivery was set and an address was
2431 redirected to /dev/null, the delivery process crashed because it assumed
2432 that a return path would always be set for a "successful" delivery. In this
2433 case, the whole delivery is bypassed as an optimization, and therefore no
2436 21. Internal re-arrangement: the function for sending a challenge and reading
2437 a response while authentication was assuming a zero-terminated challenge
2438 string. It's now changed to take a pointer and a length, to allow for
2439 binary data in such strings.
2441 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2443 23. Exim was not respecting finduser_retries when seeking the login of the
2444 uid under which it was called; it was always trying 10 times. (The default
2445 setting of finduser_retries is zero.) Also, it was sleeping after the final
2446 failure, which is pointless.
2448 24. Implemented tls_on_connect_ports.
2450 25. Implemented acl_smtp_predata.
2452 26. If the domain in control=submission is set empty, Exim assumes that the
2453 authenticated id is a complete email address when it generates From: or
2454 Sender: header lines.
2456 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2457 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2458 chown and chgrp in /bin and hostname in /usr/bin.
2460 28. Exim was keeping the "process log" file open after each use, just as it
2461 does for the main log. This opens the possibility of it remaining open for
2462 long periods when the USR1 signal hits a daemon. Occasional processlog
2463 errors were reported, that could have been caused by this. Anyway, it seems
2464 much more sensible not to leave this file open at all, so that is what now
2467 29. The long-running daemon process does not normally write to the log once it
2468 has entered its main loop, and it closes the log before doing so. This is
2469 so that log files can straightforwardly be renamed and moved. However,
2470 there are a couple of unusual error situations where the daemon does write
2471 log entries, and I had neglected to close the log afterwards.
2473 30. The text of an SMTP error response that was received during a remote
2474 delivery was being truncated at 512 bytes. This is too short for some of
2475 the long messages that one sometimes sees. I've increased the limit to
2478 31. It is now possible to make retry rules that apply only when a message has a
2479 specific sender, in particular, an empty sender.
2481 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2482 it possible to be selective about when SMTP synchronization is enforced.
2484 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2486 32. Implemented hosts_connection_nolog.
2488 33. Added an ACL for QUIT.
2490 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2493 35. Added mailbox_size and mailbox_filecount to appendfile.
2495 36. Added control = no_multiline_responses to ACLs.
2497 37. There was a bug in the logic of the code that waits for the clock to tick
2498 in the case where the clock went backwards by a substantial amount such
2499 that the microsecond fraction of "now" was more than the microsecond
2500 fraction of "then" (but the whole seconds number was less).
2502 38. Added support for the libradius Radius client library this is found on
2503 FreeBSD (previously only the radiusclient library was supported).
2509 1. When certain lookups returned multiple values in the form name=value, the
2510 quoting of the values was not always being done properly. Specifically:
2511 (a) If the value started with a double quote, but contained no whitespace,
2513 (b) If the value contained whitespace other than a space character (i.e.
2514 tabs or newlines or carriage returns) it was not quoted.
2515 This fix has been applied to the mysql and pgsql lookups by writing a
2516 separate quoting function and calling it from the lookup code. The fix
2517 should probably also be applied to nisplus, ibase and oracle lookups, but
2518 since I cannot test any of those, I have not disturbed their existing code.
2520 2. A hit in the callout cache for a specific address caused a log line with no
2521 reason for rejecting RCPT. Now it says "Previous (cached) callout
2522 verification failure".
2524 3. There was an off-by-one bug in the queryprogram router. An over-long
2525 return line was truncated at 256 instead of 255 characters, thereby
2526 overflowing its buffer with the terminating zero. As well as fixing this, I
2527 have increased the buffer size to 1024 (and made a note to document this).
2529 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2530 when Exim is waiting for an SMTP response from a remote server, Exim
2531 restarts its select() call on the socket, thereby resetting its timeout.
2532 This is not a problem when such interrupts are rare. Somebody set up a cron
2533 job to run exiwhat every 2 minutes, which is less than the normal select()
2534 timeout (5 or 10 minutes). This meant that the select() timeout never
2535 kicked in because it was always reset. I have fixed this by comparing the
2536 time when an interrupt arrives with the time at the start of the first call
2537 to select(). If more time than the timeout has elapsed, the interrupt is
2538 treated as a timeout.
2540 5. Some internal re-factoring in preparation for the addition of Sieve
2541 extensions (by MH). In particular, the "personal" test is moved to a
2542 separate function, and given an option for scanning Cc: and Bcc: (which is
2543 not set for Exim filters).
2545 6. When Exim created an email address using the login of the caller as the
2546 local part (e.g. when creating a From: or Sender: header line), it was not
2547 quoting the local part when it contained special characters such as @.
2549 7. Installed new OpenBSD configuration files.
2551 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2552 try to make them clearer.
2554 9. Callout options, other than the timeout value, were being ignored when
2555 verifying sender addresses in header lines. For example, when using
2557 verify = header_sender/callout=no_cache
2559 the cache was (incorrectly) being used.
2561 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2562 only the Cygwin environment.
2564 11. When return_path_on_delivery was set as a log selector, if different remote
2565 addresses in the same message used different return paths and parallel
2566 remote delivery occurred, the wrong values would sometimes be logged.
2567 (Whenever a remote delivery process finished, the return path value from
2568 the most recently started remote delivery process was logged.)
2570 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2571 lines when AUTH and/or TLS are in use. This is the "received protocol"
2572 field. Exim used to use "asmtp" for authenticated SMTP, without any
2573 indication (in the protocol name) for TLS use. Now it follows the RFC and
2574 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2575 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2576 names appear in log lines as well as in Received: header lines.
2578 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2579 extensions, and comparison tests, and to fix some bugs.
2581 14. Changes to the "personal" filter test:
2583 (1) The test was buggy in that it was just doing the equivalent of
2584 "contains" tests on header lines. For example, if a user's address was
2585 anne@some.where, the "personal" test would incorrectly be true for
2587 To: susanne@some.where
2589 This test is now done by extracting each address from the header in turn,
2590 and checking the entire address. Other tests that are part of "personal"
2591 are now done using regular expressions (for example, to check local parts
2592 of addresses in From: header lines).
2594 (2) The list of non-personal local parts in From: addresses has been
2595 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2596 taken from the Sieve specification recommendations.
2598 (3) If the message contains any header line starting with "List-" it is
2599 treated as non-personal.
2601 (4) The test for "circular" in the Subject: header line has been removed
2602 because it now seems ill-conceived.
2604 15. Minor typos in src/EDITME comments corrected.
2606 16. Installed latest exipick from John Jetmore.
2608 17. If headers_add on a router specified a text string that was too long for
2609 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2610 of string_sprintf() is now avoided.
2612 18. $message_body_size was not set (it was always zero) when running the DATA
2613 ACL and the local_scan() function.
2615 19. For the "mail" command in an Exim filter, no default was being set for
2616 the once_repeat time, causing a random time value to be used if "once" was
2617 specified. (If the value happened to be <= 0, no repeat happened.) The
2618 default is now 0s, meaning "never repeat". The "vacation" command was OK
2619 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2620 (I found it when inspecting the code).
2622 20. There is now an overall timeout for performing a callout verification. It
2623 defaults to 4 times the callout timeout, which applies to individual SMTP
2624 commands during the callout. The overall timeout applies when there is more
2625 than one host that can be tried. The timeout is checked before trying the
2626 next host. This prevents very long delays if there are a large number of
2627 hosts and all are timing out (e.g. when the network connections are timing
2628 out). The value of the overall timeout can be changed by specifying an
2629 additional sub-option for "callout", called "maxwait". For example:
2631 verify = sender/callout=5s,maxwait=20s
2633 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2634 to the end before writing, but this should make it even safer).
2636 22. Exim was forgetting that it had advertised PIPELINING for the second and
2637 subsequent messages on an SMTP connection. It was also not resetting its
2638 memory on STARTTLS and an internal HELO.
2640 23. When Exim logs an SMTP synchronization error within a session, it now
2641 records whether PIPELINING has been advertised or not.
2643 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2644 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2645 rather than long int.
2647 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2649 26. Added the never_mail option to autoreply.
2655 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2656 crash if the getsockname() call failed; this can happen if a connection is
2657 closed very soon after it is established. The problem was simply in the
2658 order in which certain operations were done, causing Exim to try to write
2659 to the SMTP stream before it had set up the file descriptor. The bug has
2660 been fixed by making things happen in the correct order.
2666 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2667 before the connection was closed, thus losing the rejection response.
2669 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2670 some early Solaris releases, but causes trouble in current releases where
2671 socklen_t is defined.
2673 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2676 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2679 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2680 that releasing the top of it at the end releases what was used for sub-
2681 expansions (unless the block got too big). However, discard this block if
2682 the first thing is a variable or header, so that we can use its block when
2683 it is dynamic (useful for very large $message_headers, for example).
2685 6. Lookups now cache *every* query, not just the most recent. A new, separate
2686 store pool is used for this. It can be recovered when all lookup caches are
2687 flushed. Lookups now release memory at the end of their result strings.
2688 This has involved some general refactoring of the lookup sources.
2690 7. Some code has been added to the store_xxx() functions to reduce the amount
2691 of flapping under certain conditions.
2693 8. log_incoming_interface used to affect only the <= reception log lines. Now
2694 it causes the local interface and port to be added to several more SMTP log
2695 lines, for example "SMTP connection from", and rejection lines.
2697 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2699 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2701 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2702 could be overwritten at the end of the current message (or the start of a
2703 new message if it was set in a HELO ACL). The value is now preserved for
2704 the duration of the SMTP connection.
2706 12. If a transport had a headers_rewrite setting, and a matching header line
2707 contained an unqualified address, that address was qualified, even if it
2708 did not match any rewriting rules. The underlying bug was that the values
2709 of the flags that permit the existence of unqualified sender and recipient
2710 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2711 non-local messages, and by -bnq for local messages) were not being
2712 preserved with the message after it was received.
2714 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2715 "next input=" as part of the text comprising the host identity instead of
2716 the correct text. The code was using the same buffer for two different
2717 strings. However, depending on which order the printing function evaluated
2718 its arguments, the bug did not always show up. Under Linux, for example, my
2719 test suite worked just fine.
2721 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2722 doesn't work with some older versions of Perl. It has been changed to "my",
2723 which in any case is probably the better facility to use.
2725 15. A really picky compiler found some instances of statements for creating
2726 error messages that either had too many or two few arguments for the format
2729 16. The size of the buffer for calls to the DNS resolver has been increased
2730 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2731 for addresses that have a lot of PTR records. This alleviates a problem; it
2732 does not fully solve it.
2734 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2735 buffer now truncates the list and logs the incident, which is the same
2736 action as happens when Exim is looking up a host name and its aliases.
2737 Previously in this situation something unpredictable would happen;
2738 sometimes it was "internal error: store_reset failed".
2740 18. If a server dropped the connection unexpectedly when an Exim client was
2741 using GnuTLS and trying to read a response, the client delivery process
2742 crashed while trying to generate an error log message.
2744 19. If a "warn" verb in an ACL added multiple headers to a message in a single
2745 string, for example:
2747 warn message = H1: something\nH2: something
2749 the text was added as a single header line from Exim's point of view
2750 though it ended up OK in the delivered message. However, searching for the
2751 second and subsequent header lines using $h_h2: did not work. This has been
2752 fixed. Similarly, if a system filter added multiple headers in this way,
2753 the routers could not see them.
2755 20. Expanded the error message when iplsearch is called with an invalid key to
2756 suggest using net-iplsearch in a host list.
2758 21. When running tests using -bh, any delays imposed by "delay" modifiers in
2759 ACLs are no longer actually imposed (and a message to that effect is
2762 22. If a "gecos" field in a passwd entry contained escaped characters, in
2763 particular, if it contained a \" sequence, Exim got it wrong when building
2764 a From: or a Sender: header from that name. A second bug also caused
2765 incorrect handling when an unquoted " was present following a character
2766 that needed quoting.
2768 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2769 was not being matched caselessly.
2771 24. Arranged for all hyphens in the exim.8 source to be escaped with
2774 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2775 properly. Recipient callout cache records were still being keyed to include
2776 the sender, even when use_sender was set false. This led to far more
2777 callouts that were necessary. The sender is no longer included in the key
2778 when use_sender is false.
2780 26. Added "control = submission" modifier to ACLs.
2782 27. Added the ${base62d: operator to decode base 62 numbers.
2784 28. dnsdb lookups can now access SRV records.
2786 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2787 the configuration file.
2789 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2790 (-v) mode. This makes the output for a verbose queue run more intelligible.
2792 31. Added a use_postmaster feature to recipient callouts.
2794 32. Added the $body_zerocount variable, containing the number of binary zero
2795 bytes in the message body.
2797 33. The time of last modification of the "new" subdirectory is now used as the
2798 "mailbox time last read" when there is a quota error for a maildir
2801 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2803 35. Added +ignore_unknown as a special item in host lists.
2805 36. Code for decoding IPv6 addresses in host lists is now included, even if
2806 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2807 address was recognized as an IP address, but was then not correctly decoded
2808 into binary, causing unexpected and incorrect effects when compared with
2815 1. Very minor rewording of debugging text in manualroute to say "list of
2816 hosts" instead of "hostlist".
2818 2. If verify=header_syntax was set, and a header line with an unqualified
2819 address (no domain) and a large number of spaces between the end of the
2820 name and the colon was received, the reception process suffered a buffer
2821 overflow, and (when I tested it) crashed. This was caused by some obsolete
2822 code that should have been removed. The fix is to remove it!
2824 3. When running in the test harness, delay a bit after writing a bounce
2825 message to get a bit more predictability in the log output.
2827 4. Added a call to search_tidyup() just before forking a reception process. In
2828 theory, someone could use a lookup in the expansion of smtp_accept_max_
2829 per_host which, without the tidyup, could leave open a database connection.
2831 5. Added the variables $recipient_data and $sender_data which get set from a
2832 lookup success in an ACL "recipients" or "senders" condition, or a router
2833 "senders" option, similar to $domain_data and $local_part_data.
2835 6. Moved the writing of debug_print from before to after the "senders" test
2838 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2839 problems for message scanning, either using a data ACL, or using
2840 local_scan() because the Received: header was not generated till after they
2841 were called (in order to set the time as the time of reception completion).
2842 I have revised the way this works. The header is now generated after the
2843 body is received, but before the ACL or local_scan() are called. After they
2844 are run, the timestamp in the header is updated.
2850 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2851 before starting a queue runner without re-exec. This happened only when
2852 deliver_drop_privilege was set or when the Exim user was set to root. The
2853 effect of the bug was that timeouts during subsequent deliveries caused
2854 crashes instead of being properly handled. The handler is now left at its
2855 default (and expected) setting.
2857 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2858 message, again when deliver_drop_privilege is set or Exim is run as root.
2859 The bug described in (1) was not present in this case, but the tidying up
2860 of the other signals was missing. I have made the two cases consistent.
2862 3. The ignore_target_hosts setting on a manualroute router was being ignored
2863 for hosts that were looked up using the /MX notation.
2865 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2868 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2869 operated on the sender address. After changing the $sender_address to <>
2870 for the sender address verify, Exim was re-instated it as the original
2871 (before rewriting) address, but remembering that it had rewritten it, so it
2872 wasn't rewriting it again. This bug also had the effect of breaking the
2873 sender address verification caching when the sender address was rewritten.
2875 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2876 This has been changed so that if the ip literal address matches
2877 ignore_target_hosts, the router declines.
2879 7. Added expansion conditions match_domain, match_address, and match_local_
2880 part (NOT match_host).
2882 8. The placeholder for the Received: header didn't have a length field set.
2884 9. Added code to Exim itself and to exim_lock to test for a specific race
2885 condition that could lead to file corruption when using MBX delivery. The
2886 issue is with the lockfile that is created in /tmp. If this file is removed
2887 after a process has opened it but before that process has acquired a lock,
2888 there is the potential for a second process to recreate the file and also
2889 acquire a lock. This could lead to two Exim processes writing to the file
2890 at the same time. The added code performs the same test as UW imapd; it
2891 checks after acquiring the lock that its file descriptor still refers to
2892 the same named file.
2894 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2895 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2898 11. Added the smtp_active_hostname option. If used, this will typically be made
2899 to depend on the incoming interface address. Because $interface_address is
2900 not set up until the daemon has forked a reception process, error responses
2901 that can happen earlier (such as "too many connections") no longer contain
2904 12. If an expansion in a condition on a "warn" statement fails because a lookup
2905 defers, the "warn" statement is abandoned, and the next ACL statement is
2906 processed. Previously this caused the whole ACL to be aborted.
2908 13. Added the iplsearch lookup type.
2910 14. Added ident_timeout as a log selector.
2912 15. Added tls_certificate_verified as a log selector.
2914 16. Added a global option tls_require_ciphers (compare the smtp transport
2915 option of the same name). This controls incoming TLS connections.
2917 17. I finally figured out how to make tls_require_ciphers do a similar thing
2918 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2919 before starting the TLS session.
2921 18. Tabs are now shown as \t in -bP output.
2923 19. If the log selector return_path_on_delivery was set, Exim crashed when
2924 bouncing a message because it had too many Received: header lines.
2926 20. If two routers both had headers_remove settings, and the first one included
2927 a superfluous trailing colon, the final name in the first list and the
2928 first name in the second list were incorrectly joined into one item (with a
2929 colon in the middle).
2935 1. Added -C and -D options to the exinext utility, mainly to make it easier
2936 to include in the automated testing, but these could be helpful when
2937 multiple configurations are in use.
2939 2. The exinext utility was not formatting the output nicely when there was
2940 an alternate port involved in the retry record key, nor when there was a
2941 message id as well (for retries that were specific to a specific message
2942 and a specific host). It was also confused by IPv6 addresses, because of
2943 the additional colons they contain. I have fixed the IPv4 problem, and
2944 patched it up to do a reasonable job for IPv6.
2946 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2947 delivery, the log line now contains "pipelined" if PIPELINING was used.
2949 4. An SMTP transport process used to panic and die if the bind() call to set
2950 an explicit outgoing interface failed. This has been changed; it is now
2951 treated in the same way as a connect() failure.
2953 5. A reference to $sender_host_name in the part of a conditional expansion
2954 that was being skipped was still causing a DNS lookup. This no longer
2957 6. The def: expansion condition was not recognizing references to header lines
2958 that used bh_ and bheader_.
2960 7. Added the _cache feature to named lists.
2962 8. The code for checking quota_filecount in the appendfile transport was
2963 allowing one more file than it should have been.
2965 9. For compatibility with Sendmail, the command line option
2973 and sets the incoming protocol and host name (for trusted callers). The
2974 host name and its colon can be omitted when only the protocol is to be set.
2975 Note the Exim already has two private options, -pd and -ps, that refer to
2976 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2977 "s", but I don't think that's a major issue.
2979 10. A number of refactoring changes to the code, none of which should affect
2982 (a) The number of logging options was getting close to filling up the
2983 32-bit word that was used as a bit map. I have split them into two classes:
2984 those that are passed in the argument to log_write(), and those that are
2985 only ever tested independently outside of that function. These are now in
2986 separate 32-bit words, so there is plenty of room for expansion again.
2987 There is no change in the user interface or the logging behaviour.
2989 (b) When building, for example, log lines, the code previously used a
2990 macro that called string_cat() twice, in order to add two strings. This is
2991 not really sufficiently general. Furthermore, there was one instance where
2992 it was actually wrong because one of the argument was used twice, and in
2993 one call a function was used. (As it happened, calling the function twice
2994 did not affect the overall behaviour.) The macro has been replaced by a
2995 function that can join an arbitrary number of extra strings onto a growing
2998 (c) The code for expansion conditions now uses a table and a binary chop
2999 instead of a serial search (which was left over from when there were very
3000 few conditions). Also, it now recognizes conditions like "pam" even when
3001 the relevant support is not compiled in: a suitably worded error message is
3002 given if an attempt is made to use such a condition.
3004 11. Added ${time_interval:xxxxx}.
3006 12. A bug was causing one of the ddress fields not to be passed back correctly
3007 from remote delivery subprocesses. The field in question was not being
3008 subsequently used, so this caused to problems in practice.
3010 13. Added new log selectors queue_time and deliver_time.
3012 14. Might have fixed a bug in maildirsizefile handling that threw up
3013 "unexpected character" debug warnings, and recalculated the data
3014 unnecessarily. In any case, I expanded the warning message to give more
3017 15. Added the message "Restricted characters in address" to the statements in
3018 the default ACL that block characters like @ and % in local parts.
3020 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3021 Three changes have been made:
3023 (a) There was a serious bug; a negative response to MAIL caused the whole
3024 recipient domain to be cached as invalid, thereby blocking all messages
3025 to all local parts at the same domain, from all senders. This bug has
3026 been fixed. The domain is no longer cached after a negative response to
3027 MAIL if the sender used is not empty.
3029 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3032 (c) A new callout option, "use_sender" has been added for people who want
3033 the modified behaviour.
3039 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3042 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3043 indeed breaks things for older releases.
3045 3. Added additional logging to the case where there is a problem reading data
3046 from a filter that is running in a subprocess using a pipe, in order to
3047 try to track down a specific problem.
3049 4. Testing facility fudge: when running in the test harness and attempting
3050 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3051 getting "No route to host". Convert this to a timeout.
3053 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3056 6. Some OS don't have socklen_t but use size_t instead. This affects the
3057 fifth argument of getsockopt() amongst other things. This is now
3058 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3059 can be set for individual OS. I have set it for SunOS5, OSF1, and
3060 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3061 some earlier ones do not.
3063 7. Change 4.30/15 was not doing the test caselessly.
3065 8. The standard form for an IPv6 address literal was being rejected by address
3066 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3067 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3068 this, as well as the form without the "IPv6" on the front (but only when
3069 address literals are enabled, of course).
3071 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3073 10. Exim crashed if a message with an empty sender address specified by -f
3074 encountered a router with an errors_to setting. This could be provoked only
3075 by a command such as
3079 where an empty string was supplied; "<>" did not hit this bug.
3081 11. Installed PCRE release 4.5.
3083 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3084 remained set. It is now erased.
3086 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3087 times from message ids (which are base 36 rather than the normal 62).
3089 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3090 were being counted as actual protocol errors, and logged if the log
3091 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3092 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3093 and DATA following a set of rejected RCPTs do not count as protocol errors.
3094 In other words, Exim assumes they were pipelined, though this may not
3095 actually be the case. Of course, in all cases the client gets an
3096 appropriate error code.
3098 15. If a lookup fails in an ACL condition, a message about the failure may
3099 be available; it is used if testing the ACL cannot continue, because most
3100 such messages specify what the cause of the deferral is. However, some
3101 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3102 that caused an old message to be retained and used if a later statement
3103 caused a defer, replacing the real cause of the deferral.
3105 16. If an IP address had so many PTR records that the DNS lookup buffer
3106 was not large enough to hold them, Exim could crash while trying to process
3107 the truncated data. It now detects and logs this case.
3109 17. Further to 4.21/58, another change has been made: if (and only if) the
3110 first line of a message (the first header line) ends with CRLF, a bare LF
3111 in a subsequent header line has a space inserted after it, so as not to
3112 terminate the header.
3114 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3115 unnecessarily, used atoi() instead of strtol(), and didn't check the
3116 termination when getting file sizes from file names by regex.
3118 19. Completely re-implemented the support for maildirsize files, in the light
3119 of a number of problems with the previous contributed implementation
3120 (4.30/29). In particular:
3122 . If the quota is zero, the maildirsize file is maintained, but no quota is
3125 . If the maildir directory does not exist, it is created before any attempt
3126 to write a maildirsize file.
3128 . The quota value in the file is just a cache; if the quota is changed in
3129 the transport, the new value overrides.
3131 . A regular expression is available for excluding directories from the
3134 20. The autoreply transport checks the characters in options that define the
3135 message's headers; it allows continued headers, but it was checking with
3136 isspace() after an embedded newline instead of explicitly looking for a
3139 21. If all the "regular" hosts to which an address was routed had passed their
3140 expiry times, and had not reached their retry times, the address was
3141 bounced, even if fallback hosts were defined. Now Exim should go on to try
3144 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3145 equivalent code in the SMTP transport. Some hosts send humungous responses
3146 to HELO/EHLO, more than 1024 it seems.
3148 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3149 gives compiler warnings in some environments. I've now done it "properly",
3152 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3153 (because of problems with the built-in one) was declared to return uschar *
3154 instead of char *, causing compiler failure.
3156 25. Fixed a file descriptor leak when processing alias/forward files.
3158 26. Fixed a minor format string issue in dbfn.c.
3160 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3162 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3163 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3166 29. When a sender address is verified, it is cached, to save repeating the test
3167 when there is more than one recipient in a message. However, when the
3168 verification involves a callout, it is possible for different callout
3169 options to be set for different recipients. It is too complicated to keep
3170 track of this in the cache, so now Exim always runs a verification when a
3171 callout is required, relying on the callout cache for the optimization.
3172 The overhead is duplication of the address routing, but this should not be
3175 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3176 to be verified with callout=postmaster, and the main callout worked but the
3177 postmaster check failed, the verification correctly failed. However, if a
3178 subsequent RCPT command asked for sender verification *without* the
3179 postmaster check, incorrect caching caused this verification also to fail,
3182 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3183 it was not caching the DNS options (qualify_single, search_parents) that
3184 were used when the lookup failed. A subsequent lookup with different
3185 options therefore always gave the same answer, though there were cases
3186 where it should not have. (Example: a "domains = !$mx_any" option on a
3187 dnslookup router: the "domains" option is always processed without any
3188 widening, but the router might have qualify_single set.) Now Exim uses the
3189 cached value only when the same options are set.
3191 32. Added John Jetmore's "exipick" utility to the distribution.
3193 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3194 than a timeout (e.g. a certificate is required, and is not provided), an
3195 Exim server now closes the connection immediately. Previously it waited for
3196 the client to close - but if the client is SSL, it seems that they each
3197 wait for each other, leading to a delay before one of them times out.
3199 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3200 maintained 0.8.x compatibility because I don't think many are using it, and
3201 it is clearly obsolete.
3203 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3206 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3207 client certificate was expired. A simple patch fixes this, though I don't
3208 understand the full logic of why the verify callback is called multiple
3211 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3212 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3213 which causes problems with some clients (such as the Certicom SSL Plus
3214 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3215 disables the coutermeasure allowing Eudora to connect."
3217 38. Exim was not checking that a write() to a log file succeeded. This could
3218 lead to Bad Things if a log got too big, in particular if it hit a file
3219 size limit. Exim now panics and dies if it cannot write to a log file, just
3220 as it does if it cannot open a log file.
3222 39. Modified OS/Makefile-Linux so that it now contains
3224 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3226 The two -D definitions ensure that Exim is compiled with large file
3227 support, which makes it possible to handle log files that are bigger than
3230 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3231 instance) a domain was checked against a named list that involved a lookup,
3232 causing $domain_data to be set, then another domain was checked against the
3233 same list, then the first domain was re-checked, the value of $domain_data
3234 after the final check could be wrong. In particular, if the second check
3235 failed, it could be set empty. This bug probably also applied to
3238 41. The strip_trailing_dot option was not being applied to the address given
3239 with the -f command-line option.
3241 42. The code for reading a message's header from the spool was incrementing
3242 $received_count, but never initializing it. This meant that the value was
3243 incorrect (doubled) while delivering a message in the same process in which
3244 it was received. In the most common configuration of Exim, this never
3245 happens - a fresh exec is done - but it can happen when
3246 deliver_drop_privilege is set.
3248 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3249 it now includes up to 150 characters of the unexpected data in the log
3252 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3253 and building data strings. The size of both of these buffers was 10 000
3254 bytes - far larger than anybody would *ever* want, thought I. Needless to
3255 say, somebody hit the limit. I have increased the maximum line length to
3256 20 000 and the maximum data length of concatenated lines to 100 000. I have
3257 also fixed two bugs, because there was no checking on these buffers. Tsk,
3258 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3259 buffer is too small.
3261 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3262 lsearch lookups. Now it does.
3264 46. When parsing a route_list item in a manualroute router, a fixed-length
3265 buffer was used for the list of hosts. I made this 1024 bytes long,
3266 thinking that nobody would ever have a list of hosts that long. Wrong.
3267 Somebody had a whole pile of complicated expansion conditions, and the
3268 string was silently truncated, leading to an expansion error. It turns out
3269 that it is easier to change to an unlimited length (owing to other changes
3270 that have happened since this code was originally written) than to build
3271 structure for giving a limitation error. The length of the item that
3272 expands into the list of hosts is now unlimited.
3274 47. The lsearch lookup could not handle data where the length of text line was
3275 more than 4095 characters. Such lines were truncated, leading to shortened
3276 data being returned. It should now handle lines of any length.
3278 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3279 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3282 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3283 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3284 no longer happens in comment lines. A list of replacements is now placed
3285 at the head of all of the source files, except those whose only change is
3286 to replace PERL_COMMAND in the very first #! line.
3288 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3289 messages on the queue, with a bottom-up merge sort, using code contributed
3290 by Michael Haardt. This should make operations like -bp somewhat faster on
3291 large queues. It won't affect queue runners, except when queue_run_in_order
3294 51. Installed eximstats 1.31 in the distribution.
3296 52. Added support for SRV lookups to the dnslookup router.
3298 53. If an ACL referred to $message_body or $message_body_end, the value was not
3299 reset for any messages that followed in the same SMTP session.
3301 54. The store-handling optimization for building very long strings was not
3302 differentiating between the different store pools. I don't think this
3303 actually made any difference in practice, but I've tidied it.
3305 55. While running the routers to verify a sender address, $sender_address
3306 was still set to the sender address. This is wrong, because when routing to
3307 send a bounce to the sender, it would be empty. Therefore, I have changed
3308 it so that, while verifying a sender address, $sender_address is set to <>.
3309 (There is no change to what happens when verifying a recipient address.)
3311 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3312 target A or AAAA records (if not already returned) without resetting the
3313 qualify_single or search_parents options of the DNS resolver. These are
3314 inappropriate in this case because the targets of MX and SRV records must
3315 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3316 target that, when qualified, matched something in the local domain. These
3317 two options are now turned off when doing these lookups.
3319 57. It seems that at least some releases of Reiserfs (which does not have the
3320 concept of a fixed number of inodes) returns zero and not -1 for the
3321 number of available inodes. This interacted badly with check_spool_inodes,
3322 which assumed that -1 was the "no such thing" setting. What I have done is
3323 to check that the total number of inodes is greater than zero before doing
3324 the test of how many are available.
3326 58. When a "warn" ACL statement has a log_message modifier, the message is
3327 remembered, and not repeated. This is to avoid a lot of repetition when a
3328 message has many recipients that cause the same warning to be written.
3329 Howewer, Exim was preserving the list of already written lines for an
3330 entire SMTP session, which doesn't seem right. The memory is now reset if a
3331 new message is started.
3333 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3334 debugging output unless log_rewrite was also set.
3336 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3337 of (int)(handle) when we know that handle contains (void *)(-1).
3339 61. The Exim daemon panic-logs an error return when it closes the incoming
3340 connection. However "connection reset by peer" seems to be common, and
3341 isn't really an error worthy of noting specially, so that particular error
3344 62. When Exim is trying to find all the local interfaces, it used to panic and
3345 die if the ioctl to get the interface flags failed. However, it seems that
3346 on at least one OS (Solaris 9) it is possible to have an interface that is
3347 included in the list of interfaces, but for which you get a failure error
3348 for this call. This happens when the interface is not "plumbed" into a
3349 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3350 failure of the "get flags" call assumes that the interface is down.
3352 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3353 makes life easier for people who are doing arithmetic on fields extracted
3354 from dates, where you often get leading zeros that should not be
3355 interpreted as octal.
3357 64. Added qualify_domain to the redirect router, to override the global
3360 65. If a pathologically long header line contained very many addresses (the
3361 report of this problem mentioned 10 000) and each of them was rewritten,
3362 Exim could use up a very large amount of memory. (It kept on making new
3363 copies of the header line as it rewrote, and never released the old ones.)
3364 At the expense of a bit more processing, the header rewriting function has
3365 been changed so that it no longer eats memory in this way.
3367 66. The generation of the Received: header has been moved from the time that a
3368 message starts to be received, to the time that it finishes. The timestamp
3369 in the Received: header should now be very close to that of the <= log
3370 line. There are two side-effects of this change:
3372 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3373 logged header lines no longer include the local Received: line, because
3374 it has not yet been created. The same applies to a copy of the message
3375 that is returned to a non-SMTP sender when a message is rejected.
3377 (b) When a filter file is tested using -bf, no additional Received: header
3378 is added to the test message. After some thought, I decided that this
3381 This change does not affect the value of $received_for. It is still set
3382 after address rewriting, but before local_scan() is called.
3384 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3386 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3387 gave an unhelpful panic error message, and a defer error. I have managed to
3388 change this behaviour so that it now rejects any supplied certificate,
3389 which seems right, as the list of acceptable certificates is empty.
3391 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3392 gave an unhelpful defer error. I have not managed to make this reject any
3393 supplied certificates, but the error message it gives is "no certificate
3394 supplied", which is not helpful.
3396 70. exigrep's output now also includes lines that are not associated with any
3397 message, but which match the given pattern. Implemented by a patch from
3398 Martin Sluka, which also tidied up the Perl a bit.
3400 71. Recipient callout verification, like sender verification, was using <> in
3401 the MAIL FROM command. This isn't really the right thing, since the actual
3402 sender may affect whether the remote host accepts the recipient or not. I
3403 have changed it to use the actual sender in the callout; this means that
3404 the cache record is now keyed on a recipient/sender pair, not just the
3405 recipient address. There doesn't seem to be a real danger of callout loops,
3406 since a callout by the remote host to check the sender would use <>.
3407 [SEE ABOVE: changed after hitting problems.]
3409 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3410 temporary errors. However, in the case of such a code being given after
3411 the end of a data transmission (i.e. after ".") Exim was failing to write
3412 a retry record for the message. (Yes, there was some broken host that was
3413 actually sending 8xx at this point.)
3415 73. An unknown lookup type in a host list could cause Exim to panic-die when
3416 the list was checked. (An example that provoked this was putting <; in the
3417 middle of a list instead of at the start.) If this happened during a DATA
3418 ACL check, a -D file could be left lying around. This kind of configuration
3419 error no longer causes Exim to die; instead it causes a defer errror. The
3420 incident is still logged to the main and panic logs.
3422 74. Buglet left over from Exim 3 conversion. The message "too many messages
3423 in one connection" was written to the rejectlog but not the mainlog, except
3424 when address rewriting (yes!) was being logged.
3426 75. Added write_rejectlog option.
3428 76. When a system filter was run not as root (that is, when system_filter_user
3429 was set), the values of the $n variables were not being returned to the
3430 main process; thus, they were not subsequently available in the $sn
3433 77. Added +return_path_on_delivery log selector.
3435 78. A connection timeout was being treated differently from recipients deferred
3436 when testing hosts_max_try with a message that was older than the host's
3437 retry timeout. (The host should not be counted, thus allowing all hosts to
3438 be tried at least once before bouncing.) This may have been the cause of an
3439 occasionally reported bug whereby a message would remain on the queue
3440 longer than the retry timeout, but would be bounced if a delivery was
3441 forced. I say "may" because I never totally pinned down the problem;
3442 setting up timeout/retry tests is difficult. See also the next item.
3444 79. The ultimate address timeout was not being applied to errors that involved
3445 a combination of host plus message (for example, a timeout on a MAIL
3446 command). When an address resolved to a number of possible hosts, and they
3447 were not all tried for each delivery (e.g. because of hosts_max_try), a
3448 message could remain on the queue longer than the retry timeout.
3450 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3453 81. Fixed an obscure SMTP outgoing bug which required at least the following
3454 conditions: (a) there was another message waiting for the same server;
3455 (b) the server returned 5xx to all RCPT commands in the first message so
3456 that the message was not completed; (c) the server dropped the connection
3457 or gave a negative response to the RSET that Exim sends to abort the
3458 transaction. The observed case was a dropped connection after DATA that had
3459 been sent in pipelining mode. That is, the server had advertised PIPELINING
3460 but was not implementing it correctly. The effect of the bug was incorrect
3461 behaviour, such as trying another host, and this could lead to a crash.
3467 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3468 and daemon.c were passed as pointers to ints; they should have been
3469 pointers to socklen_t variables (which are typically unsigned ints).
3471 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3474 3. Fixed a really odd bug that affected only the testing scheme; patching a
3475 certain fixed string in the binary changed the value of another string that
3476 happened to be identical to the end of the original first string.
3478 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3479 name", it returns that address as the IP address. On some operating
3480 systems (e.g. Solaris), it also passes back the IP address string as the
3481 "host name". However, on others (e.g. Linux), it passes back an empty
3482 string. Exim wasn't checking for this, and was changing the host name to an
3483 empty string, assuming it had been canonicized.
3485 5. Although rare, it is permitted to have more than one PTR record for a given
3486 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3487 all the names associated with an address, because they do in Solaris.
3488 However, it seems that they do not in Linux for data that comes from the
3489 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3490 I found this out when I moved to a new Linux workstation and tried to run
3491 the Exim test suite.
3493 To get round this problem I have changed the code so that it now does its
3494 own call to the DNS to look up PTR records when searching for a host name.
3495 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3496 addresses that are only in /etc/hosts are still found.
3498 This behaviour is, however, controlled by an option called host_lookup_
3499 order, which defaults to "bydns:byaddr". If people want to use the other
3500 order, or indeed, just use one or the other means of lookup, they can
3501 specify it in this variable.
3503 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3504 some operating systems, this comes back from gethostbyaddr() as an empty
3505 string, and this is what Exim used to test for. However, it seems that in
3506 other systems, "." is yielded. Exim now tests for this case too.
3508 7. The values of check_spool_space and check_log_space are now held internally
3509 as a number of kilobytes instead of an absolute number of bytes. If a
3510 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3511 kilobyte. This means that much larger values can be stored.
3513 8. Exim monitor: an attempt to get the action menu when not actually pointing
3514 at a message produces an empty menu entitled "No message selected". This
3515 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3516 no entries in it ("Shell widget menu has zero width and/or height"). So I
3517 have added a single, blank menu entry in this case.
3519 9. Added ${quote_local_part.
3521 10. MIME decoding is now applied to the contents of Subject: header lines when
3524 11. Now that a reference to $sender_host_address automatically causes a reverse
3525 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3526 host lookup before query-style lookups in lists that might use this
3527 variable. This has therefore been abolished, and the "net-" prefix is no
3528 longer necessary for query-style lookups.
3530 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3531 have been a typo for LFLAGS, so it has been changed.
3533 13. The install script calls Exim with "-C /dev/null" in order to find the
3534 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3535 to be output. Howeve, since Exim outputs its version number before the
3536 error, it didn't break the script. It just looked ugly. I fixed this by
3537 always allowing "-C /dev/null" if the caller is root.
3539 14. Ignore overlarge ACL variable number when reading spool file - insurance
3540 against a later release with more variables having written the file.
3542 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3543 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3544 this, as well as the form without the "IPv6" on the front.
3546 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3547 OS/Makefile-Darwin file.
3549 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3550 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3552 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3553 variable. However, some OS use ints for inodes. I've added cast to long int
3554 to get rid of the compiler warning.
3556 19. I had forgotten to lock out "/../" in configuration file names when
3557 ALT_CONFIG_PREFIX was set.
3559 20. Routers used for verification do not need to specify transports. However,
3560 if such a router generated a host list, and callout was configured, Exim
3561 crashed, because it could not find a port number from the (non-existent)
3562 transport. It now assumes port 25 in this circumstance.
3564 21. Added the -t option to exigrep.
3566 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3567 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3568 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3569 them is set, LOOKUP_LSEARCH is forced.
3571 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3573 24. Added sender and host information to the "rejected by local_scan()" log
3574 line; previously there was no indication of these.
3576 25. Added .include_if_exists.
3578 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3579 Linux. It turns out that not all file systems support this. Apparently some
3580 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3581 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3582 not supported on the file descriptor, is now ignored when Exim is trying to
3583 sync a directory. This applies only to Linux.
3585 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3587 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3588 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3591 29. Added support for maildirsize files from supplied patch (modified a bit).
3593 30. The use of :fail: followed by an empty string could lead Exim to respond to
3594 sender verification failures with (e.g.):
3596 550 Verification failed for <xxx>
3597 550 Sender verify failed
3599 where the first response line was missing the '-' that indicates it is not
3600 the final line of the response.
3602 31. The loop for finding the name of the user that called Exim had a hardwired
3603 limit of 10; it now uses the value of finduser_retries, which is used for
3604 all other user lookups.
3606 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3609 33. Exim was neglecting to zero errno before one call of strtol() when
3610 expanding a string and expecting an integer value. On some systems this
3611 resulted in spurious "integer overflow" errors. Also, it was casting the
3612 result into an int without checking.
3614 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3615 did not work. The code looks as if it has *never* worked, though it appears
3616 to have been documented since at least releast 1.62. I have made it work.
3618 35. The "timeout_DNS" error in retry rules, also documented since at least
3619 1.62, also never worked. As it isn't clear exactly what this means, and
3620 clearly it isn't a major issue, I have abolished the feature by treating it
3621 as "timeout", and writing a warning to the main and panic logs.
3623 36. The display of retry rules for -brt wasn't always showing the error code
3626 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3627 timeout_connect_A, timeout_connect_MX.
3629 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3630 to the empty sender.
3632 39. The daemon was not analysing the content of -oX till after it had closed
3633 stderr and disconnected from the controlling terminal. This meant that any
3634 syntax errors were only noted on the panic log, and the return code from
3635 the command was 0. By re-arranging the code a little, I've made the
3636 decoding happen first, so such errors now appear on stderr, and the return
3637 code is 1. However, the actual setting up of the sockets still happens in
3638 the disconnected process, so errors there are still only recorded on the
3641 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3642 connections (as happens on some IP stacks) was logged at start up time as
3643 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3644 it from "IPv6 and IPv4", which means that two separate sockets are being
3647 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3648 says whether AF_INET or AF_INET6 was passed as an argument.
3650 42. Exiwhat output was messed up when time zones were included in log
3653 43. Exiwhat now gives more information about the daemon's listening ports,
3654 and whether -tls-on-connect was used.
3656 44. The "port" option of the smtp transport is now expanded.
3658 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3659 silently ignored. Now an error message is written to the main and panic
3662 46. There's a new ACL modifier called "logwrite" which writes to a log file
3663 as soon as it is encountered.
3665 47. Added $local_user_uid and $local_user_gid at routing time.
3667 48. Exim crashed when trying to verify a sender address that was being
3670 49. Exim was recognizing only a space character after ".include". It now also
3671 recognizes a tab character.
3673 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3674 extracting the relevant information from the specification. The man page no
3675 longer contains scrambled data for the -d option, and I've added a section
3676 at the front about calling Exim under different names.
3678 51. Added "extra_headers" argument to the "mail" command in filter files.
3680 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3683 53. Installed eximstats 1.29.
3685 54. Added transport_filter_timeout as a generic transport option.
3687 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3688 Cc: header lines. This was required by RFC 822, but it not required by RFC
3691 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3692 incoming messages that did not have them. Now it does so only if the
3693 message originates locally, that is, if there is no associated remote host
3694 address. When Resent- header lines are present, this applies to the Resent-
3695 lines rather than the non-Resent- lines.
3697 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3698 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3700 58. Messages for configuration errors now include the name of the main
3701 configuration file - useful now that there may be more than one file in a
3702 list (.included file names were always shown).
3704 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3705 for those rare installations that do not start the daemon as root or run it
3706 setuid root. I've cut out the call to initgroups() if the daemon is not
3709 60. The Exim user and group can now be bound into the binary as text strings
3710 that are looked up at the start of Exim's processing.
3712 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3714 62. Added $mailstore_basename variable.
3716 63. Installed patch to sieve.c from Michael Haardt.
3718 64. When Exim failed to open the panic log after failing to open the main log,
3719 the original message it was trying to log was written to stderr and debug
3720 output, but if they were not available (the usual case in production), it
3721 was lost. Now it is written to syslog before the two lines that record the
3722 failures to open the logs.
3724 65. Users' Exim filters run in subprocesses under the user's uid. It is
3725 possible for a "deliver" command or an alias in a "personal" command to
3726 provoke an address rewrite. If logging of address rewriting is configured,
3727 this fails because the process is not running as root or exim. There may be
3728 a better way of dealing with this, but for the moment (because 4.30 needs
3729 to be released), I have disabled address rewrite logging when running a
3730 filter in a non-root, non-exim process.
3736 1. The buildconfig auxiliary program wasn't quoting the value set for
3737 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3738 not defined. This bug was masked in 4.22 by the effect that was fixed in
3741 2. Some messages that were rejected after a message id was allocated were
3742 shown as "incomplete" by exigrep. It no longer does this for messages that
3743 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3745 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3746 have allow_domain_literals set, the ID did not get logged in the <= line.
3747 Domain literals are now always recognized in Message-ID: header lines.
3749 4. The first argument for a ${extract expansion item is the key name or field
3750 number. Leading and trailing spaces in this item were not being ignored,
3751 causing some misleading effects.
3753 5. When deliver_drop_privilege was set, single queue runner processes started
3754 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3755 same command in the process it spins off) were not dropping privilege.
3757 6. When the daemon running as "exim" started a queue runner, it always
3758 re-executed Exim in the spun-off process. This is a waste of effort when
3759 deliver_drop_privilege is set. The new process now just calls the
3760 queue-runner function directly.
3766 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3769 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3770 ignored. Though the use of -oP was forcing the writing of a pid file, it
3771 was always written to the default place.
3773 3. If the message "no IP address found for host xxxx" is generated during
3774 incoming verification, it is now followed by identification of the incoming
3775 connection (so you can more easily find what provoked it).
3777 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3779 5. Added some features to "harden" Exim a bit more against certain attacks:
3781 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3782 be put in Local/Makefile. This is like the never_users runtime option,
3783 but it cannot be overridden. The default setting is "root".
3785 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3786 prefix string with which any file named in a -C command line option
3789 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3790 is retained for -C and -D only if the caller of Exim is root. Without
3791 it, the exim user may also use -C and -D and retain privilege.
3793 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3794 command line option is disabled.
3796 6. Macro names set by the -D option must start with an upper case letter, just
3797 like macro names defined in the configuration file.
3799 7. Added "dereference=" facility to LDAP.
3801 8. Two instances of the typo "uknown" in the source files are fixed.
3803 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3804 the Configure-Makefile script screwed up while processing it.
3806 10. Incorporated PCRE 4.4.
3808 11. The SMTP synchronization check was not operating right at the start of an
3809 SMTP session. For example, it could not catch a HELO sent before the client
3810 waited for the greeting. There is now a check for outstanding input at the
3811 point when the greeting is written. Because of the duplex, asynchronous
3812 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3813 way, but not yet received, when the check is performed.
3815 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3816 on TCP/IP sockets, because this apparently causes some broken clients to
3819 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3820 unchanged) from the Cygwin maintainer.
3822 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3823 was supported instead of testing for mbx. Effectively a typo.
3825 15. The spa authenticator server code was not checking that the input it
3826 received was valid base64.
3828 16. The debug output line for the "set" modifier in ACLs was not showing the
3829 name of the variable that was being set.
3831 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3833 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3834 lookup on its own. Something else was needed to trigger the lookup. For
3835 example, a match in host_lookup or the need for a host name in a host list.
3836 Now, if $sender_host_name is referenced and the host name has not yet been
3837 looked up, a lookup is performed. If the lookup fails, the variable remains
3838 empty, and $host_lookup_failed is set to "1".
3840 19. Added "eqi" as a case-independent comparison operator.
3842 20. The saslauthd authentication condition could segfault if neither service
3843 nor realm was specified.
3845 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3846 error message that was logged was misleading, and incoming SMTP
3847 connections were dropped. The message is now more accurate, and temporary
3848 errors are given to SMTP connections.
3850 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3851 (except RSET) with a 421 error, until QUIT is received. However, it was
3852 failing to send a response to QUIT.
3854 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3855 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3856 ACL is now run after the helo_try_verify_hosts code.
3858 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3859 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3860 case-independent, so other case variants are also recognized). Apparently
3861 some systems use these upper case variants.
3863 25. If more than two messages were waiting for the same host, and a transport
3864 filter was specified for the transport, Exim sent two messages over the
3865 same TCP/IP connection, and then failed with "socket operation on non-
3866 socket" when it tried to send the third.
3868 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3870 27. The extern definition of crypt16() in expand.c was not being excluded when
3871 the OS had its own crypt16() function.
3873 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3874 as a preferred synonym for return_size_limit, both as an option and as an
3877 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3879 30. Changed the default configuration ACL to relax the local part checking rule
3880 for addresses that are not in any local domains. For these addresses,
3881 slashes and pipe symbols are allowed within local parts, but the sequence
3882 /../ is explicitly forbidden.
3884 31. SPA server authentication was not clearing the challenge buffer before
3887 32. log_message in a "warn" ACL statement was writing to the reject log as
3888 well as to the main log, which contradicts the documentation and doesn't
3889 seem right (because no rejection is happening). So I have stopped it.
3891 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3892 However, I am unable to do any testing of this.
3894 34. Fixed an infelicity in the appendfile transport. When checking directories
3895 for a mailbox, to see if any needed to be created, it was accidentally
3896 using path names with one or more superfluous leading slashes; tracing
3897 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3899 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3900 contents are added to the log line that is written for every discarded
3901 recipient. (Previously a log_message setting was ignored.)
3903 36. The ${quote: operator now quotes the string if it is empty.
3905 37. The install script runs exim in order to find its version number. If for
3906 some reason other than non-existence or emptiness, which it checks, it
3907 could not run './exim', it was installing it with an empty version number,
3908 i.e. as "exim-". This error state is now caught, and the installation is
3911 38. An argument was missing from the function that creates an error message
3912 when Exim fails to connect to the socket for saslauthd authentication.
3913 This could cause Exim to crash, or give a corrupted message.
3915 39. Added isip, isip4, and isip6 to ${if conditions.
3917 40. The ACL variables $acl_xx are now saved with the message, and can be
3918 accessed later in routers, transports, and filters.
3920 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3921 strings in the file are not string-expanded.
3923 42. If a MAIL command specified a SIZE value that was too large to fit into an
3924 int variable, the check against message_size_limit failed. Such values are
3925 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3926 day this will have to be increased, but I don't think I want to be around
3927 when emails are that large.
3934 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3935 iconv() is not standard in FreeBSD.
3937 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3938 IPv6 enabled. The observed symptom was a segmentation fault on return from
3939 the function os_common_find_running_interfaces() in src/os.c.
3941 3. In the check_special_case() function in daemon.c I had used "errno" as an
3942 argument name, which causes warnings on some systems. This was basically a
3943 typo, since it was named "eno" in the comments!
3945 4. The code that waits for the clock to tick (at a resolution of some fraction
3946 of a second) so as to ensure message-id uniqueness was always waiting for
3947 at least one whole tick, when it could have waited for less. [This is
3948 almost certainly not relevant at current processor speeds, where it is
3949 unlikely to ever wait at all. But we try to future-proof.]
3951 5. The function that sleeps for a time interval that includes fractions of a
3952 second contained a race. It did not block SIGALRM between setting the
3953 timer, and suspending (a couple of lines later). If the interval was short
3954 and the sigsuspend() was delayed until after it had expired, the suspension
3955 never ended. On busy systems this could lead to processes getting stuck for
3958 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3959 process, before it forks any delivery processes. The open lookup caching
3960 mechanism meant that the open file or database connection was passed into
3961 the delivery process. The problem was that delivery processes always tidy
3962 up cached lookup data. This could cause a problem for the next delivery
3963 process started by the queue runner, because the external queue runner
3964 process does not know about the closure. So the next delivery process
3965 still has data in the lookup cache. In the case of a file lookup, there was
3966 no problem because closing a file descriptor in a subprocess doesn't affect
3967 the parent. However, if the lookup was caching a connection to a database,
3968 the connection was closed, and the second delivery process was likely to
3969 see errors such as "PGSQL: query failed: server closed the connection
3970 unexpectedly". The problem has been fixed by closing all cached lookups
3971 in a queue runner before running a delivery process.
3973 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3974 seem to have the "const" qualifier which it has on other OS. I've
3977 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3978 *of the same type* (client or server) with the same public name that an
3979 error should be diagnosed.
3981 9. When Exim looked up a host name for an IP address, but failed to find the
3982 original IP address when looking up the host name (a safety check), it
3983 output the message "<ip address> does not match any IP for NULL", which was
3984 confusing, to say the least. The bug was that the host name should have
3985 appeared instead of "NULL".
3987 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3988 user that is built into the binary, and the -C or -D options is used, root
3989 privilege is dropped before the configuration file is read. In addition,
3990 logging is switched to stderr instead of the normal log files. If the
3991 configuration then re-defines the Exim user, the unprivileged environment
3992 is probably not what is expected, so Exim logs a panic warning message (but
3995 However, if deliver_drop_privilege is set, the unprivileged state may well
3996 be exactly what is intended, so the warning has been cut out in that case,
3997 and Exim is allowed to try to write to its normal log files.
4003 1. smtp_return_error_details was not giving details for temporary sender
4004 or receiver verification errors.
4006 2. Diagnose a configuration error if two authenticators have the same public
4009 3. Exim used not to create the message log file for a message until the first
4010 delivery attempt. This could be confusing when incoming messages were held
4011 for policy or load reasons. The message log file is now created at the time
4012 the message is received, and an initial "Received" line is written to it.
4014 4. The automatically generated man page for command line options had a minor
4015 bug that caused no ill effects; however, a more serious problem was that
4016 the procedure for building the man page automatically didn't always
4017 operate. Consequently, release 4.20 contains an out-of-date version. This
4018 shouldn't happen again.
4020 5. When building Exim with embedded Perl support, the script that builds the
4021 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4022 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4024 6. The freeze_tell option was not being used for messages that were frozen on
4025 arrival, either by an ACL or by local_scan().
4027 7. Added the smtp_incomplete_transaction log selector.
4029 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4030 was accepting AUTH without a new EHLO.
4032 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4033 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4036 10. Logging of TCP/IP connections (when configured) now happens in the main
4037 daemon process instead of the child process, so that the TCP/IP connection
4038 count is more accurate (but it can never be perfect).
4040 11. The use of "drop" in a nested ACL was not being handled correctly in the
4041 outer ACL. Now, if condition failure induced by the nested "drop" causes
4042 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4043 or "require"), the connection is dropped.
4045 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4046 that yield "discard" can now be used with an "accept" or a "discard" verb,
4047 but an error is generated for any others (because I can't see a useful way
4048 to define what should happen).
4050 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4051 are now processed in the same way as lines in the Exim configuration file.
4052 In particular, continuation lines are supported.
4054 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4056 15. Added -ti meaning -t -i.
4058 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4059 domains, and warn by logging if others are found.
4061 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4062 returned from GIFCONF when Exim is trying to find the list of interfaces on
4063 a host. The code in os.c has been modified to copy each ifreq to an aligned
4064 structure in all cases.
4066 Also, in some cases, the returned ifreq's were being copied to a 'struct
4067 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4068 means the last couple of bytes of an IPv6 address could be chopped if the
4069 ifreq contained only a normal sockaddr (14 bytes storage).
4071 18. Named domain lists were not supported in the hosts_treat_as_local option.
4072 An entry such as +xxxx was not recognized, and was treated as a literal
4075 19. Ensure that header lines added by a DATA ACL are included in the reject log
4076 if the ACL subsequently rejects the message.
4078 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4079 MD5 (which is deprecated).
4081 21. When testing a filter file using -bf, Exim was writing a message when it
4082 took the sender from a "From " line in the message, but it was not doing so
4083 when it took $return_path from a Return-Path: header line. It now does.
4085 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4086 with a valid header line field name (a series of printing characters
4087 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4089 23. Changed "disc" in the source to "disk" to conform to the documentation and
4090 the book and for uniformity.
4092 24. Ignore Sendmail's -Ooption=value command line item.
4094 25. When execve() failed while trying to run a command in a pipe transport,
4095 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4096 could be confused with a return value of 69 from the command itself. This
4097 has been changed to 127, the value the shell returns if it is asked to run
4098 a non-existent command. The wording for the related log line suggests a
4099 non-existent command as the problem.
4101 26. If received_header_text expands to an empty string, do not add a Received:
4102 header line to the message. (Well, it adds a token one on the spool, but
4103 marks it "old" so that it doesn't get used or transmitted.)
4105 27. Installed eximstats 1.28 (addition of -nt option).
4107 28. There was no check for failure on the call to getsockname() in the daemon
4108 code. This can fail if there is a shortage of resources on the system, with
4109 ENOMEM, for example. A temporary error is now given on failure.
4111 29. Contrary to the C standard, it seems that in some environments, the
4112 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4113 program. Exim now does this explicitly; it affects the formatting of
4114 timestamps using strftime().
4116 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4117 complaints. I've now initialized all the variables, to avoid this.
4119 32. Header lines added by a system filter were not being "seen" during
4120 transport-time rewrites.
4122 33. The info_callback() function passed to OpenSSL is set up with type void
4123 (*)(SSL *, int, int), as described somewhere. However, when calling the
4124 function (actually a macro) that sets it up, the type void(*)() is
4125 expected. I've put in a cast to prevent warnings from picky compilers.
4127 34. If a DNS black list lookup found a CNAME record, but there were no A
4128 records associated with the domain it pointed at, Exim crashed.
4130 35. If a DNS black list lookup returned more than one A record, Exim ignored
4131 all but the first. It now scans all returned addresses if a particular IP
4132 value is being sought. In this situation, the contents of the
4133 $dnslist_value variable are a list of all the addresses, separated by a
4136 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4137 to accept a host name and all its aliases if the forward lookup for any of
4138 them yielded the IP address of the incoming connection. Now it accepts only
4139 those names whose forward lookup yields the correct IP address. Any other
4140 names are discarded. This closes a loophole whereby a rogue DNS
4141 administrator could create reverse DNS records to break through a
4142 wildcarded host restriction in an ACL.
4144 37. If a user filter or a system filter that ran in a subprocess used any of
4145 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4146 the wrong values were passed to the pipe command ($thisaddress had the
4147 value of $0, $0 had the value of $1, etc). This bug was introduced by
4148 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4150 38. Improved the line breaking for long SMTP error messages from ACLs.
4151 Previously, if there was no break point between 40 and 75 characters, Exim
4152 left the rest of the message alone. Two changes have been made: (a) I've
4153 reduced the minimum length to 35 characters; (b) if it can't find a break
4154 point between 35 and 75 characters, it looks ahead and uses the first one
4155 that it finds. This may give the occasional overlong line, but at least the
4156 remaining text gets split now.
4158 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4159 file descriptors might be low, and that setting 1000 would always raise it.
4160 It turns out that in some environments, the limit is already over 1000 and
4161 that lowering it causes trouble. So now Exim takes care not to decrease it.
4163 40. When delivering a message, the value of $return_path is set to $sender_
4164 address at the start of routing (routers may change the value). By an
4165 oversight, this default was not being set up when an address was tested by
4166 -bt or -bv, which affected the outcome if any router or filter referred to
4169 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4170 log only when "message" or "log_message" are set. However, if one of the
4171 conditions was an address verification, or a call to a nested ACL, the
4172 messages generated by the underlying test were being passed through. This
4173 no longer happens. The underlying message is available in $acl_verify_
4174 message for both "message" and "log_message" expansions, so it can be
4175 passed through if needed.
4177 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4178 new expansion $bh_ to give the encoded byte string without charset
4179 translation. Translation happens only if iconv() is available; HAVE_ICONV
4180 indicates this at build time. HEADERS_CHARSET gives the charset to
4181 translate to; headers_charset can change it in the configuration, and
4182 "headers charset" can change it in an individual filter file.
4184 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4185 that creates RFC 2047 encoded "words" labels them as that charset instead
4186 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4187 expansion operator; (ii) when Exim creates a From: line for a local
4188 message; (iii) when a header line is rewritten to include a "phrase" part.
4190 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4191 buggy, causing it to skip the first lines of messages whose message ID
4192 ended in 'D'. This would not have bitten before Exim release 4.14, because
4193 message IDs were unlikely to end in 'D' before then. The effect was to have
4194 incorrect size information for certain domains.
4196 45. #include "config.h" was missing at the start of the crypt16.c module. This
4197 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4200 46. If there was a timeout during a "random" callout check, Exim treated it as
4201 a failure of the random address, and carried on sending RSET and the real
4202 address. If the delay was just some slowness somewhere, the response to the
4203 original RCPT would be taken as a response to RSET and so on, causing
4204 mayhem of various kinds.
4206 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4207 when I implemented it. It didn't allow for the fact that some option values
4208 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4209 the right test for positive values.
4211 48. Domain names in DNS records are case-independent. Exim always looks them up
4212 in lower case. Some resolvers return domain names in exactly the case they
4213 appear in the zone file, that is, they may contain uppercase letters. Not
4214 all resolvers do this - some return always lower case. Exim was treating a
4215 change of case by a resolver as a change of domain, similar to a widening
4216 of a domain abbreviation. This triggered its re-routing code and so it was
4217 trying to route what was effectively the same domain again. This normally
4218 caused routing to fail (because the router wouldn't handle the domain
4219 twice). Now Exim checks for this case specially, and just changes the
4220 casing of the domain that it ultimately uses when it transmits the message
4223 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4226 50. If a filter generated a file delivery with a non-absolute name (possible if
4227 no home directory exists for the router), the forbid_file option was not
4230 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4231 the existing equality matching.
4233 52. Exim was using ints instead of ino_t variables in some places where it was
4234 dealing with inode numbers.
4236 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4237 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4238 called TMPDIR, and if it finds it is different, it changes its value.
4240 54. The smtp_printf() function is now made available to local_scan() so
4241 additional output lines can be written before returning. There is also an
4242 smtp_fflush() function to enable the detection of a dropped connection.
4243 The variables smtp_input and smtp_batched_input are exported to
4246 55. Changed the default runtime configuration: the message "Unknown user"
4247 has been removed from the ACL, and instead placed on the localuser router,
4248 using the cannot_route_message feature. This means that any verification
4249 failures that generate their own messages won't get overridden. Similarly,
4250 the "Unrouteable address" message that was in the ACL for unverifiable
4251 relay addresses has also been removed.
4253 56. Added hosts_avoid_esmtp to the smtp transport.
4255 57. The exicyclog script was not checking for the esoteric option
4256 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4257 will work only if exicyclog is run under the appropriate euid.
4259 58. Following a discussion on the list, the rules by which Exim recognises line
4260 endings on incoming messages have been changed. The -dropcr and drop_cr
4261 options are now no-ops, retained only for backwards compatibility. The
4262 following line terminators are recognized: LF CRLF CR. However, special
4263 processing applies to CR:
4265 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4266 nor a local message in the state where . is a terminator.
4268 (ii) If a bare CR is encountered in a header line, an extra space is added
4269 after the line terminator so as not to end the header. The reasoning
4270 behind this is that bare CRs in header lines are most likely either
4271 to be mistakes, or people trying to play silly games.
4273 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4274 was being incorrectly given as 18 bytes larger than it should have been.
4275 This is a VOB (very old bug).
4277 60. This may never have affected anything current, but just in case it has:
4278 When the local host is found other than at the start of a list of hosts,
4279 the local host, those with the same MX, and any that follow, are discarded.
4280 When the list in question was part of a longer list of hosts, the following
4281 hosts (not currently being processed) were also being discarded. This no
4282 longer happens. I'm not sure if this situation could ever has previously
4285 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4288 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4289 header. This is something that is recommended in a new Internet Draft, and
4290 is something that is documented as being done by Sendmail. There are two
4291 possible values. For messages generated by the autoreply transport, Exim
4294 Auto-Submitted: auto-replied
4296 whereas for all other generated messages (e.g. bounces) it adds
4298 Auto-Submitted: auto-generated
4300 63. The "personal" condition in filters now includes a test for the
4301 Auto-Submitted: header. If it contains the string "auto-" the message it
4302 not considered personal.
4304 64. Added rcpt_include_affixes as a generic transport option.
4306 65. Added queue_only_override (default true).
4308 66. Added the syslog_duplication option.
4310 67. If what should have been the first header line of a message consisted of
4311 a space followed by a colon, Exim was mis-interpreting it as a header line.
4312 It isn't of course - it is syntactically invalid and should therefore be
4313 treated as the start of the message body. The misbehaviour could have
4314 caused a number of strange effects, including loss of data in subsequent
4315 header lines, and spool format errors.
4317 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4318 client host had authenticated. This control can now be exercised by an ACL
4319 for more flexibility.
4321 69. By default, callouts do not happen when testing with -bh. There is now a
4322 variant, -bhc, which does actually run the callout code, including
4323 consulting and updating the callout cache.
4325 70. Added support for saslauthd authentication, courtesy of Alexander
4328 71. If statvfs() failed on the spool or log directories while checking their
4329 size for availability, Exim confusingly gave the error "space shortage".
4330 Furthermore, in debugging mode it crashed with a floating point exception.
4331 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4332 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4333 serious problem, Exim now writes to the main and panic logs when this
4334 happens, with details of the failure. It then refuses to accept the
4335 incoming message, giving the message "spool directory problem" or "log
4336 directory problem" with a 421 code for SMTP messages.
4338 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4339 0, 1, and 2 exist, because some OS complain for execs without them (see
4340 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4341 descriptors. However, the code omitted to check that the open succeeded,
4342 causing mysterious errors if for some reason the permissions on /dev/null
4343 got screwed. Now Exim writes a message to the main and panic logs, and
4344 bombs out if it can't open /dev/null.
4346 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4347 interact so that it is all more flexible. It is supposed to remain
4348 backwards compatible. Also added extra_local_interfaces.
4350 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4351 to bomb out with an assertion failure - to the client this appears as a
4352 connection drop. This problem occurs in the part of the code that was taken
4353 from the Samba project. Fortunately, the assertion is in a very simple
4354 function, so I have fixed this by reproducing the function inline in the
4355 one place where it is called, and arranging for authentication to fail
4356 instead of killing the process with assert().
4358 75. The SPA client code was not working when the server requested OEM rather
4359 than Unicode encoding.
4361 76. Added code to make require_files with a specific uid setting more usable in
4362 the case where statting the file as root fails - usually a non-root-mounted
4363 NFS file system. When this happens and the failure is EACCES, Exim now
4364 forks a subprocess and does the per-uid checking as the relevant uid.
4366 77. Added process_log_path.
4368 78. If log_file_path was not explicitly set, a setting of check_log_space or
4369 check_log_inodes was ignored.
4371 79. If a space check for the spool or log partitions fails, the incident is now
4372 logged. Of course, in the latter case the data may get lost...
4374 80. Added the %p formatting code to string_format() so that it can be used to
4375 print addresses in debug_print(). Adjusted all the address printing in the
4376 debugging in store.c to use %p rather than %d.
4378 81. There was a concern that a line of code in smtp_in.c could overflow a
4379 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4380 initially expressed, the concern was not well-founded, because trailing
4381 spaces are removed early. However, if the trailing spaces were followed by
4382 a NULL, they did not get removed, so the overflow was possible. Two fixes
4385 (a) I re-wrote the offending code in a cleaner fashion.
4386 (b) If an incoming SMTP command contains a NULL character, it is rejected
4389 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4390 runs initgroups(), so that if the Exim user is in any additional groups,
4391 they will be used during message reception.
4397 The change log for 4.20 and earlier releases has been archived.