1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.430 2006/11/13 11:26:37 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).
271 SC/01 Use a glob alias rather than an array ref in eximstats generated
272 parser. This improves both readability and performance.
274 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
275 Don't display local sender or destination tables in eximstats unless
276 there is data to show.
277 Added average volumes into the eximstats top table text output.
279 SC/03 Collect data on the number of addresses (recipients) as well
280 as the number of messages in eximstats.
282 TF/01 Correct an error in the documentation for the redirect router. Exim
283 does (usually) call initgroups() when daemonizing.
285 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
286 with consistent privilege compared to when running as a daemon.
288 TF/03 Note in the spec that $authenticated_id is not set for local
289 submissions from trusted users.
291 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
292 Thanks to Dean Brooks <dean@iglou.com> for the patch.
294 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
295 by adding some example configuration directives to the default
296 configuration file. A little bit of work is required to uncomment the
297 directives and define how usernames and passwords are checked, but
298 there is now a framework to start from.
300 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
301 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
302 without this. I don't know how relevant this is to other LDAP libraries.
304 PH/02 Add the verb name to the "unknown ACL verb" error.
306 PH/03 Magnus Holmgren's patch for filter_prepend_home.
308 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
310 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
311 directory not expanded when it should be if an expanded home directory
312 was set for the address (which is overridden by the transport).
314 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
317 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
318 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
319 because it is too late at that time, and has no effect.
321 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
322 security issue with \' (bugzilla #107). I could not use the
323 PQescapeStringConn() function, because it needs a PGconn value as one of
326 PH/08 When testing addresses using -bt, indicate those final addresses that
327 are duplicates that would not cause an additional delivery. At least one
328 person was confused, thinking that -bt output corresponded to deliveries.
329 (Suppressing duplicates isn't a good idea as you lose the information
330 about possibly different redirections that led to the duplicates.)
332 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
333 systems where poll() doesn't work, in particular OS X.
335 PH/10 Added more information to debugging output for retry time not reached.
337 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
338 operations in malware.c.
340 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
343 PH/13 If write_rejectlog was set false when logging was sent to syslog with
344 syslog_duplication set false, log lines that would normally be written
345 both the the main log and to the reject log were not written to syslog at
348 PH/14 In the default configuration, change the use of "message" in ACL warn
349 statements to "add_header".
351 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
352 not followed by a command (e.g. "seen endif").
354 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
355 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
358 PH/17 Added extra conditions to the default value of delay_warning_condition
362 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
363 { match{$h_precedence:}{(?i)bulk|list|junk} } \
364 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
367 The Auto-Submitted: and various List- headers are standardised, whereas I
368 don't think Precedence: ever was.
370 PH/18 Refactored debugging code in route_finduser() to show more information,
371 in particular, the error code if getpwnam() issues one.
373 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
374 This is apparently needed in addition to the PH/07 change above to avoid
375 any possible encoding problems.
377 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
378 but not after initializing Perl.
380 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
381 output them only if debugging. By default they are written stderr,
382 apparently, which is not desirable.
384 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
387 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
390 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
392 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
393 authenticated or an ident call has been made. Suppress the default
394 values for $authenticated_id and $authenticated_sender (but permit -oMai
395 and -oMas) when testing with -bh.
397 PH/24 Re-jigged the order of the tests in the default configuration so that the
398 tests for valid domains and recipients precede the DNS black list and CSA
399 tests, on the grounds that those ones are more expensive.
401 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
402 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
403 This bug exists in every version of Exim that I still have, right back to
406 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
407 However, an attempt to turn on case-sensitivity in a regex key by
408 including (?-i) didn't work because the subject string was already
409 lowercased, and the effects were non-intuitive. It turns out that a
410 one-line patch can be used to allow (?-i) to work as expected.
416 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
417 other effects) broke the use of negated acl sub-conditions.
419 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
422 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
423 "Deny" causes Exim to reject the incoming connection with a 554 error.
424 Unfortunately, if there is a major crisis, such as a disk failure,
425 tcp-wrappers gives "deny", whereas what one would like would be some
426 kind of temporary error. A kludge has been added to help with this.
427 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
428 554 error is used if errno is still zero or contains ENOENT (which occurs
429 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
432 PH/03 Add -lutil to the default FreeBSD LIBS setting.
434 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
435 errors. Otherwise a message that provokes a temporary error (when other
436 messages do not) can cause a whole host to time out.
438 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
439 addresses were routed directly to files or pipes from a redirect router.
440 File deliveries just didn't batch; pipe deliveries might have suffered
443 PH/06 A failure to get a lock for a hints database would erroneously always say
444 "Failed to get write lock", even when it was really a read lock.
446 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
447 of 0600. This has been changed to use the value of the lockfile_mode
448 option (which defaults to 0600).
450 PH/08 Applied small patch from the Sieve maintainer.
452 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
453 folder from quota calculations, a direct delivery into this folder messed
454 up the contents of the maildirsize file. This was because the regex was
455 used only to exclude .Trash (or whatever) when the size of the mailbox
456 was calculated. There was no check that a delivery was happening into an
457 excluded directory. This bug has been fixed by ignoring all quota
458 processing for deliveries into excluded directories.
460 PH/10 Added the maildirfolder_create_regex option to appendfile.
466 PH/01 The code for finding all the local interface addresses on a FreeBSD
467 system running IPv6 was broken. This may well have applied to all BSD
468 systems, as well as to others that have similar system calls. The broken
469 code found IPv4 interfaces correctly, but gave incorrect values for the
470 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
471 that it would not match correctly against @[] and not recognize the IPv6
474 PH/02 The ipliteral router was not recognizing addresses of the form user@
475 [ipv6:....] because it didn't know about the "ipv6:" prefix.
477 PH/03 Added disable_ipv6.
479 PH/04 Changed $reply_address to use the raw form of the headers instead of the
480 decoded form, because it is most often used to construct To: headers
481 lines in autoreplies, and the decoded form may well be syntactically
482 invalid. However, $reply_address has leading white space removed, and all
483 newlines turned into spaces so that the autoreply transport does not
486 PH/05 If group was specified without a user on a router, and no group or user
487 was specified on a transport, the group from the router was ignored.
489 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
490 for visible compile-time settings that can be used to change these
491 numbers, for those that want even more. Backwards compatibility with old
492 spool files has been maintained. However, going back to a previous Exim
493 release will lost any variables that are in spool files.
495 PH/07 Two small changes when running in the test harness: increase delay when
496 passing a TCP/IP connection to a new process, in case the original
497 process has to generate a bounce, and remove special handling of
498 127.0.0.2 (sic), which is no longer necessary.
500 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
501 be the same on different OS.
503 PH/09 Moved a debug statement in filter processing to avoid a race problem when
506 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
507 whether --show-vars was specified or not
509 JJ/02 exipick: Added support for new ACL variable spool format introduced
512 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
513 syntactically invalid From: or Reply-to: line, and a filter used this to
514 generate an autoreply, and therefore failed to obtain an address for the
515 autoreply, Exim could try to deliver to a non-existent relative file
516 name, causing unrelated and misleading errors. What now happens is that
517 it logs this as a hard delivery error, but does not attempt to create a
520 PH/11 The exinext utility has a -C option for testing purposes, but although
521 the given file was scanned by exinext itself; it wasn't being passed on
522 when Exim was called.
524 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
525 an end-of-file indication when reading a command response.
527 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
528 compiled. In many other places in Exim, IPv6 addresses are always
529 recognized, so I have changed this. It also means that IPv4 domain
530 literals of the form [IPV4:n.n.n.n] are now always recognized.
532 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
533 used if the router is not running as root, for example, when verifying at
534 ACL time, or when using -bh. The debugging output from this situation was
535 non-existent - all you got was a failure to exec. I have made two
538 (a) Failures to set uid/gid, the current directory, or a process leader
539 in a subprocess such as that created by queryprogram now generate
540 suitable debugging ouput when -d is set.
542 (b) The queryprogram router detects when it is not running as root,
543 outputs suitable debugging information if -d is set, and then runs
544 the subprocess without attempting to change uid/gid.
546 PH/15 Minor change to Makefile for building test_host (undocumented testing
549 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
550 additional section of a DNS packet that returns MX or SRV records.
551 Instead, it always explicitly searches for A/AAAA records. This avoids
552 major problems that occur when a DNS server includes only records of one
553 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
554 fixed another bug: if SRV records were looked up and the corresponding
555 address records were *not* found in the additional section, the port
556 values from the SRV records were lost.
558 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
559 using the correct key (the original address) when searching the retry
560 rules in order to find which one to use for generating the retry hint.
562 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
563 adding the default one. Similarly, if it contains a Reply-To: header, the
564 errors_reply_to option, if set, is not used.
566 PH/19 When calculating a retry time, Exim used to measure the "time since
567 failure" by looking at the "first failed" field in the retry record. Now
568 it does not use this if it is later than than the arrival time of the
569 message. Instead it uses the arrival time. This makes for better
570 behaviour in cases where some deliveries succeed, thus re-setting the
571 "first failed" field. An example is a quota failure for a huge message
572 when small messages continue to be delivered. Without this change, the
573 "time since failure" will always be short, possible causing more frequent
574 delivery attempts for the huge message than are intended.
575 [Note: This change was subsequently modified - see PH/04 for 4.62.]
577 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
578 $1, $2, $3) because the numerical variables can be reset during some
579 expansion items (e.g. "match"), thereby losing the authentication data.
581 PH/21 Make -bV show the size of off_t variables so that the test suite can
582 decide whether to run tests for quotas > 2G.
584 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
585 mailbox_size, and mailbox_filecount in the appendfile transport. If a
586 filecount value is greater than 2G or if a quota value is greater than 2G
587 on a system where the size of off_t is not greater than 4, a panic error
590 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
591 never match. The debug and -bh output now contains an explicit error
592 message indicating a malformed IPv4 address or mask.
594 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
595 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
598 PH/25 Do not write to syslog when running in the test harness. The only
599 occasion when this arises is a failure to open the main or panic logs
600 (for which there is an explicit test).
602 PH/26 Added the /no_tell option to "control=freeze".
604 PH/27 If a host name lookup failed very early in a connection, for example, if
605 the IP address matched host_lookup and the reverse lookup yielded a name
606 that did not have a forward lookup, an error message of the form "no IP
607 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
608 could be logged. Now it outputs the IP address instead of "NULL".
610 PH/28 An enabling patch from MH: add new function child_open_exim2() which
611 allows the sender and the authenticated sender to be set when
612 submitting a message from within Exim. Since child_open_exim() is
613 documented for local_scan(), the new function should be too.
615 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
616 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
617 results in an empty string is now treated as unset.
619 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
621 PH/31 Added sender_verify_fail logging option.
623 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
624 needed by RFC 822 but not by RFC 2822 was commented out. I have now
625 tidied the source and removed it altogether.
627 PH/33 When a queue run was abandoned because the load average was too high, a
628 log line was always written; now it is written only if the queue_run log
629 selector is set. In addition, the log line for abandonment now contains
630 information about the queue run such as the pid. This is always present
631 in "start" and "stop" lines but was omitted from the "abandon" line.
633 PH/34 Omit spaces between a header name and the colon in the error message that
634 is given when verify = headers_syntax fails (if there are lots of them,
635 the message gets confusing).
637 PH/35 Change the default for dns_check_names_pattern to allow slashes within
638 names, as there are now some PTR records that contain slashes. This check
639 is only to protect against broken name servers that fall over on strange
640 characters, so the fact that it applies to all lookups doesn't matter.
642 PH/36 Now that the new test suite is complete, we can remove some of the
643 special code in Exim that was needed for the old test suite. For example,
644 sorting DNS records because real resolvers return them in an arbitrary
645 order. The new test suite's fake resolver always returns records in the
648 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
649 bounces) except when queue_only is set, to avoid logging races between
650 the different processes.
652 PH/38 Panic-die if .include specifies a non-absolute path.
654 PH/39 A tweak to the "H" retry rule from its user.
656 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
657 a label. They prevented compilation on older perls.
659 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
660 a warning to be raised on newish perls.
662 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
663 on queue. Changes to match documented behaviour of showing count of
664 messages matching specified criteria.
666 PH/40 Changed the default ident timeout from 30s to 5s.
668 PH/41 Added support for the use of login_cap features, on those BSD systems
669 that have them, for controlling the resources used by pipe deliveries.
671 PH/42 The content-scanning code uses fopen() to create files in which to put
672 message data. Previously it was not paying any attention to the mode of
673 the files. Exim runs with umask(0) because the rest of the code creates
674 files with open(), and sets the required mode explicitly. Thus, these
675 files were ending up world-writeable. This was not a big issue, because,
676 being within the spool directory, they were not world-accessible. I have
677 created a function called modefopen, which takes an additional mode
678 argument. It sets umask(777), creates the file, chmods it to the required
679 mode, then resets the umask. All the relevant calls to fopen() in the
680 content scanning code have been changed to use this function.
682 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
683 to 24 hours. This avoids potential overflow problems when processing G
684 and H retry rules. I suspect nobody ever tinkers with this value.
686 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
688 PH/45 When the plaintext authenticator is running as a client, the server's
689 challenges are checked to ensure they are valid base64 strings. By
690 default, the authentication attempt is cancelled if an invalid string is
691 received. Setting client_ignore_invalid_base64 true ignores these errors.
692 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
693 they are received. Thus, the responses can be made to depend on the
694 challenges. If an invalid string is ignored, an empty string is placed in
697 PH/46 Messages that are created by the autoreply transport now contains a
698 References: header, in accordance with RFCs 2822 and 3834.
700 PH/47 Added authenticated_sender_force to the smtp transport.
702 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
704 PH/49 Installed latest patch from the Sieve maintainer.
706 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
707 was also set, the appendfile transport was unnecessarily scanning a
708 directory of message files (e.g. for maildir delivery) to find the count
709 of files (along with the size), even though it did not need this
710 information. It now does the scan only if it needs to find either the
711 size of the count of files.
713 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
715 PH/52 Two bugs concerned with error handling when the smtp transport is
718 (i) Exim was not creating retry information for temporary errors given
719 for individual recipients after the DATA command when the smtp transport
720 was used in LMTP mode. This meant that they could be retried too
721 frequently, and not timed out correctly.
723 (ii) Exim was setting the flag that allows error details to be returned
724 for LMTP errors on RCPT commands, but not for LMTP errors for individual
725 recipients that were returned after the DATA command.
727 PH/53 This is related to PH/52, but is more general: for any failing address,
728 when detailed error information was permitted to be returned to the
729 sender, but the error was temporary, then after the final timeout, only
730 "retry timeout exceeded" was returned. Now it returns the full error as
731 well as "retry timeout exceeded".
733 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
734 do this, and (what is worse) MTAs that accept it.
736 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
737 will now be deprecated.
739 PH/56 New os.c-cygwin from the Cygwin maintainer.
741 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
742 formats (previously only available in exim formats via -bpr, -bpru,
743 and -bpra. Now also available in native and exiqgrep formats)
745 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
746 with very large, slow to parse queues
748 JJ/08 exipick: added ! as generic prefix to negate any criteria format
750 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
752 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
753 responses to authentication challenges, though it was showing the
754 challenges; (ii) I've removed the CR characters from the debug output for
757 PH/58 Allow for the insertion of a newline as well as a space when a string
758 is turned into more than one encoded-word during RFC 2047 encoding. The
759 Sieve code now uses this.
761 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
762 data_4xx, lost_connection, tls_required.
764 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
765 message was being sent as an SMTP response.
767 PH/61 Add -l and -k options to exicyclog.
769 PH/62 When verifying, if an address was redirected to one new address, so that
770 verification continued, and the new address failed or deferred after
771 having set something in $address_data, the value of $address_data was not
772 passed back to the ACL. This was different to the case when no
773 redirection occurred. The value is now passed back in both cases.
775 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
776 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
777 use login_cap.h, so on its own it isn't the distinguishing feature. The
778 new name refers directly to the setclassresources() function.
780 PH/65 Added configuration files for NetBSD3.
782 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
784 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
785 is preferred over IPv4.
787 PH/68 The bounce_return_message and bounce_return_body options were not being
788 honoured for bounces generated during the reception of non-SMTP messages.
789 In particular, this applied to messages rejected by the ACL. This bug has
790 been fixed. However, if bounce_return_message is true and bounce_return_
791 body is false, the headers that are returned for a non-SMTP message
792 include only those that have been read before the error was detected.
793 (In the case of an ACL rejection, they have all been read.)
795 PH/69 The HTML version of the specification is now built in a directory called
796 spec_html instead of spec.html, because the latter looks like a path with
797 a MIME-type, and this confuses some software.
799 PH/70 Catch two compiler warnings in sieve.c.
801 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
802 function verify_get_ident() calls ip_connect() to connect a socket, but
803 if the "connect()" function timed out, ip_connect() used to close the
804 socket. However, verify_get_ident() also closes the socket later, and in
805 between Exim writes to the log, which may get opened at this point. When
806 the socket was closed in ip_connect(), the log could get the same file
807 descriptor number as the socket. This naturally causes chaos. The fix is
808 not to close the socket in ip_connect(); the socket should be closed by
809 the function that creates it. There was only one place in the code where
810 this was missing, in the iplookup router, which I don't think anybody now
811 uses, but I've fixed it anyway.
813 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
814 well as to direct DNS lookups. Otherwise the handling of names in host
815 lists is inconsistent and therefore confusing.
821 PH/01 Two changes to the default runtime configuration:
823 (1) Move the checks for relay_from_hosts and authenticated clients from
824 after to before the (commented out) DNS black list checks.
826 (2) Add control=submission to the relay_from_hosts and authenticated
827 clients checks, on the grounds that messages accepted by these
828 statements are most likely to be submissions.
830 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
832 (1) Generate an error if the third argument for the ${prvs expansion is
835 (2) Treat a missing third argument of ${prvscheck as if it were an empty
838 (3) Reset the variables that are obtained from the first argument of
839 ${prvscheck and used in the second argument before leaving the code,
840 because their memory is reclaimed, so using them afterwards may do
843 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
844 one (it's much easier than Tom thought :-).
846 (5) Because of (4), we can now allow for the use of $prvscheck_result
847 inside the third argument.
849 PH/03 For some reason, the default setting of PATH when running a command from
850 a pipe transport was just "/usr/bin". I have changed it to
853 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
854 anything to be listed in the output from -bV.
856 PH/05 When a filter generated an autoreply, the entire To: header line was
857 quoted in the delivery log line, like this:
859 => >A.N.Other <ano@some.domain> <original@ddress> ...
861 This has been changed so that it extracts the operative address. There
862 may be more than one such address. If so, they are comma-separated, like
865 => >ano@some.domain,ona@other.domain <original@ddress> ...
867 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
868 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
869 not being looked up in the rDNS to get a host name, Exim was showing the
870 IP address twice in Received: lines, even though the IP addresses were
871 identical. For example:
873 Received: from [1.2.3.4] (helo=[1.2.3.4])
875 However, if the real host name was known, it was omitting the HELO data
876 if it matched the actual IP address. This has been tidied up so that it
877 doesn't show the same IP address twice.
879 PH/07 When both +timestamp and +memory debugging was on, the value given by
880 $tod_xxx expansions could be wrong, because the tod_stamp() function was
881 called by the debug printing, thereby overwriting the timestamp buffer.
882 Debugging no longer uses the tod_stamp() function when +timestamp is set.
884 PH/08 When the original message was included in an autoreply transport, it
885 always said "this is a copy of the message, including all the headers",
886 even if body_only or headers_only was set. It now gives an appropriate
889 PH/09 Applied a patch from the Sieve maintainer which:
891 o fixes some comments
892 o adds the (disabled) notify extension core
893 o adds some debug output for the result of if/elsif tests
894 o points to the current vacation draft in the documentation
895 and documents the missing references header update
899 o fixes a bug in processing the envelope test (when testing
900 multiple envelope elements, the last element determinted the
903 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
904 Electronic Mail") by including:
906 Auto-submitted: auto-generated
908 in the messages that it generates (bounce messages and others, such as
909 warnings). In the case of bounce messages for non-SMTP mesages, there was
910 also a typo: it was using "Auto_submitted" (underscore instead of
911 hyphen). Since every message generated by Exim is necessarily in response
912 to another message, thes have all been changed to:
914 Auto-Submitted: auto-replied
916 in accordance with these statements in the RFC:
918 The auto-replied keyword:
920 - SHOULD be used on messages sent in direct response to another
921 message by an automatic process,
923 - MUST NOT be used on manually-generated messages,
925 - MAY be used on Delivery Status Notifications (DSNs) and Message
926 Disposition Notifications (MDNs),
928 - MUST NOT be used on messages generated by automatic or periodic
929 processes, except for messages which are automatic responses to
932 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
933 to the default Received: header definition.
935 PH/12 Added log selector acl_warn_skipped (default on).
937 PH/13 After a successful wildlsearch lookup, discard the values of numeric
938 variables because (a) they are in the wrong storage pool and (b) even if
939 they were copied, it wouldn't work properly because of the caching.
941 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
942 checking when decoding. Apparently there are clients that generate
943 overlong encoded strings. Why am I not surprised?
945 PH/15 If the first argument of "${if match_address" was not empty, but did not
946 contain an "@" character, Exim crashed. Now it writes a panic log message
947 and treats the condition as false.
949 PH/16 In autoreply, treat an empty string for "once" the same as unset.
951 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
952 extension "envelope-auth". The code is finished and in agreement with
953 other implementations, but there is no documentation so far and in fact,
954 nobody wrote the draft yet. This extension is currently #undef'ed, thus
955 not changing the active code.
957 Print executed "if" and "elsif" statements when debugging is used. This
958 helps a great deal to understand what a filter does.
960 Document more things not specified clearly in RFC3028. I had all this
961 sorted out, when out of a sudden new issues came to my mind. Oops."
963 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
966 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
967 canonical form (as documented). However, after a host name lookup from
968 the IP address, check_host() was doing a simple string comparison with
969 addresses acquired from the DNS when checking that the found name did
970 have the original IP as one of its addresses. Since any found IPv6
971 addresses are likely to be in abbreviated form, the comparison could
972 fail. Luckily, there already exists a function for doing the comparison
973 by converting both addresses to binary, so now that is used instead of
976 PH/20 There was another similar case to PH/19, when a complete host name was
977 given in a host list; looking up its IP address could give an abbreviated
978 form, whereas the current host's name might or might not be abbreviated.
979 The same fix has been applied.
985 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
986 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
989 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
990 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
992 PH/03 Typo: missing ".o" in src/pcre/Makefile.
994 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
995 header line, restrict the check to what is listed in RFCs 2369 and 2929.
996 Also, for "Auto-Submitted", treat anything other than "no" as
997 non-personal, in accordance with RFC 3834. (Previously it treated
998 anything starting "auto-" as non-personal.)
1000 TF/01 The control=submission/name=... option had a problem with syntax
1001 errors if the name included a slash character. The /name= option
1002 now slurps the rest of the string, so it can include any characters
1003 but it must come last in the list of options (after /sender_retain
1006 PH/05 Some modifications to the interface to the fake nameserver for the new
1014 TK/01 Added the "success_on_redirect" address verification option. See
1015 NewStuff for rationale and an example.
1017 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1019 PH/02 Patch to exigrep to allow it to work on syslog lines.
1021 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1022 fread() to skip over the body file's header line, because in Cygwin the
1023 header line is locked and is inaccessible.
1025 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1026 co-exist for some time) to make it clear that it is the Exim ID that is
1027 referenced, not the Message-ID: header line.
1029 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1030 string_format() function, because snprintf() does not exist on all
1033 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1034 ${stat: expansion item.
1036 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1037 protocol synchronization error", to keep the pedants happy.
1039 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1040 well as for IRIX systems, when gcc is being used. See the host.c source
1043 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1045 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1048 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1049 transport and to the smtp transport in LMTP mode.
1051 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1053 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1054 run a filter in a subprocess. This could lead to confusion in subsequent
1055 lookups in the parent process. There should also be a search_tidyup() at
1056 the end of the subprocess.
1058 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1059 only if the host matched helo_try_verify_hosts, which caused the
1060 verification to occur when the EHLO/HELO command was issued. The ACL just
1061 tested the remembered result. Now, if a previous verification attempt has
1062 not happened, "verify = helo" does it there and then.
1064 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1066 TK/03 Fix log output including CR from clamd.
1068 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1069 exist provoked a memory error which could cause a segfault.
1071 PH/15 Installed PCRE 6.2
1073 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1075 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1076 of the problem. Specifically, suggested +O2 rather than +O1 for the
1079 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1081 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1082 record, the port was not correctly logged when the outgoing_port log
1083 selector was set (it logged the transort's default port).
1085 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1086 fallback_hosts, and "hosts" in the smtp transport.
1088 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1089 host errors such as "Connection refused".
1091 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1092 authentication with radiusclient 0.4.9:
1094 - Error returned from rc_read_config was caught wrongly
1095 - Username/password not passed on to radius server due to wrong length.
1097 The presumption is that some radiusclient API changes for 4.51/PH/17
1098 were not taken care of correctly. The code is still untested by me (my
1099 Linux distribution still has 0.3.2 of radiusclient), but it was
1100 contributed by a Radius user.
1102 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1103 expanding the "port" option of the smtp transport.
1105 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1106 while reading a MIME header. Thanks to Tom Hughes for a patch.
1108 PH/24 Include config.h inside local_scan.h so that configuration settings are
1111 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1112 that in an ACL for RCPT (for example), you can examine exactly what was
1115 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1116 commands, but it was not correctly comparing the address with the actual
1117 client host address. Thus, it would show the EHLO address in Received:
1118 header lines when this was not necessary.
1120 PH/27 Added the % operator to ${eval:}.
1122 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1123 it should be ignoring failures (because with -C, for example, it has lost
1124 privilege). It wasn't ignoring creation failures other than "already
1127 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1130 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1131 because an input line was too long, either on its own, or by virtue of
1132 too many continuations, the temporary file was not being removed, and the
1133 return code was incorrect.
1135 PH/31 Missing "BOOL" in function definition in filtertest.c.
1137 PH/32 Applied Sieve patches from the maintainer.
1139 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1141 PH/33 Added "verify = not_blind".
1143 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1144 Local/Makefile (with some defaults set). These are used in built scripts
1145 such as exicyclog, but they have never been used in the exim_install
1146 script (though there are many overriding facilities there). I have
1147 arranged that the exim_install script now takes note of these two
1150 PH/35 Installed configuration files for Dragonfly.
1152 PH/36 When a locally submitted message by a trusted user did not contain a
1153 From: header, and the sender address was obtained from -f or from an SMTP
1154 MAIL command, and the trusted user did not use -F to supply a sender
1155 name, $originator_name was incorrectly used when constructing a From:
1156 header. Furthermore, $originator_name was used for submission mode
1157 messages from external hosts without From: headers in a similar way,
1158 which is clearly wrong.
1160 PH/37 Added control=suppress_local_fixups.
1162 PH/38 When log_selector = +received_sender was set, and the addition of the
1163 sender made the log line's construction buffer exactly full, or one byte
1164 less than full, an overflow happened when the terminating "\n" was
1167 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1168 when the result of a list match is failure because a DNS lookup failed.
1170 PH/40 RM_COMMAND is now used in the building process.
1172 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1173 the "build-* directories that it finds.
1175 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1176 address in a domain literal was a prefix of an interface address.
1178 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1179 when verifying a sender address, unless rewrite_headers is false.
1181 PH/44 Wrote a long comment about why errors_to addresses are verified as
1182 recipients, not senders.
1184 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1185 the ratelimit ACL was added.
1187 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1189 PH/46 Added extra information about PostgreSQL errors to the error string.
1191 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1192 avoiding the need to install special zones in a real server. This is
1193 backwards compatible; if it can't find the fake resolver, it drops back.
1194 Thus, both old and new test suites can be run.
1196 TF/02 Added util/ratelimit.pl
1198 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1199 clock is set back in time.
1201 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1202 Candler <B.Candler@pobox.com>.
1204 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1205 OK for addresses that are the result of redirections.
1207 PH/48 A number of further additions for the benefit of the new test suite,
1208 including a fake gethostbyname() that interfaces to the fake DNS resolver
1211 TF/06 The fix for widen_domains has also been applied to qualify_single and
1212 search_parents which are the other dnslookup options that can cause
1215 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1218 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1220 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1221 service" if it encountered an MX record with an empty target hostname.
1222 The message is now "an MX or SRV record indicated no SMTP service".
1224 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1225 if the DNS of the sending site is misconfigured. This is quite a
1226 common situation. This change restores the behaviour of treating a
1227 helo verification defer as a failure.
1229 PH/51 If self=fail was set on a router, the bounce message did not include the
1230 actual error message.
1236 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1238 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1239 command itself ended in error, the underlying message about the transport
1240 filter timeout was being overwritten with the pipe command error. Now the
1241 underlying error message should be appended to the second error message.
1243 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1245 PH/02 Reduce the amount of output that "make" produces by default. Full output
1246 can still be requested.
1248 PH/03 The warning log line about a condition test deferring for a "warn" verb
1249 was being output only once per connection, rather than after each
1250 occurrence (because it was using the same function as for successful
1251 "warn" verbs). This seems wrong, so I have changed it.
1253 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1254 it should not have, which might have caused a crash in the right
1255 circumstances, but probably never did.
1257 PH/04 Installed a modified version of Tony Finch's patch to make submission
1258 mode fix the return path as well as the Sender: header line, and to
1259 add a /name= option so that you can make the user's friendly name appear
1262 TF/03 Added the control = fakedefer ACL modifier.
1264 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1265 Mark Lowes for thorough testing.
1267 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1269 TK/03 Merged latest SRS patch from Miles Wilton.
1271 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1272 with the definition in sysexits.h (which is #included earlier).
1273 Fortunately, Exim does not actually use EX_OK. The code used to try to
1274 preserve the sysexits.h value, by assumimg that macro definitions were
1275 scanned for macro replacements. I have been disabused of this notion,
1276 so now the code just undefines EX_OK before #including unistd.h.
1278 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1279 in the smtp transport. When a block could not be written in a single
1280 write() function, the timeout was being re-applied to each part-write.
1281 This seems wrong - if the receiver was accepting one byte at a time it
1282 would take for ever. The timeout is now adjusted when this happens. It
1283 doesn't have to be particularly precise.
1285 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1286 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1288 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1289 without a domain if the check to <postmaster@domain> fails.
1291 SC/01 Eximstats: added -xls and the ability to specify output files
1292 (patch written by Frank Heydlauf).
1294 SC/02 Eximstats: use FileHandles for outputing results.
1296 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1298 SC/04 Eximstats: fixed display of large numbers with -nvr option
1300 SC/05 Eximstats: fixed merging of reports with empty tables.
1302 SC/06 Eximstats: added the -include_original_destination flag
1304 SC/07 Eximstats: removed tabs and trailing whitespace.
1306 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1308 TK/06 MBOX spool code: Add real "From " MBOX separator line
1309 so the .eml file is really in mbox format (even though
1310 most programs do not really care). Patch from Alex Miller.
1312 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1313 The latter is generated from $received_to and is only set if the
1314 message has one envelope recipient. SA can use these headers,
1315 obviously out-of-the-box. Patch from Alex Miller.
1317 PH/08 The ${def test on a variable was returning false if the variable's
1318 value was "0", contrary to what the specification has always said!
1319 The result should be true unless the variable is empty.
1321 PH/09 The syntax error of a character other than { following "${if
1322 def:variable_name" (after optional whitespace) was not being diagnosed.
1323 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1324 accidental colon was present, for example, could give incorrect results.
1326 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1327 result is used (not including appendfile, where other changes are about
1330 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1331 This involved changing a lot of size variables from int to off_t. It
1332 should work with maildirs and everything.
1334 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1335 spamd dying while we are connected to it.
1337 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1340 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1341 function for BDB 4.3.
1343 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1345 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1346 the sources for PCRE have been re-arranged and more files are now
1349 PH/15 The code I had for printing potentially long long variables in PH/11
1350 above was not the best (it lost precision). The length of off_t variables
1351 is now inspected at build time, and an appropriate printing format (%ld
1352 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1353 to be "long long int" or "long int". This is needed for the internal
1354 formatting function string_vformat().
1356 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1357 the configuration file to be ":syslog", then the script "guesses" where
1358 the logs files are, rather than using the compiled in default. In our
1359 case the guess is not the same as the compiled default, so the script
1360 suddenly stopped working when I started to use syslog. The patch checks
1361 to see if log_file_path is "". If so, it attempts to read it from exim
1362 with no configuration file to get the compiled in version, before it
1363 falls back to the previous guessing code."
1365 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1366 implementing BATV in an Exim configuration. See NewStuff for the gory
1369 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1370 Makefile that are specific to HP-UX.
1372 PH/18 If the "use_postmaster" option was set for a recipient callout together
1373 with the "random" option, the postmaster address was used as the MAIL
1374 FROM address for the random test, but not for the subsequent recipient
1375 test. It is now used for both.
1377 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1378 patch removes a few documentation additions to RFC 3028, because the
1379 latest draft now contains them. It adds the new en;ascii-case comparator
1380 and a new error check for 8bit text in MIME parts. Comparator and
1381 require names are now matched exactly. I enabled the subaddress
1382 extension, but it is not well tested yet (read: it works for me)."
1384 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1385 rework some of the code of TK/09 above to avoid the hardwired use of
1386 "%lld" and "long long". Replaced the call to snprintf() with a call to
1389 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1390 records point to non-existent hosts", "retry timeout exceeded", and
1391 "retry time not reached for any host after a long failure period".
1393 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1394 experimental DomainKeys support:
1396 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1397 (2) On an error such as an illegally used "control", the wrong name for
1398 the control was given.
1400 These problems did NOT occur unless DomainKeys support was compiled.
1402 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1404 PH/24 Added ${if match_ip condition.
1406 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1407 databases so that it will be absolutely obvious if a crash occurs in the
1408 DB library. This is a regular occurrence (often caused by mis-matched
1411 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1412 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1413 picked up on a user's system that detects such things. There doesn't seem
1414 to be a gcc warning option for this - only an attribute that has to be
1415 put on the function's prototype. It seems that in Fedora Core 4 they have
1416 set this on a number of new functions. No doubt there will be more in due
1419 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1420 specify a transport. However, if an address that was verified by such a
1421 router was the subject of a callout, Exim crashed because it tried to
1422 read the rcpt_include_affixes from the non-existent transport. Now it
1423 just assumes that the setting of that option is false. This bug was
1424 introduced by 4.51/PH/31.
1426 PH/28 Changed -d+all to exclude +memory, because that information is very
1427 rarely of interest, but it makes the output a lot bigger. People tend to
1428 do -d+all out of habit.
1430 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1431 code in os-type was giving problems when libc.so lives in lib64, like on
1434 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1435 aren't the modern standard, and it seems that some systems' include files
1436 don't always have them. Exim was already checking for some of the newer
1437 ones like T_AAAA, and defining it itself. I've added checks for all the
1438 record types that Exim uses.
1440 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1441 not automatically generating a new one, as it is supposed to. This
1442 prevented TLS from working. If the file did exist, but contained invalid
1443 data, a new version was generated, as expected. It was only the case of a
1444 non-existent file that was broken.
1446 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1447 with a change in libdomainkeys > 0.64.
1449 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1450 from DNS. If the selector record carries the flag, it now has
1451 precedence over the domain-wide flag.
1453 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1455 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1456 the use of a port name that isn't defined in /etc/services) occurred, the
1457 message was deferred as in a normal delivery, and thus remained on the
1458 spool, instead of being failed because of the mua_wrapper setting. This
1459 is now fixed, and I tidied up some of the mua_wrapper messages at the
1462 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1463 the messages in a hash of arrays rather than using individual hashes.
1464 This is a bit cleaner and results in dramatic memory savings, albeit
1465 at a slight CPU cost.
1467 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1468 as requested by Marc Sherman.
1470 SC/10 Eximstats: added histograms for user specified patterns as requested
1473 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1475 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1476 fopen() in the content-scanning modules that did not already have it.
1482 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1483 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1485 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1487 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1489 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1490 to test Sieve filters that use "vacation".
1492 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1493 that changes the way the GnuTLS parameters are stored in the cache file.
1494 The new format can be generated externally. For backward compatibility,
1495 if the data in the cache doesn't make sense, Exim assumes it has read an
1496 old-format file, and it generates new data and writes a new file. This
1497 means that you can't go back to an older release without removing the
1500 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1501 work if there are any delivery delays because "one_time" forces the
1502 parent to be marked "delivered", so its unseen clone is never tried
1503 again. For this reason, Exim now forbids the simultaneous setting of
1506 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1507 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1508 entry if you want to know the details. The fix, however, neglected to
1509 consider the case when local delivery batching is involved. The test for
1510 "previously delivered" was not happening when checking to see if an
1511 address could be batched with a previous (undelivered) one; under
1512 certain circumstances this could lead to multiple deliveries to the same
1515 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1516 in its include files, and this causes problems building Exim.
1518 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1519 header_syntax) but Exim was just ignoring anything given after a slash.
1520 In particular, this caused confusion with an attempt to use "verify =
1521 reverse_host_lookup/defer_ok". An error is now given when options are
1522 supplied for verify items that do not have them. (Maybe reverse_host_
1523 lookup should have a defer_ok option, but that's a different point.)
1525 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1526 defined by RFC 821) to 2048, because there were problems with some AUTH
1527 commands, and RFC 1869 says the size should be increased for extended
1528 SMTP commands that take arguments.
1530 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1533 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1534 "unknown" error; now it says that the functionality isn't in the binary.
1536 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1537 an address' error message when a string expansion fails (syntax or
1538 whatever). Otherwise the password may appear in the log. Following change
1539 PH/42 below, there is no longer a chance of it appearing in a bounce
1542 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1544 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1545 of its addresses was ever tried. (Bugzilla bug #2.)
1547 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1548 the result incorrectly in the debug output. (It correctly added a newline
1549 to what was transported.)
1551 TF/01 Added $received_time.
1553 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1554 commented out examples of how to interface to a virus scanner and to
1555 SpamAssassin. Also added commented examples of av_scanner and
1556 spamd_address settings.
1558 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1559 and controls are allowed in which ACLs. There were a couple of minor
1560 errors. Some of the entries in the conditions table (which is a table of
1561 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1562 negation of where the condition IS allowed.
1564 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1566 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1567 header file does not have a version number, so I've had to invent a new
1568 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1569 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1570 radiusclient), but it was contributed by a Radius user.
1572 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1573 files or directories, for OpenSSL.
1575 PH/19 When an Exim process that is running as root has to create an Exim log
1576 file, it does so in a subprocess that runs as exim:exim so as to get the
1577 ownership right at creation (otherwise, other Exim processes might see
1578 the file with the wrong ownership). There was no test for failure of this
1579 fork() call, which would lead to the process getting stuck as it waited
1580 for a non-existent subprocess. Forks do occasionally fail when resources
1581 run out. I reviewed all the other calls to fork(); they all seem to check
1584 PH/20 When checking for unexpected SMTP input at connect time (before writing
1585 the banner), Exim was not dealing correctly with a non-positive return
1586 from the read() function. If the client had disconnected by this time,
1587 the result was a log entry for a synchronization error with an empty
1588 string after "input=" when read() returned zero. If read() returned -1
1589 (an event I could not check), uninitialized data bytes were printed.
1590 There were reports of junk text (parts of files, etc) appearing after
1593 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1595 PH/22 Added support for macro redefinition, and (re)definition in between
1596 driver and ACL definitions.
1598 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1599 forgetting to use the resulting value; it was using the unexpanded value.
1601 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1602 hadn't been configured. The fix is from Juergen Kreileder, who
1603 understands it better than I do:
1605 "Here's what I see happening with three configured cyrus_sasl
1606 authenticators configured (plain, login, cram-md5):
1608 On startup auth_cyrus_sasl_init() gets called for each of these.
1609 This means three calls to sasl_listmech() without a specified mech_list.
1610 => SASL tests which mechs of all available mechs actually work
1611 => three warnings about OTP not working
1612 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1614 With the patch, sasl_listmech() also gets called three times. But now
1615 SASL's mech_list option is set to the server_mech specified in the the
1616 authenticator. Or in other words, the answer from sasl_listmech()
1617 gets limited to just the mech you're testing for (which is different
1619 => the return list contains just 'plain' or 'login', 'cram-md5' or
1620 nothing depending on the value of ob->server_mech.
1622 I've just tested the patch: Authentication still works fine,
1623 unavailable mechs specified in the exim configuration are still
1624 caught, and the auth.log warnings about OTP are gone."
1626 PH/25 When debugging is enabled, the contents of the command line are added
1627 to the debugging output, even when log_selector=+arguments is not
1630 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1631 answer is "GNU", and only if the return is "GNU/something" is the answer
1634 PH/27 $acl_verify_message is now set immediately after the failure of a
1635 verification in an ACL, and so is available in subsequent modifiers. In
1636 particular, the message can be preserved by coding like this:
1638 warn !verify = sender
1639 set acl_m0 = $acl_verify_message
1641 Previously, $acl_verify_message was set only while expanding "message"
1642 and "log_message" when a very denied access.
1644 PH/28 Modified OS/os.c-Linux with
1646 -#ifndef OS_LOAD_AVERAGE
1647 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1649 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1650 nomenclature these days.)
1652 PH/29 Installed patch from the Sieve maintainer that adds the options
1653 sieve_useraddress and sieve_subaddress to the redirect router.
1655 PH/30 In these circumstances:
1656 . Two addresses routed to the same list of hosts;
1657 . First host does not offer TLS;
1658 . First host accepts first address;
1659 . First host gives temporary error to second address;
1660 . Second host offers TLS and a TLS session is established;
1661 . Second host accepts second address.
1662 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1663 and peerdn, if requested) that were in fact used only for the second
1666 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1667 attention to any local part prefix or suffix that was matched by the
1668 router that accepted the address. It now behaves in the same way as it
1669 does for delivery: the affixes are removed from the local part unless
1670 rcpt_include_affixes is set on the transport.
1672 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1673 timeout during the DATA phase of an incoming message.
1675 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1676 applied a patch sanctioned by the Sieve maintainer.
1678 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1679 the uid or gid is negative. A case of a negative gid caused this to be
1680 noticed. The fix allows for either to be negative.
1682 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1683 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1686 PH/36 The change PH/12 above was broken. Fixed it.
1688 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1689 the grounds that routing the same address twice would always produce the
1690 same answer. This might have been true once, but it is certainly no
1691 longer true now. Routing a child address may depend on the previous
1692 routing that produced that child. Some complicated redirection strategies
1693 went wrong when messages had multiple recipients, and made Exim's
1694 behaviour dependent on the order in which the addresses were given.
1696 I have moved the duplicate checking until after the routing is complete.
1697 Exim scans the addresses that are assigned to local and remote
1698 transports, and removes any duplicates. This means that more work will be
1699 done, as duplicates will always all be routed, but duplicates are
1700 presumably rare, so I don't expect this is of any significance.
1702 For deliveries to pipes, files, and autoreplies, the duplicate checking
1703 still happens during the routing process, since they are not going to be
1706 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1707 It corrects a timeout issue with spamd. This is Ian's comment: "The
1708 background is that sometimes spamd either never reads data from a
1709 connection it has accepted, or it never writes response data. The exiscan
1710 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1711 blindly assumes that writes won't block so it may never time out."
1713 PH/39 Allow G after quota size as well as K and M.
1715 PH/40 The value set for $authenticated_id in an authenticator may not contain
1716 binary zeroes or newlines because the value is written to log lines and
1717 to spool files. There was no check on this. Now the value is run through
1718 the string_printing() function so that such characters are converted to
1719 printable escape sequences.
1721 PH/41 $message_linecount is a new variable that contains the total number of
1722 lines in the message. Compare $body_linecount, which is the count for the
1725 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1726 bounce and delay warning messages, except in certain special cases, which
1729 (a) An SMTP error message from a remote host;
1730 (b) A message specified in a :fail: redirection;
1731 (c) A message specified in a "fail" command in a system filter;
1732 (d) A message specified in a FAIL return from the queryprogram router;
1733 (e) A message specified by the cannot_route_message router option.
1735 In these cases only, Exim does include the error details in bounce and
1736 warning messages. There are also a few cases where bland messages such
1737 as "unrouteable address" or "local delivery error" are given.
1739 PH/43 $value is now also set for the "else" part of a ${run expansion.
1741 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1742 being worked on, but at least Exim now implements the latest version to
1745 PH/45 In a pipe transport, although a timeout while waiting for the pipe
1746 process to complete was treated as a delivery failure, a timeout while
1747 writing the message to the pipe was logged, but erroneously treated as a
1748 successful delivery. Such timeouts include transport filter timeouts. For
1749 consistency with the overall process timeout, these timeouts are now
1750 treated as errors, giving rise to delivery failures by default. However,
1751 there is now a new Boolean option for the pipe transport called
1752 timeout_defer, which, if set TRUE, converts the failures into defers for
1753 both kinds of timeout. A transport filter timeout is now identified in
1756 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1757 systems where "make" and "gmake" are different, calling "gmake" at top
1758 level broke things. I've arranged for the value of $(MAKE) to be passed
1759 from the Makefile to this script so that it can call the same version of
1763 A note about Exim versions 4.44 and 4.50
1764 ----------------------------------------
1766 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1767 changes of various kinds. As a consequence, a big documentation update was
1768 needed. This delayed the release for rather longer than seemed good, especially
1769 in the light of a couple of (minor) security issues. Therefore, the changes
1770 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1771 release. So 4.44 and 4.50 are in effect two different branches that both start
1774 I have left the 4.50 change log unchanged; it contains all the changes since
1775 4.43. The change log for 4.44 is below; many of its items are identical to
1776 those for 4.50. This seems to be the most sensible way to preserve the
1777 historical information.
1783 1. Minor wording change to the doc/README.SIEVE file.
1785 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1786 computation of the current number of files was incorrect.
1788 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1789 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1792 4. Give more explanation in the error message when the command for a transport
1793 filter fails to execute.
1795 5. There are several places where Exim runs a non-Exim command in a
1796 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1797 was being done only for the command run by the queryprogram router. It is
1798 now done for all such subprocesses. The other cases are: ${run, transport
1799 filters, and the commands run by the lmtp and pipe transports.
1801 6. Added CONFIGURE_GROUP build-time option.
1803 7. Some older OS have a limit of 256 on the maximum number of file
1804 descriptors. Exim was using setrlimit() to set 1000 as a large value
1805 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1806 systems. I've change it so that if it can't get 1000, it tries for 256.
1808 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1809 was an oversight, and furthermore, ever since the addition of extra
1810 controls (e.g. 4.43/32), the checks on when to allow different forms of
1811 "control" were broken. There should now be diagnostics for all cases when a
1812 control that does not make sense is encountered.
1814 9. Added the /retain_sender option to "control=submission".
1816 10. $recipients is now available in the predata ACL (oversight).
1818 11. Tidy the search cache before the fork to do a delivery from a message
1819 received from the command line. Otherwise the child will trigger a lookup
1820 failure and thereby defer the delivery if it tries to use (for example) a
1821 cached ldap connection that the parent has called unbind on.
1823 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1824 of $address_data from the recipient verification was clobbered by the
1825 sender verification.
1827 13. The value of address_data from a sender verification is now available in
1828 $sender_address_data in subsequent conditions in the ACL statement.
1830 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1832 15. Added a new option "connect=<time>" to callout options, to set a different
1835 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1836 was its contents. (It was OK if the option was not defined at all.)
1838 17. A "Completed" log line is now written for messages that are removed from
1839 the spool by the -Mrm option.
1841 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1842 information about exactly what failed.
1844 19. Added -dd to debug only the daemon process.
1846 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1847 handles timeouts, both on the server side and network timeouts. Renamed the
1848 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1850 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1851 It is now set to "smtps".
1853 22. $host_address is now set to the target address during the checking of
1854 ignore_target_hosts.
1856 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1857 being passed; this would have caused $sender_host_name to have been used if
1858 matching the list had actually called for a host name (not very likely,
1859 since this list is usually IP addresses). A host name is now passed as
1862 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1863 code that specifies a non-restarting handler (typically sigaction() in
1864 modern systems) in an attempt to fix a rare and obscure crash bug.
1866 25. Narrowed the window for a race in the daemon that could cause it to ignore
1867 SIGCHLD signals. This is not a major problem, because they are used only to
1868 wake it up if nothing else does.
1870 26. A malformed maildirsize file could cause Exim to calculate negative values
1871 for the mailbox size or file count. Odd effects could occur as a result.
1872 The maildirsize information is now recalculated if the size or filecount
1875 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1876 support for a long time. Removed HAVE_SYS_VFS_H.
1878 28. Installed the latest version of exipick from John Jetmore.
1880 29. In an address list, if the pattern was not a regular expression, an empty
1881 subject address (from a bounce message) matched only if the pattern was an
1882 empty string. Non-empty patterns were not even tested. This was the wrong
1883 because it is perfectly reasonable to use an empty address as part of a
1884 database query. An empty address is now tested by patterns that are
1885 lookups. However, all the other forms of pattern expect the subject to
1886 contain a local part and a domain, and therefore, for them, an empty
1887 address still always fails if the pattern is not itself empty.
1889 30. Exim went into a mad DNS loop when attempting to do a callout where the
1890 host was specified on an smtp transport, and looking it up yielded more
1891 than one IP address.
1893 31. Re-factored the code for checking spool and log partition space into a
1894 function that finds that data and another that does the check. The former
1895 is then used to implement four new variables: $spool_space, $log_space,
1896 $spool_inodes, and $log_inodes.
1898 32. The RFC2047 encoding function was originally intended for short strings
1899 such as real names; it was not keeping to the 75-character limit for
1900 encoded words that the RFC imposes. It now respects the limit, and
1901 generates multiple encoded words if necessary. To be on the safe side, I
1902 have increased the buffer size for the ${rfc2047: expansion operator from
1905 33. It is now permitted to omit both strings after an "if" condition; if the
1906 condition is true, the result is "true". As before, when the second string
1907 is omitted, a false condition yields an empty string. This makes it less
1908 cumbersome to write custom ACL and router conditions.
1910 34. Failure to deliver a bounce message always caused it to be frozen, even if
1911 there was an errors_to setting on the router. The errors_to setting is now
1914 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1915 canonical form (fully expanded) before being placed in
1916 $sender_host_address.
1918 36. The table in the code that translates DNS record types into text (T_A to
1919 "A" for instance) was missing entries for NS and CNAME. It is just possible
1920 that this could have caused confusion if both these types were looked up
1921 for the same domain, because the text type is used as part of Exim's
1922 per-process caching. But the chance of anyone hitting this buglet seems
1925 37. The dnsdb lookup has been extended in a number of ways.
1927 (1) There is a new type, "zns", which walks up the domain tree until it
1928 finds some nameserver records. It should be used with care.
1930 (2) There is a new type, "mxh", which is like "mx" except that it returns
1931 just the host names, not the priorities.
1933 (3) It is now possible to give a list of domains (or IP addresses) to be
1934 looked up. The behaviour when one of the lookups defers can be
1935 controlled by a keyword.
1937 (4) It is now possible to specify the separator character for use when
1938 multiple records are returned.
1940 38. The dnslists ACL condition has been extended: it is now possible to supply
1941 a list of IP addresses and/or domains to be looked up in a particular DNS
1944 39. Added log_selector=+queue_time_overall.
1946 40. When running the queue in the test harness, wait just a tad after forking a
1947 delivery process, to get repeatability of debugging output.
1949 41. Include certificate and key file names in error message when GnuTLS fails
1950 to set them up, because the GnuTLS error message doesn't include the name
1951 of the failing file when there is a problem reading it.
1953 42. Allow both -bf and -bF in the same test run.
1955 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1957 44. The "Exiscan patch" is now merged into the mainline Exim source.
1959 45. Sometimes the final signoff response after QUIT could fail to get
1960 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1961 < 0 before doing a fflush(). This bug looks as though it goes back to the
1962 introduction of TLS in release 3.20, but "sometimes" must have been rare
1963 because the tests only now provoked it.
1965 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1966 (this can affect the format of dates).
1968 47. exim_tidydb, when checking for the continued existence of a message for
1969 which it has found a message-specific retry record, was not finding
1970 messages that were in split spool directories. Consequently, it was
1971 deleting retry records that should have stayed in existence.
1973 48. Steve fixed some bugs in eximstats.
1975 49. The SPA authentication driver was not abandoning authentication and moving
1976 on to the next authenticator when an expansion was forced to fail,
1977 contradicting the general specification for all authenticators. Instead it
1978 was generating a temporary error. It now behaves as specified.
1980 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1981 (the order specifies the preference for clients). The order is now AES256,
1982 AES128, 3DES, ARCFOUR128.
1984 51. Small patch to Sieve code - explicitly set From: when generating an
1987 52. Exim crashed if a remote delivery caused a very long error message to be
1988 recorded - for instance if somebody sent an entire SpamAssassin report back
1989 as a large number of 550 error lines. This bug was coincidentally fixed by
1990 increasing the size of one of Exim's internal buffers (big_buffer) that
1991 happened as part of the Exiscan merge. However, to be on the safe side, I
1992 have made the code more robust (and fixed the comments that describe what
1995 53. Now that there can be additional text after "Completed" in log lines (if
1996 the queue_time_overall log selector is set), a one-byte patch to exigrep
1997 was needed to allow it to recognize "Completed" as not the last thing in
2000 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2001 patch that reportedly fixes this has been added. I am not expert enough to
2002 create a test for it. This is what the patch creator wrote:
2004 "I found a little strange behaviour of ldap code when working with
2005 Windows 2003 AD Domain, where users was placed in more than one
2006 Organization Units. When I tried to give exim partial DN, the exit code
2007 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2008 But simultaneously result of request was absolutely normal ldap result,
2009 so I produce this patch..."
2011 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2012 REFERENCE, so I have modified the code to exclude the patch when that macro
2015 55. Some experimental protocols are using DNS PTR records for new purposes. The
2016 keys for these records are domain names, not reversed IP addresses. The
2017 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2018 leaves it alone. Component reversal etc. now happens only for IP addresses.
2021 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2023 57. Double the size of the debug message buffer (to 2048) so that more of very
2024 long debug lines gets shown.
2026 58. The exicyclog utility now does better if the number of log files to keep
2027 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2029 59. Two changes related to the smtp_active_hostname option:
2031 (1) $smtp_active_hostname is now available as a variable.
2032 (2) The default for smtp_banner uses $smtp_active_hostname instead
2033 of $primary_hostname.
2035 60. The host_aton() function is supposed to be passed a string that is known
2036 to be a valid IP address. However, in the case of IPv6 addresses, it was
2037 not checking this. This is a hostage to fortune. Exim now panics and dies
2038 if the condition is not met. A case was found where this could be provoked
2039 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2040 components; fortuitously, this particular loophole had already been fixed
2041 by change 4.50/55 above.
2043 If there are any other similar loopholes, the new check in host_aton()
2044 itself should stop them being exploited. The report I received stated that
2045 data on the command line could provoke the exploit when Exim was running as
2046 exim, but did not say which command line option was involved. All I could
2047 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2048 running as the user.
2051 61. There was a buffer overflow vulnerability in the SPA authentication code
2052 (which came originally from the Samba project). I have added a test to the
2053 spa_base64_to_bits() function which I hope fixes it.
2056 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2057 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2059 63. The daemon start-up calls getloadavg() while still root for those OS that
2060 need the first call to be done as root, but it missed one case: when
2061 deliver_queue_load_max is set with deliver_drop_privilege. This is
2062 necessary for the benefit of the queue runner, because there is no re-exec
2063 when deliver_drop_privilege is set.
2065 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2066 This has been fixed.
2068 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2069 was in use, was not putting the data itself into the right store pool;
2070 consequently, it could be overwritten for a subsequent message in the same
2071 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2074 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2076 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2077 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2078 the function were treating the return as a boolean value, which happened to
2079 work because 0=false and not-0=true, but is not correct code.
2081 68. The host_aton() function was not handling scoped IPv6 addresses (those
2082 with, for example, "%eth0" on the end) correctly.
2084 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2085 negated items (that is, ~something) in unsigned ints. Some compilers
2086 apparently mutter when there is no cast.
2088 70. If an address verification called from an ACL failed, and did not produce a
2089 user-specific message (i.e. there was only a "system" message), nothing was
2090 put in $acl_verify_message. In this situation, it now puts the system
2093 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2094 session; change 4.31/43 added the unwanted input to the log line - except
2095 that it did not do this in the start of session case. It now does.
2097 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2098 This is wrong and can cause the other end to generate a synchronization
2099 error if it is another Exim or anything else that does the synchronization
2100 check. A QUIT command is no longer sent after a timeout.
2102 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2103 during host lookups.
2105 74. The defer_ok option of callout verification was not working if it was used
2106 when verifying addresses in header lines, that is, for this case:
2108 verify = header_sender/callout=defer_ok
2110 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2111 those file descriptors could be used for SMTP connections. If anything
2112 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2113 could be sent to the SMTP client, causing chaos. The daemon now opens
2114 stdin, stdout, and stderr to /dev/null when it puts itself into the
2117 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2118 log by default. The user can override this with suitable Perl magic.
2120 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2121 the log message when discard triggers, was not working for the DATA ACL or
2122 for the non-SMTP ACL.
2124 78. Error message wording change in sieve.c.
2126 79. If smtp_accept_max_per_host was set, the number of connections could be
2127 restricted to fewer than expected, because the daemon was trying to set up
2128 a new connection before checking whether the processes handling previous
2129 connections had finished. The check for completed processes is now done
2130 earlier. On busy systems, this bug wouldn't be noticed because something
2131 else would have woken the daemon, and it would have reaped the completed
2134 80. If a message was submitted locally by a user whose login name contained one
2135 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2136 It caused a spool format error. I have fixed the spool reading code. A
2137 related problem was that the "from" clause in the Received: line became
2138 illegal because of the space(s). It is now covered by ${quote_local_part.
2140 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2143 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2145 83. Patch from Sieve maintainer to fix unterminated string problem in
2146 "vacation" handling.
2148 84. Some minor changes to the Linux configuration files to help with other
2149 OS variants using glibc.
2151 85. One more patch for Sieve to update vacation handling to latest spec.
2154 ----------------------------------------------------
2155 See the note above about the 4.44 and 4.50 releases.
2156 ----------------------------------------------------
2162 1. Change 4.43/35 introduced a bug that caused file counts to be
2163 incorrectly computed when quota_filecount was set in an appendfile
2166 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2167 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2170 3. Give more explanation in the error message when the command for a transport
2171 filter fails to execute.
2173 4. There are several places where Exim runs a non-Exim command in a
2174 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2175 was being done only for the command run by the queryprogram router. It is
2176 now done for all such subprocesses. The other cases are: ${run, transport
2177 filters, and the commands run by the lmtp and pipe transports.
2179 5. Some older OS have a limit of 256 on the maximum number of file
2180 descriptors. Exim was using setrlimit() to set 1000 as a large value
2181 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2182 systems. I've change it so that if it can't get 1000, it tries for 256.
2184 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2185 was an oversight, and furthermore, ever since the addition of extra
2186 controls (e.g. 4.43/32), the checks on when to allow different forms of
2187 "control" were broken. There should now be diagnostics for all cases when a
2188 control that does not make sense is encountered.
2190 7. $recipients is now available in the predata ACL (oversight).
2192 8. Tidy the search cache before the fork to do a delivery from a message
2193 received from the command line. Otherwise the child will trigger a lookup
2194 failure and thereby defer the delivery if it tries to use (for example) a
2195 cached ldap connection that the parent has called unbind on.
2197 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2198 of $address_data from the recipient verification was clobbered by the
2199 sender verification.
2201 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2202 was its contents. (It was OK if the option was not defined at all.)
2204 11. A "Completed" log line is now written for messages that are removed from
2205 the spool by the -Mrm option.
2207 12. $host_address is now set to the target address during the checking of
2208 ignore_target_hosts.
2210 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2211 being passed; this would have caused $sender_host_name to have been used if
2212 matching the list had actually called for a host name (not very likely,
2213 since this list is usually IP addresses). A host name is now passed as
2216 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2217 code that specifies a non-restarting handler (typically sigaction() in
2218 modern systems) in an attempt to fix a rare and obscure crash bug.
2220 15. Narrowed the window for a race in the daemon that could cause it to ignore
2221 SIGCHLD signals. This is not a major problem, because they are used only to
2222 wake it up if nothing else does.
2224 16. A malformed maildirsize file could cause Exim to calculate negative values
2225 for the mailbox size or file count. Odd effects could occur as a result.
2226 The maildirsize information is now recalculated if the size or filecount
2229 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2230 support for a long time. Removed HAVE_SYS_VFS_H.
2232 18. Updated exipick to current release from John Jetmore.
2234 19. Allow an empty sender to be matched against a lookup in an address list.
2235 Previously the only cases considered were a regular expression, or an
2238 20. Exim went into a mad DNS lookup loop when doing a callout where the
2239 host was specified on the transport, if the DNS lookup yielded more than
2242 21. The RFC2047 encoding function was originally intended for short strings
2243 such as real names; it was not keeping to the 75-character limit for
2244 encoded words that the RFC imposes. It now respects the limit, and
2245 generates multiple encoded words if necessary. To be on the safe side, I
2246 have increased the buffer size for the ${rfc2047: expansion operator from
2249 22. Failure to deliver a bounce message always caused it to be frozen, even if
2250 there was an errors_to setting on the router. The errors_to setting is now
2253 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2254 canonical form (fully expanded) before being placed in
2255 $sender_host_address.
2257 24. Updated eximstats to version 1.33
2259 25. Include certificate and key file names in error message when GnuTLS fails
2260 to set them up, because the GnuTLS error message doesn't include the name
2261 of the failing file when there is a problem reading it.
2263 26. Expand error message when OpenSSL has problems setting up cert/key files.
2266 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2267 (this can affect the format of dates).
2269 28. exim_tidydb, when checking for the continued existence of a message for
2270 which it has found a message-specific retry record, was not finding
2271 messages that were in split spool directories. Consequently, it was
2272 deleting retry records that should have stayed in existence.
2274 29. eximstats updated to version 1.35
2275 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2276 1.35 - bugfix such that pie charts by volume are generated correctly
2278 30. The SPA authentication driver was not abandoning authentication and moving
2279 on to the next authenticator when an expansion was forced to fail,
2280 contradicting the general specification for all authenticators. Instead it
2281 was generating a temporary error. It now behaves as specified.
2283 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2284 (the order specifies the preference for clients). The order is now AES256,
2285 AES128, 3DES, ARCFOUR128.
2287 31. Small patch to Sieve code - explicitly set From: when generating an
2290 32. Exim crashed if a remote delivery caused a very long error message to be
2291 recorded - for instance if somebody sent an entire SpamAssassin report back
2292 as a large number of 550 error lines. This bug was coincidentally fixed by
2293 increasing the size of one of Exim's internal buffers (big_buffer) that
2294 happened as part of the Exiscan merge. However, to be on the safe side, I
2295 have made the code more robust (and fixed the comments that describe what
2298 33. Some experimental protocols are using DNS PTR records for new purposes. The
2299 keys for these records are domain names, not reversed IP addresses. The
2300 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2301 leaves it alone. Component reversal etc. now happens only for IP addresses.
2304 34. The host_aton() function is supposed to be passed a string that is known
2305 to be a valid IP address. However, in the case of IPv6 addresses, it was
2306 not checking this. This is a hostage to fortune. Exim now panics and dies
2307 if the condition is not met. A case was found where this could be provoked
2308 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2309 components; fortuitously, this particular loophole had already been fixed
2310 by change 4.50/55 or 4.44/33 above.
2312 If there are any other similar loopholes, the new check in host_aton()
2313 itself should stop them being exploited. The report I received stated that
2314 data on the command line could provoke the exploit when Exim was running as
2315 exim, but did not say which command line option was involved. All I could
2316 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2317 running as the user.
2320 35. There was a buffer overflow vulnerability in the SPA authentication code
2321 (which came originally from the Samba project). I have added a test to the
2322 spa_base64_to_bits() function which I hope fixes it.
2325 36. The daemon start-up calls getloadavg() while still root for those OS that
2326 need the first call to be done as root, but it missed one case: when
2327 deliver_queue_load_max is set with deliver_drop_privilege. This is
2328 necessary for the benefit of the queue runner, because there is no re-exec
2329 when deliver_drop_privilege is set.
2331 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2332 was in use, was not putting the data itself into the right store pool;
2333 consequently, it could be overwritten for a subsequent message in the same
2334 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2337 38. Sometimes the final signoff response after QUIT could fail to get
2338 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2339 < 0 before doing a fflush(). This bug looks as though it goes back to the
2340 introduction of TLS in release 3.20, but "sometimes" must have been rare
2341 because the tests only now provoked it.
2347 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2348 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2349 Now it returns an integer. A number of calls were still expecting a T/F
2350 return. Fortuitously, in all cases, the tests worked in OK situations,
2351 which is the norm. However, things would have gone wrong on any write
2352 failures on the smtp file descriptor. This function is used when sending
2353 messages over SMTP and also when doing verify callouts.
2355 2. When Exim is called to do synchronous delivery of a locally submitted
2356 message (the -odf or -odi options), it no longer closes stderr before doing
2359 3. Implemented the mua_wrapper option.
2361 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2363 5. Implemented the functions header_remove(), header_testname(),
2364 header_add_at_position(), and receive_remove_recipient(), and exported them
2367 6. If an ACL "warn" statement specified the addition of headers, Exim already
2368 inserted X-ACL-Warn: at the start if there was no header name. However, it
2369 was not making this test for the second and subsequent header lines if
2370 there were newlines in the string. This meant that an invalid header could
2371 be inserted if Exim was badly configured.
2373 7. Allow an ACL "warn" statement to add header lines at the start or after all
2374 the Received: headers, as well as at the end.
2376 8. Added the rcpt_4xx retry error code.
2378 9. Added postmaster_mailfrom=xxx to callout verification option.
2380 10. Added mailfrom=xxxx to the callout verification option, for verify=
2383 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2384 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2386 12. Inserted some casts to stop certain compilers warning when using pointer
2387 differences as field lengths or precisions in printf-type calls (mostly
2388 affecting debugging statements).
2390 13. Added optional readline() support for -be (dynamically loaded).
2392 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2393 same clock tick as a message's arrival, so that its received time was the
2394 same as the "first fail" time on the retry record, and that message
2395 remained on the queue past the ultimate address timeout, every queue runner
2396 would try a delivery (because it was past the ultimate address timeout) but
2397 after another failure, the ultimate address timeout, which should have then
2398 bounced the address, did not kick in. This was a "< instead of <=" error;
2399 in most cases the first failure would have been in the next clock tick
2400 after the received time, and all would be well.
2402 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2403 being recognized when the domain list was tested by the match_domain
2404 condition in an expansion string.
2406 16. Added the ${str2b64: operator.
2408 17. Exim was always calling setrlimit() to set a large limit for the number of
2409 processes, without checking whether the existing limit was already
2410 adequate. (It did check for the limit on file descriptors.) Furthermore,
2411 errors from getrlimit() and setrlimit() were being ignored. Now they are
2412 logged to the main and panic logs, but Exim does carry on, to try to do its
2413 job under whatever limits there are.
2415 18. Imported PCRE 5.0.
2417 19. Trivial typo in log message " temporarily refused connection" (the leading
2420 20. If the log selector return_path_on_delivery was set and an address was
2421 redirected to /dev/null, the delivery process crashed because it assumed
2422 that a return path would always be set for a "successful" delivery. In this
2423 case, the whole delivery is bypassed as an optimization, and therefore no
2426 21. Internal re-arrangement: the function for sending a challenge and reading
2427 a response while authentication was assuming a zero-terminated challenge
2428 string. It's now changed to take a pointer and a length, to allow for
2429 binary data in such strings.
2431 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2433 23. Exim was not respecting finduser_retries when seeking the login of the
2434 uid under which it was called; it was always trying 10 times. (The default
2435 setting of finduser_retries is zero.) Also, it was sleeping after the final
2436 failure, which is pointless.
2438 24. Implemented tls_on_connect_ports.
2440 25. Implemented acl_smtp_predata.
2442 26. If the domain in control=submission is set empty, Exim assumes that the
2443 authenticated id is a complete email address when it generates From: or
2444 Sender: header lines.
2446 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2447 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2448 chown and chgrp in /bin and hostname in /usr/bin.
2450 28. Exim was keeping the "process log" file open after each use, just as it
2451 does for the main log. This opens the possibility of it remaining open for
2452 long periods when the USR1 signal hits a daemon. Occasional processlog
2453 errors were reported, that could have been caused by this. Anyway, it seems
2454 much more sensible not to leave this file open at all, so that is what now
2457 29. The long-running daemon process does not normally write to the log once it
2458 has entered its main loop, and it closes the log before doing so. This is
2459 so that log files can straightforwardly be renamed and moved. However,
2460 there are a couple of unusual error situations where the daemon does write
2461 log entries, and I had neglected to close the log afterwards.
2463 30. The text of an SMTP error response that was received during a remote
2464 delivery was being truncated at 512 bytes. This is too short for some of
2465 the long messages that one sometimes sees. I've increased the limit to
2468 31. It is now possible to make retry rules that apply only when a message has a
2469 specific sender, in particular, an empty sender.
2471 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2472 it possible to be selective about when SMTP synchronization is enforced.
2474 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2476 32. Implemented hosts_connection_nolog.
2478 33. Added an ACL for QUIT.
2480 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2483 35. Added mailbox_size and mailbox_filecount to appendfile.
2485 36. Added control = no_multiline_responses to ACLs.
2487 37. There was a bug in the logic of the code that waits for the clock to tick
2488 in the case where the clock went backwards by a substantial amount such
2489 that the microsecond fraction of "now" was more than the microsecond
2490 fraction of "then" (but the whole seconds number was less).
2492 38. Added support for the libradius Radius client library this is found on
2493 FreeBSD (previously only the radiusclient library was supported).
2499 1. When certain lookups returned multiple values in the form name=value, the
2500 quoting of the values was not always being done properly. Specifically:
2501 (a) If the value started with a double quote, but contained no whitespace,
2503 (b) If the value contained whitespace other than a space character (i.e.
2504 tabs or newlines or carriage returns) it was not quoted.
2505 This fix has been applied to the mysql and pgsql lookups by writing a
2506 separate quoting function and calling it from the lookup code. The fix
2507 should probably also be applied to nisplus, ibase and oracle lookups, but
2508 since I cannot test any of those, I have not disturbed their existing code.
2510 2. A hit in the callout cache for a specific address caused a log line with no
2511 reason for rejecting RCPT. Now it says "Previous (cached) callout
2512 verification failure".
2514 3. There was an off-by-one bug in the queryprogram router. An over-long
2515 return line was truncated at 256 instead of 255 characters, thereby
2516 overflowing its buffer with the terminating zero. As well as fixing this, I
2517 have increased the buffer size to 1024 (and made a note to document this).
2519 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2520 when Exim is waiting for an SMTP response from a remote server, Exim
2521 restarts its select() call on the socket, thereby resetting its timeout.
2522 This is not a problem when such interrupts are rare. Somebody set up a cron
2523 job to run exiwhat every 2 minutes, which is less than the normal select()
2524 timeout (5 or 10 minutes). This meant that the select() timeout never
2525 kicked in because it was always reset. I have fixed this by comparing the
2526 time when an interrupt arrives with the time at the start of the first call
2527 to select(). If more time than the timeout has elapsed, the interrupt is
2528 treated as a timeout.
2530 5. Some internal re-factoring in preparation for the addition of Sieve
2531 extensions (by MH). In particular, the "personal" test is moved to a
2532 separate function, and given an option for scanning Cc: and Bcc: (which is
2533 not set for Exim filters).
2535 6. When Exim created an email address using the login of the caller as the
2536 local part (e.g. when creating a From: or Sender: header line), it was not
2537 quoting the local part when it contained special characters such as @.
2539 7. Installed new OpenBSD configuration files.
2541 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2542 try to make them clearer.
2544 9. Callout options, other than the timeout value, were being ignored when
2545 verifying sender addresses in header lines. For example, when using
2547 verify = header_sender/callout=no_cache
2549 the cache was (incorrectly) being used.
2551 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2552 only the Cygwin environment.
2554 11. When return_path_on_delivery was set as a log selector, if different remote
2555 addresses in the same message used different return paths and parallel
2556 remote delivery occurred, the wrong values would sometimes be logged.
2557 (Whenever a remote delivery process finished, the return path value from
2558 the most recently started remote delivery process was logged.)
2560 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2561 lines when AUTH and/or TLS are in use. This is the "received protocol"
2562 field. Exim used to use "asmtp" for authenticated SMTP, without any
2563 indication (in the protocol name) for TLS use. Now it follows the RFC and
2564 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2565 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2566 names appear in log lines as well as in Received: header lines.
2568 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2569 extensions, and comparison tests, and to fix some bugs.
2571 14. Changes to the "personal" filter test:
2573 (1) The test was buggy in that it was just doing the equivalent of
2574 "contains" tests on header lines. For example, if a user's address was
2575 anne@some.where, the "personal" test would incorrectly be true for
2577 To: susanne@some.where
2579 This test is now done by extracting each address from the header in turn,
2580 and checking the entire address. Other tests that are part of "personal"
2581 are now done using regular expressions (for example, to check local parts
2582 of addresses in From: header lines).
2584 (2) The list of non-personal local parts in From: addresses has been
2585 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2586 taken from the Sieve specification recommendations.
2588 (3) If the message contains any header line starting with "List-" it is
2589 treated as non-personal.
2591 (4) The test for "circular" in the Subject: header line has been removed
2592 because it now seems ill-conceived.
2594 15. Minor typos in src/EDITME comments corrected.
2596 16. Installed latest exipick from John Jetmore.
2598 17. If headers_add on a router specified a text string that was too long for
2599 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2600 of string_sprintf() is now avoided.
2602 18. $message_body_size was not set (it was always zero) when running the DATA
2603 ACL and the local_scan() function.
2605 19. For the "mail" command in an Exim filter, no default was being set for
2606 the once_repeat time, causing a random time value to be used if "once" was
2607 specified. (If the value happened to be <= 0, no repeat happened.) The
2608 default is now 0s, meaning "never repeat". The "vacation" command was OK
2609 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2610 (I found it when inspecting the code).
2612 20. There is now an overall timeout for performing a callout verification. It
2613 defaults to 4 times the callout timeout, which applies to individual SMTP
2614 commands during the callout. The overall timeout applies when there is more
2615 than one host that can be tried. The timeout is checked before trying the
2616 next host. This prevents very long delays if there are a large number of
2617 hosts and all are timing out (e.g. when the network connections are timing
2618 out). The value of the overall timeout can be changed by specifying an
2619 additional sub-option for "callout", called "maxwait". For example:
2621 verify = sender/callout=5s,maxwait=20s
2623 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2624 to the end before writing, but this should make it even safer).
2626 22. Exim was forgetting that it had advertised PIPELINING for the second and
2627 subsequent messages on an SMTP connection. It was also not resetting its
2628 memory on STARTTLS and an internal HELO.
2630 23. When Exim logs an SMTP synchronization error within a session, it now
2631 records whether PIPELINING has been advertised or not.
2633 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2634 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2635 rather than long int.
2637 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2639 26. Added the never_mail option to autoreply.
2645 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2646 crash if the getsockname() call failed; this can happen if a connection is
2647 closed very soon after it is established. The problem was simply in the
2648 order in which certain operations were done, causing Exim to try to write
2649 to the SMTP stream before it had set up the file descriptor. The bug has
2650 been fixed by making things happen in the correct order.
2656 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2657 before the connection was closed, thus losing the rejection response.
2659 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2660 some early Solaris releases, but causes trouble in current releases where
2661 socklen_t is defined.
2663 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2666 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2669 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2670 that releasing the top of it at the end releases what was used for sub-
2671 expansions (unless the block got too big). However, discard this block if
2672 the first thing is a variable or header, so that we can use its block when
2673 it is dynamic (useful for very large $message_headers, for example).
2675 6. Lookups now cache *every* query, not just the most recent. A new, separate
2676 store pool is used for this. It can be recovered when all lookup caches are
2677 flushed. Lookups now release memory at the end of their result strings.
2678 This has involved some general refactoring of the lookup sources.
2680 7. Some code has been added to the store_xxx() functions to reduce the amount
2681 of flapping under certain conditions.
2683 8. log_incoming_interface used to affect only the <= reception log lines. Now
2684 it causes the local interface and port to be added to several more SMTP log
2685 lines, for example "SMTP connection from", and rejection lines.
2687 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2689 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2691 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2692 could be overwritten at the end of the current message (or the start of a
2693 new message if it was set in a HELO ACL). The value is now preserved for
2694 the duration of the SMTP connection.
2696 12. If a transport had a headers_rewrite setting, and a matching header line
2697 contained an unqualified address, that address was qualified, even if it
2698 did not match any rewriting rules. The underlying bug was that the values
2699 of the flags that permit the existence of unqualified sender and recipient
2700 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2701 non-local messages, and by -bnq for local messages) were not being
2702 preserved with the message after it was received.
2704 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2705 "next input=" as part of the text comprising the host identity instead of
2706 the correct text. The code was using the same buffer for two different
2707 strings. However, depending on which order the printing function evaluated
2708 its arguments, the bug did not always show up. Under Linux, for example, my
2709 test suite worked just fine.
2711 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2712 doesn't work with some older versions of Perl. It has been changed to "my",
2713 which in any case is probably the better facility to use.
2715 15. A really picky compiler found some instances of statements for creating
2716 error messages that either had too many or two few arguments for the format
2719 16. The size of the buffer for calls to the DNS resolver has been increased
2720 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2721 for addresses that have a lot of PTR records. This alleviates a problem; it
2722 does not fully solve it.
2724 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2725 buffer now truncates the list and logs the incident, which is the same
2726 action as happens when Exim is looking up a host name and its aliases.
2727 Previously in this situation something unpredictable would happen;
2728 sometimes it was "internal error: store_reset failed".
2730 18. If a server dropped the connection unexpectedly when an Exim client was
2731 using GnuTLS and trying to read a response, the client delivery process
2732 crashed while trying to generate an error log message.
2734 19. If a "warn" verb in an ACL added multiple headers to a message in a single
2735 string, for example:
2737 warn message = H1: something\nH2: something
2739 the text was added as a single header line from Exim's point of view
2740 though it ended up OK in the delivered message. However, searching for the
2741 second and subsequent header lines using $h_h2: did not work. This has been
2742 fixed. Similarly, if a system filter added multiple headers in this way,
2743 the routers could not see them.
2745 20. Expanded the error message when iplsearch is called with an invalid key to
2746 suggest using net-iplsearch in a host list.
2748 21. When running tests using -bh, any delays imposed by "delay" modifiers in
2749 ACLs are no longer actually imposed (and a message to that effect is
2752 22. If a "gecos" field in a passwd entry contained escaped characters, in
2753 particular, if it contained a \" sequence, Exim got it wrong when building
2754 a From: or a Sender: header from that name. A second bug also caused
2755 incorrect handling when an unquoted " was present following a character
2756 that needed quoting.
2758 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2759 was not being matched caselessly.
2761 24. Arranged for all hyphens in the exim.8 source to be escaped with
2764 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2765 properly. Recipient callout cache records were still being keyed to include
2766 the sender, even when use_sender was set false. This led to far more
2767 callouts that were necessary. The sender is no longer included in the key
2768 when use_sender is false.
2770 26. Added "control = submission" modifier to ACLs.
2772 27. Added the ${base62d: operator to decode base 62 numbers.
2774 28. dnsdb lookups can now access SRV records.
2776 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2777 the configuration file.
2779 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2780 (-v) mode. This makes the output for a verbose queue run more intelligible.
2782 31. Added a use_postmaster feature to recipient callouts.
2784 32. Added the $body_zerocount variable, containing the number of binary zero
2785 bytes in the message body.
2787 33. The time of last modification of the "new" subdirectory is now used as the
2788 "mailbox time last read" when there is a quota error for a maildir
2791 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2793 35. Added +ignore_unknown as a special item in host lists.
2795 36. Code for decoding IPv6 addresses in host lists is now included, even if
2796 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2797 address was recognized as an IP address, but was then not correctly decoded
2798 into binary, causing unexpected and incorrect effects when compared with
2805 1. Very minor rewording of debugging text in manualroute to say "list of
2806 hosts" instead of "hostlist".
2808 2. If verify=header_syntax was set, and a header line with an unqualified
2809 address (no domain) and a large number of spaces between the end of the
2810 name and the colon was received, the reception process suffered a buffer
2811 overflow, and (when I tested it) crashed. This was caused by some obsolete
2812 code that should have been removed. The fix is to remove it!
2814 3. When running in the test harness, delay a bit after writing a bounce
2815 message to get a bit more predictability in the log output.
2817 4. Added a call to search_tidyup() just before forking a reception process. In
2818 theory, someone could use a lookup in the expansion of smtp_accept_max_
2819 per_host which, without the tidyup, could leave open a database connection.
2821 5. Added the variables $recipient_data and $sender_data which get set from a
2822 lookup success in an ACL "recipients" or "senders" condition, or a router
2823 "senders" option, similar to $domain_data and $local_part_data.
2825 6. Moved the writing of debug_print from before to after the "senders" test
2828 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2829 problems for message scanning, either using a data ACL, or using
2830 local_scan() because the Received: header was not generated till after they
2831 were called (in order to set the time as the time of reception completion).
2832 I have revised the way this works. The header is now generated after the
2833 body is received, but before the ACL or local_scan() are called. After they
2834 are run, the timestamp in the header is updated.
2840 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2841 before starting a queue runner without re-exec. This happened only when
2842 deliver_drop_privilege was set or when the Exim user was set to root. The
2843 effect of the bug was that timeouts during subsequent deliveries caused
2844 crashes instead of being properly handled. The handler is now left at its
2845 default (and expected) setting.
2847 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2848 message, again when deliver_drop_privilege is set or Exim is run as root.
2849 The bug described in (1) was not present in this case, but the tidying up
2850 of the other signals was missing. I have made the two cases consistent.
2852 3. The ignore_target_hosts setting on a manualroute router was being ignored
2853 for hosts that were looked up using the /MX notation.
2855 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2858 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2859 operated on the sender address. After changing the $sender_address to <>
2860 for the sender address verify, Exim was re-instated it as the original
2861 (before rewriting) address, but remembering that it had rewritten it, so it
2862 wasn't rewriting it again. This bug also had the effect of breaking the
2863 sender address verification caching when the sender address was rewritten.
2865 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2866 This has been changed so that if the ip literal address matches
2867 ignore_target_hosts, the router declines.
2869 7. Added expansion conditions match_domain, match_address, and match_local_
2870 part (NOT match_host).
2872 8. The placeholder for the Received: header didn't have a length field set.
2874 9. Added code to Exim itself and to exim_lock to test for a specific race
2875 condition that could lead to file corruption when using MBX delivery. The
2876 issue is with the lockfile that is created in /tmp. If this file is removed
2877 after a process has opened it but before that process has acquired a lock,
2878 there is the potential for a second process to recreate the file and also
2879 acquire a lock. This could lead to two Exim processes writing to the file
2880 at the same time. The added code performs the same test as UW imapd; it
2881 checks after acquiring the lock that its file descriptor still refers to
2882 the same named file.
2884 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2885 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2888 11. Added the smtp_active_hostname option. If used, this will typically be made
2889 to depend on the incoming interface address. Because $interface_address is
2890 not set up until the daemon has forked a reception process, error responses
2891 that can happen earlier (such as "too many connections") no longer contain
2894 12. If an expansion in a condition on a "warn" statement fails because a lookup
2895 defers, the "warn" statement is abandoned, and the next ACL statement is
2896 processed. Previously this caused the whole ACL to be aborted.
2898 13. Added the iplsearch lookup type.
2900 14. Added ident_timeout as a log selector.
2902 15. Added tls_certificate_verified as a log selector.
2904 16. Added a global option tls_require_ciphers (compare the smtp transport
2905 option of the same name). This controls incoming TLS connections.
2907 17. I finally figured out how to make tls_require_ciphers do a similar thing
2908 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2909 before starting the TLS session.
2911 18. Tabs are now shown as \t in -bP output.
2913 19. If the log selector return_path_on_delivery was set, Exim crashed when
2914 bouncing a message because it had too many Received: header lines.
2916 20. If two routers both had headers_remove settings, and the first one included
2917 a superfluous trailing colon, the final name in the first list and the
2918 first name in the second list were incorrectly joined into one item (with a
2919 colon in the middle).
2925 1. Added -C and -D options to the exinext utility, mainly to make it easier
2926 to include in the automated testing, but these could be helpful when
2927 multiple configurations are in use.
2929 2. The exinext utility was not formatting the output nicely when there was
2930 an alternate port involved in the retry record key, nor when there was a
2931 message id as well (for retries that were specific to a specific message
2932 and a specific host). It was also confused by IPv6 addresses, because of
2933 the additional colons they contain. I have fixed the IPv4 problem, and
2934 patched it up to do a reasonable job for IPv6.
2936 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2937 delivery, the log line now contains "pipelined" if PIPELINING was used.
2939 4. An SMTP transport process used to panic and die if the bind() call to set
2940 an explicit outgoing interface failed. This has been changed; it is now
2941 treated in the same way as a connect() failure.
2943 5. A reference to $sender_host_name in the part of a conditional expansion
2944 that was being skipped was still causing a DNS lookup. This no longer
2947 6. The def: expansion condition was not recognizing references to header lines
2948 that used bh_ and bheader_.
2950 7. Added the _cache feature to named lists.
2952 8. The code for checking quota_filecount in the appendfile transport was
2953 allowing one more file than it should have been.
2955 9. For compatibility with Sendmail, the command line option
2963 and sets the incoming protocol and host name (for trusted callers). The
2964 host name and its colon can be omitted when only the protocol is to be set.
2965 Note the Exim already has two private options, -pd and -ps, that refer to
2966 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2967 "s", but I don't think that's a major issue.
2969 10. A number of refactoring changes to the code, none of which should affect
2972 (a) The number of logging options was getting close to filling up the
2973 32-bit word that was used as a bit map. I have split them into two classes:
2974 those that are passed in the argument to log_write(), and those that are
2975 only ever tested independently outside of that function. These are now in
2976 separate 32-bit words, so there is plenty of room for expansion again.
2977 There is no change in the user interface or the logging behaviour.
2979 (b) When building, for example, log lines, the code previously used a
2980 macro that called string_cat() twice, in order to add two strings. This is
2981 not really sufficiently general. Furthermore, there was one instance where
2982 it was actually wrong because one of the argument was used twice, and in
2983 one call a function was used. (As it happened, calling the function twice
2984 did not affect the overall behaviour.) The macro has been replaced by a
2985 function that can join an arbitrary number of extra strings onto a growing
2988 (c) The code for expansion conditions now uses a table and a binary chop
2989 instead of a serial search (which was left over from when there were very
2990 few conditions). Also, it now recognizes conditions like "pam" even when
2991 the relevant support is not compiled in: a suitably worded error message is
2992 given if an attempt is made to use such a condition.
2994 11. Added ${time_interval:xxxxx}.
2996 12. A bug was causing one of the ddress fields not to be passed back correctly
2997 from remote delivery subprocesses. The field in question was not being
2998 subsequently used, so this caused to problems in practice.
3000 13. Added new log selectors queue_time and deliver_time.
3002 14. Might have fixed a bug in maildirsizefile handling that threw up
3003 "unexpected character" debug warnings, and recalculated the data
3004 unnecessarily. In any case, I expanded the warning message to give more
3007 15. Added the message "Restricted characters in address" to the statements in
3008 the default ACL that block characters like @ and % in local parts.
3010 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3011 Three changes have been made:
3013 (a) There was a serious bug; a negative response to MAIL caused the whole
3014 recipient domain to be cached as invalid, thereby blocking all messages
3015 to all local parts at the same domain, from all senders. This bug has
3016 been fixed. The domain is no longer cached after a negative response to
3017 MAIL if the sender used is not empty.
3019 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3022 (c) A new callout option, "use_sender" has been added for people who want
3023 the modified behaviour.
3029 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3032 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3033 indeed breaks things for older releases.
3035 3. Added additional logging to the case where there is a problem reading data
3036 from a filter that is running in a subprocess using a pipe, in order to
3037 try to track down a specific problem.
3039 4. Testing facility fudge: when running in the test harness and attempting
3040 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3041 getting "No route to host". Convert this to a timeout.
3043 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3046 6. Some OS don't have socklen_t but use size_t instead. This affects the
3047 fifth argument of getsockopt() amongst other things. This is now
3048 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3049 can be set for individual OS. I have set it for SunOS5, OSF1, and
3050 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3051 some earlier ones do not.
3053 7. Change 4.30/15 was not doing the test caselessly.
3055 8. The standard form for an IPv6 address literal was being rejected by address
3056 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3057 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3058 this, as well as the form without the "IPv6" on the front (but only when
3059 address literals are enabled, of course).
3061 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3063 10. Exim crashed if a message with an empty sender address specified by -f
3064 encountered a router with an errors_to setting. This could be provoked only
3065 by a command such as
3069 where an empty string was supplied; "<>" did not hit this bug.
3071 11. Installed PCRE release 4.5.
3073 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3074 remained set. It is now erased.
3076 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3077 times from message ids (which are base 36 rather than the normal 62).
3079 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3080 were being counted as actual protocol errors, and logged if the log
3081 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3082 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3083 and DATA following a set of rejected RCPTs do not count as protocol errors.
3084 In other words, Exim assumes they were pipelined, though this may not
3085 actually be the case. Of course, in all cases the client gets an
3086 appropriate error code.
3088 15. If a lookup fails in an ACL condition, a message about the failure may
3089 be available; it is used if testing the ACL cannot continue, because most
3090 such messages specify what the cause of the deferral is. However, some
3091 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3092 that caused an old message to be retained and used if a later statement
3093 caused a defer, replacing the real cause of the deferral.
3095 16. If an IP address had so many PTR records that the DNS lookup buffer
3096 was not large enough to hold them, Exim could crash while trying to process
3097 the truncated data. It now detects and logs this case.
3099 17. Further to 4.21/58, another change has been made: if (and only if) the
3100 first line of a message (the first header line) ends with CRLF, a bare LF
3101 in a subsequent header line has a space inserted after it, so as not to
3102 terminate the header.
3104 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3105 unnecessarily, used atoi() instead of strtol(), and didn't check the
3106 termination when getting file sizes from file names by regex.
3108 19. Completely re-implemented the support for maildirsize files, in the light
3109 of a number of problems with the previous contributed implementation
3110 (4.30/29). In particular:
3112 . If the quota is zero, the maildirsize file is maintained, but no quota is
3115 . If the maildir directory does not exist, it is created before any attempt
3116 to write a maildirsize file.
3118 . The quota value in the file is just a cache; if the quota is changed in
3119 the transport, the new value overrides.
3121 . A regular expression is available for excluding directories from the
3124 20. The autoreply transport checks the characters in options that define the
3125 message's headers; it allows continued headers, but it was checking with
3126 isspace() after an embedded newline instead of explicitly looking for a
3129 21. If all the "regular" hosts to which an address was routed had passed their
3130 expiry times, and had not reached their retry times, the address was
3131 bounced, even if fallback hosts were defined. Now Exim should go on to try
3134 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3135 equivalent code in the SMTP transport. Some hosts send humungous responses
3136 to HELO/EHLO, more than 1024 it seems.
3138 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3139 gives compiler warnings in some environments. I've now done it "properly",
3142 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3143 (because of problems with the built-in one) was declared to return uschar *
3144 instead of char *, causing compiler failure.
3146 25. Fixed a file descriptor leak when processing alias/forward files.
3148 26. Fixed a minor format string issue in dbfn.c.
3150 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3152 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3153 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3156 29. When a sender address is verified, it is cached, to save repeating the test
3157 when there is more than one recipient in a message. However, when the
3158 verification involves a callout, it is possible for different callout
3159 options to be set for different recipients. It is too complicated to keep
3160 track of this in the cache, so now Exim always runs a verification when a
3161 callout is required, relying on the callout cache for the optimization.
3162 The overhead is duplication of the address routing, but this should not be
3165 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3166 to be verified with callout=postmaster, and the main callout worked but the
3167 postmaster check failed, the verification correctly failed. However, if a
3168 subsequent RCPT command asked for sender verification *without* the
3169 postmaster check, incorrect caching caused this verification also to fail,
3172 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3173 it was not caching the DNS options (qualify_single, search_parents) that
3174 were used when the lookup failed. A subsequent lookup with different
3175 options therefore always gave the same answer, though there were cases
3176 where it should not have. (Example: a "domains = !$mx_any" option on a
3177 dnslookup router: the "domains" option is always processed without any
3178 widening, but the router might have qualify_single set.) Now Exim uses the
3179 cached value only when the same options are set.
3181 32. Added John Jetmore's "exipick" utility to the distribution.
3183 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3184 than a timeout (e.g. a certificate is required, and is not provided), an
3185 Exim server now closes the connection immediately. Previously it waited for
3186 the client to close - but if the client is SSL, it seems that they each
3187 wait for each other, leading to a delay before one of them times out.
3189 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3190 maintained 0.8.x compatibility because I don't think many are using it, and
3191 it is clearly obsolete.
3193 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3196 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3197 client certificate was expired. A simple patch fixes this, though I don't
3198 understand the full logic of why the verify callback is called multiple
3201 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3202 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3203 which causes problems with some clients (such as the Certicom SSL Plus
3204 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3205 disables the coutermeasure allowing Eudora to connect."
3207 38. Exim was not checking that a write() to a log file succeeded. This could
3208 lead to Bad Things if a log got too big, in particular if it hit a file
3209 size limit. Exim now panics and dies if it cannot write to a log file, just
3210 as it does if it cannot open a log file.
3212 39. Modified OS/Makefile-Linux so that it now contains
3214 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3216 The two -D definitions ensure that Exim is compiled with large file
3217 support, which makes it possible to handle log files that are bigger than
3220 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3221 instance) a domain was checked against a named list that involved a lookup,
3222 causing $domain_data to be set, then another domain was checked against the
3223 same list, then the first domain was re-checked, the value of $domain_data
3224 after the final check could be wrong. In particular, if the second check
3225 failed, it could be set empty. This bug probably also applied to
3228 41. The strip_trailing_dot option was not being applied to the address given
3229 with the -f command-line option.
3231 42. The code for reading a message's header from the spool was incrementing
3232 $received_count, but never initializing it. This meant that the value was
3233 incorrect (doubled) while delivering a message in the same process in which
3234 it was received. In the most common configuration of Exim, this never
3235 happens - a fresh exec is done - but it can happen when
3236 deliver_drop_privilege is set.
3238 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3239 it now includes up to 150 characters of the unexpected data in the log
3242 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3243 and building data strings. The size of both of these buffers was 10 000
3244 bytes - far larger than anybody would *ever* want, thought I. Needless to
3245 say, somebody hit the limit. I have increased the maximum line length to
3246 20 000 and the maximum data length of concatenated lines to 100 000. I have
3247 also fixed two bugs, because there was no checking on these buffers. Tsk,
3248 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3249 buffer is too small.
3251 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3252 lsearch lookups. Now it does.
3254 46. When parsing a route_list item in a manualroute router, a fixed-length
3255 buffer was used for the list of hosts. I made this 1024 bytes long,
3256 thinking that nobody would ever have a list of hosts that long. Wrong.
3257 Somebody had a whole pile of complicated expansion conditions, and the
3258 string was silently truncated, leading to an expansion error. It turns out
3259 that it is easier to change to an unlimited length (owing to other changes
3260 that have happened since this code was originally written) than to build
3261 structure for giving a limitation error. The length of the item that
3262 expands into the list of hosts is now unlimited.
3264 47. The lsearch lookup could not handle data where the length of text line was
3265 more than 4095 characters. Such lines were truncated, leading to shortened
3266 data being returned. It should now handle lines of any length.
3268 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3269 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3272 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3273 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3274 no longer happens in comment lines. A list of replacements is now placed
3275 at the head of all of the source files, except those whose only change is
3276 to replace PERL_COMMAND in the very first #! line.
3278 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3279 messages on the queue, with a bottom-up merge sort, using code contributed
3280 by Michael Haardt. This should make operations like -bp somewhat faster on
3281 large queues. It won't affect queue runners, except when queue_run_in_order
3284 51. Installed eximstats 1.31 in the distribution.
3286 52. Added support for SRV lookups to the dnslookup router.
3288 53. If an ACL referred to $message_body or $message_body_end, the value was not
3289 reset for any messages that followed in the same SMTP session.
3291 54. The store-handling optimization for building very long strings was not
3292 differentiating between the different store pools. I don't think this
3293 actually made any difference in practice, but I've tidied it.
3295 55. While running the routers to verify a sender address, $sender_address
3296 was still set to the sender address. This is wrong, because when routing to
3297 send a bounce to the sender, it would be empty. Therefore, I have changed
3298 it so that, while verifying a sender address, $sender_address is set to <>.
3299 (There is no change to what happens when verifying a recipient address.)
3301 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3302 target A or AAAA records (if not already returned) without resetting the
3303 qualify_single or search_parents options of the DNS resolver. These are
3304 inappropriate in this case because the targets of MX and SRV records must
3305 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3306 target that, when qualified, matched something in the local domain. These
3307 two options are now turned off when doing these lookups.
3309 57. It seems that at least some releases of Reiserfs (which does not have the
3310 concept of a fixed number of inodes) returns zero and not -1 for the
3311 number of available inodes. This interacted badly with check_spool_inodes,
3312 which assumed that -1 was the "no such thing" setting. What I have done is
3313 to check that the total number of inodes is greater than zero before doing
3314 the test of how many are available.
3316 58. When a "warn" ACL statement has a log_message modifier, the message is
3317 remembered, and not repeated. This is to avoid a lot of repetition when a
3318 message has many recipients that cause the same warning to be written.
3319 Howewer, Exim was preserving the list of already written lines for an
3320 entire SMTP session, which doesn't seem right. The memory is now reset if a
3321 new message is started.
3323 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3324 debugging output unless log_rewrite was also set.
3326 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3327 of (int)(handle) when we know that handle contains (void *)(-1).
3329 61. The Exim daemon panic-logs an error return when it closes the incoming
3330 connection. However "connection reset by peer" seems to be common, and
3331 isn't really an error worthy of noting specially, so that particular error
3334 62. When Exim is trying to find all the local interfaces, it used to panic and
3335 die if the ioctl to get the interface flags failed. However, it seems that
3336 on at least one OS (Solaris 9) it is possible to have an interface that is
3337 included in the list of interfaces, but for which you get a failure error
3338 for this call. This happens when the interface is not "plumbed" into a
3339 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3340 failure of the "get flags" call assumes that the interface is down.
3342 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3343 makes life easier for people who are doing arithmetic on fields extracted
3344 from dates, where you often get leading zeros that should not be
3345 interpreted as octal.
3347 64. Added qualify_domain to the redirect router, to override the global
3350 65. If a pathologically long header line contained very many addresses (the
3351 report of this problem mentioned 10 000) and each of them was rewritten,
3352 Exim could use up a very large amount of memory. (It kept on making new
3353 copies of the header line as it rewrote, and never released the old ones.)
3354 At the expense of a bit more processing, the header rewriting function has
3355 been changed so that it no longer eats memory in this way.
3357 66. The generation of the Received: header has been moved from the time that a
3358 message starts to be received, to the time that it finishes. The timestamp
3359 in the Received: header should now be very close to that of the <= log
3360 line. There are two side-effects of this change:
3362 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3363 logged header lines no longer include the local Received: line, because
3364 it has not yet been created. The same applies to a copy of the message
3365 that is returned to a non-SMTP sender when a message is rejected.
3367 (b) When a filter file is tested using -bf, no additional Received: header
3368 is added to the test message. After some thought, I decided that this
3371 This change does not affect the value of $received_for. It is still set
3372 after address rewriting, but before local_scan() is called.
3374 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3376 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3377 gave an unhelpful panic error message, and a defer error. I have managed to
3378 change this behaviour so that it now rejects any supplied certificate,
3379 which seems right, as the list of acceptable certificates is empty.
3381 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3382 gave an unhelpful defer error. I have not managed to make this reject any
3383 supplied certificates, but the error message it gives is "no certificate
3384 supplied", which is not helpful.
3386 70. exigrep's output now also includes lines that are not associated with any
3387 message, but which match the given pattern. Implemented by a patch from
3388 Martin Sluka, which also tidied up the Perl a bit.
3390 71. Recipient callout verification, like sender verification, was using <> in
3391 the MAIL FROM command. This isn't really the right thing, since the actual
3392 sender may affect whether the remote host accepts the recipient or not. I
3393 have changed it to use the actual sender in the callout; this means that
3394 the cache record is now keyed on a recipient/sender pair, not just the
3395 recipient address. There doesn't seem to be a real danger of callout loops,
3396 since a callout by the remote host to check the sender would use <>.
3397 [SEE ABOVE: changed after hitting problems.]
3399 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3400 temporary errors. However, in the case of such a code being given after
3401 the end of a data transmission (i.e. after ".") Exim was failing to write
3402 a retry record for the message. (Yes, there was some broken host that was
3403 actually sending 8xx at this point.)
3405 73. An unknown lookup type in a host list could cause Exim to panic-die when
3406 the list was checked. (An example that provoked this was putting <; in the
3407 middle of a list instead of at the start.) If this happened during a DATA
3408 ACL check, a -D file could be left lying around. This kind of configuration
3409 error no longer causes Exim to die; instead it causes a defer errror. The
3410 incident is still logged to the main and panic logs.
3412 74. Buglet left over from Exim 3 conversion. The message "too many messages
3413 in one connection" was written to the rejectlog but not the mainlog, except
3414 when address rewriting (yes!) was being logged.
3416 75. Added write_rejectlog option.
3418 76. When a system filter was run not as root (that is, when system_filter_user
3419 was set), the values of the $n variables were not being returned to the
3420 main process; thus, they were not subsequently available in the $sn
3423 77. Added +return_path_on_delivery log selector.
3425 78. A connection timeout was being treated differently from recipients deferred
3426 when testing hosts_max_try with a message that was older than the host's
3427 retry timeout. (The host should not be counted, thus allowing all hosts to
3428 be tried at least once before bouncing.) This may have been the cause of an
3429 occasionally reported bug whereby a message would remain on the queue
3430 longer than the retry timeout, but would be bounced if a delivery was
3431 forced. I say "may" because I never totally pinned down the problem;
3432 setting up timeout/retry tests is difficult. See also the next item.
3434 79. The ultimate address timeout was not being applied to errors that involved
3435 a combination of host plus message (for example, a timeout on a MAIL
3436 command). When an address resolved to a number of possible hosts, and they
3437 were not all tried for each delivery (e.g. because of hosts_max_try), a
3438 message could remain on the queue longer than the retry timeout.
3440 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3443 81. Fixed an obscure SMTP outgoing bug which required at least the following
3444 conditions: (a) there was another message waiting for the same server;
3445 (b) the server returned 5xx to all RCPT commands in the first message so
3446 that the message was not completed; (c) the server dropped the connection
3447 or gave a negative response to the RSET that Exim sends to abort the
3448 transaction. The observed case was a dropped connection after DATA that had
3449 been sent in pipelining mode. That is, the server had advertised PIPELINING
3450 but was not implementing it correctly. The effect of the bug was incorrect
3451 behaviour, such as trying another host, and this could lead to a crash.
3457 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3458 and daemon.c were passed as pointers to ints; they should have been
3459 pointers to socklen_t variables (which are typically unsigned ints).
3461 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3464 3. Fixed a really odd bug that affected only the testing scheme; patching a
3465 certain fixed string in the binary changed the value of another string that
3466 happened to be identical to the end of the original first string.
3468 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3469 name", it returns that address as the IP address. On some operating
3470 systems (e.g. Solaris), it also passes back the IP address string as the
3471 "host name". However, on others (e.g. Linux), it passes back an empty
3472 string. Exim wasn't checking for this, and was changing the host name to an
3473 empty string, assuming it had been canonicized.
3475 5. Although rare, it is permitted to have more than one PTR record for a given
3476 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3477 all the names associated with an address, because they do in Solaris.
3478 However, it seems that they do not in Linux for data that comes from the
3479 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3480 I found this out when I moved to a new Linux workstation and tried to run
3481 the Exim test suite.
3483 To get round this problem I have changed the code so that it now does its
3484 own call to the DNS to look up PTR records when searching for a host name.
3485 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3486 addresses that are only in /etc/hosts are still found.
3488 This behaviour is, however, controlled by an option called host_lookup_
3489 order, which defaults to "bydns:byaddr". If people want to use the other
3490 order, or indeed, just use one or the other means of lookup, they can
3491 specify it in this variable.
3493 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3494 some operating systems, this comes back from gethostbyaddr() as an empty
3495 string, and this is what Exim used to test for. However, it seems that in
3496 other systems, "." is yielded. Exim now tests for this case too.
3498 7. The values of check_spool_space and check_log_space are now held internally
3499 as a number of kilobytes instead of an absolute number of bytes. If a
3500 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3501 kilobyte. This means that much larger values can be stored.
3503 8. Exim monitor: an attempt to get the action menu when not actually pointing
3504 at a message produces an empty menu entitled "No message selected". This
3505 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3506 no entries in it ("Shell widget menu has zero width and/or height"). So I
3507 have added a single, blank menu entry in this case.
3509 9. Added ${quote_local_part.
3511 10. MIME decoding is now applied to the contents of Subject: header lines when
3514 11. Now that a reference to $sender_host_address automatically causes a reverse
3515 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3516 host lookup before query-style lookups in lists that might use this
3517 variable. This has therefore been abolished, and the "net-" prefix is no
3518 longer necessary for query-style lookups.
3520 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3521 have been a typo for LFLAGS, so it has been changed.
3523 13. The install script calls Exim with "-C /dev/null" in order to find the
3524 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3525 to be output. Howeve, since Exim outputs its version number before the
3526 error, it didn't break the script. It just looked ugly. I fixed this by
3527 always allowing "-C /dev/null" if the caller is root.
3529 14. Ignore overlarge ACL variable number when reading spool file - insurance
3530 against a later release with more variables having written the file.
3532 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3533 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3534 this, as well as the form without the "IPv6" on the front.
3536 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3537 OS/Makefile-Darwin file.
3539 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3540 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3542 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3543 variable. However, some OS use ints for inodes. I've added cast to long int
3544 to get rid of the compiler warning.
3546 19. I had forgotten to lock out "/../" in configuration file names when
3547 ALT_CONFIG_PREFIX was set.
3549 20. Routers used for verification do not need to specify transports. However,
3550 if such a router generated a host list, and callout was configured, Exim
3551 crashed, because it could not find a port number from the (non-existent)
3552 transport. It now assumes port 25 in this circumstance.
3554 21. Added the -t option to exigrep.
3556 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3557 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3558 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3559 them is set, LOOKUP_LSEARCH is forced.
3561 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3563 24. Added sender and host information to the "rejected by local_scan()" log
3564 line; previously there was no indication of these.
3566 25. Added .include_if_exists.
3568 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3569 Linux. It turns out that not all file systems support this. Apparently some
3570 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3571 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3572 not supported on the file descriptor, is now ignored when Exim is trying to
3573 sync a directory. This applies only to Linux.
3575 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3577 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3578 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3581 29. Added support for maildirsize files from supplied patch (modified a bit).
3583 30. The use of :fail: followed by an empty string could lead Exim to respond to
3584 sender verification failures with (e.g.):
3586 550 Verification failed for <xxx>
3587 550 Sender verify failed
3589 where the first response line was missing the '-' that indicates it is not
3590 the final line of the response.
3592 31. The loop for finding the name of the user that called Exim had a hardwired
3593 limit of 10; it now uses the value of finduser_retries, which is used for
3594 all other user lookups.
3596 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3599 33. Exim was neglecting to zero errno before one call of strtol() when
3600 expanding a string and expecting an integer value. On some systems this
3601 resulted in spurious "integer overflow" errors. Also, it was casting the
3602 result into an int without checking.
3604 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3605 did not work. The code looks as if it has *never* worked, though it appears
3606 to have been documented since at least releast 1.62. I have made it work.
3608 35. The "timeout_DNS" error in retry rules, also documented since at least
3609 1.62, also never worked. As it isn't clear exactly what this means, and
3610 clearly it isn't a major issue, I have abolished the feature by treating it
3611 as "timeout", and writing a warning to the main and panic logs.
3613 36. The display of retry rules for -brt wasn't always showing the error code
3616 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3617 timeout_connect_A, timeout_connect_MX.
3619 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3620 to the empty sender.
3622 39. The daemon was not analysing the content of -oX till after it had closed
3623 stderr and disconnected from the controlling terminal. This meant that any
3624 syntax errors were only noted on the panic log, and the return code from
3625 the command was 0. By re-arranging the code a little, I've made the
3626 decoding happen first, so such errors now appear on stderr, and the return
3627 code is 1. However, the actual setting up of the sockets still happens in
3628 the disconnected process, so errors there are still only recorded on the
3631 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3632 connections (as happens on some IP stacks) was logged at start up time as
3633 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3634 it from "IPv6 and IPv4", which means that two separate sockets are being
3637 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3638 says whether AF_INET or AF_INET6 was passed as an argument.
3640 42. Exiwhat output was messed up when time zones were included in log
3643 43. Exiwhat now gives more information about the daemon's listening ports,
3644 and whether -tls-on-connect was used.
3646 44. The "port" option of the smtp transport is now expanded.
3648 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3649 silently ignored. Now an error message is written to the main and panic
3652 46. There's a new ACL modifier called "logwrite" which writes to a log file
3653 as soon as it is encountered.
3655 47. Added $local_user_uid and $local_user_gid at routing time.
3657 48. Exim crashed when trying to verify a sender address that was being
3660 49. Exim was recognizing only a space character after ".include". It now also
3661 recognizes a tab character.
3663 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3664 extracting the relevant information from the specification. The man page no
3665 longer contains scrambled data for the -d option, and I've added a section
3666 at the front about calling Exim under different names.
3668 51. Added "extra_headers" argument to the "mail" command in filter files.
3670 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3673 53. Installed eximstats 1.29.
3675 54. Added transport_filter_timeout as a generic transport option.
3677 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3678 Cc: header lines. This was required by RFC 822, but it not required by RFC
3681 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3682 incoming messages that did not have them. Now it does so only if the
3683 message originates locally, that is, if there is no associated remote host
3684 address. When Resent- header lines are present, this applies to the Resent-
3685 lines rather than the non-Resent- lines.
3687 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3688 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3690 58. Messages for configuration errors now include the name of the main
3691 configuration file - useful now that there may be more than one file in a
3692 list (.included file names were always shown).
3694 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3695 for those rare installations that do not start the daemon as root or run it
3696 setuid root. I've cut out the call to initgroups() if the daemon is not
3699 60. The Exim user and group can now be bound into the binary as text strings
3700 that are looked up at the start of Exim's processing.
3702 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3704 62. Added $mailstore_basename variable.
3706 63. Installed patch to sieve.c from Michael Haardt.
3708 64. When Exim failed to open the panic log after failing to open the main log,
3709 the original message it was trying to log was written to stderr and debug
3710 output, but if they were not available (the usual case in production), it
3711 was lost. Now it is written to syslog before the two lines that record the
3712 failures to open the logs.
3714 65. Users' Exim filters run in subprocesses under the user's uid. It is
3715 possible for a "deliver" command or an alias in a "personal" command to
3716 provoke an address rewrite. If logging of address rewriting is configured,
3717 this fails because the process is not running as root or exim. There may be
3718 a better way of dealing with this, but for the moment (because 4.30 needs
3719 to be released), I have disabled address rewrite logging when running a
3720 filter in a non-root, non-exim process.
3726 1. The buildconfig auxiliary program wasn't quoting the value set for
3727 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3728 not defined. This bug was masked in 4.22 by the effect that was fixed in
3731 2. Some messages that were rejected after a message id was allocated were
3732 shown as "incomplete" by exigrep. It no longer does this for messages that
3733 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3735 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3736 have allow_domain_literals set, the ID did not get logged in the <= line.
3737 Domain literals are now always recognized in Message-ID: header lines.
3739 4. The first argument for a ${extract expansion item is the key name or field
3740 number. Leading and trailing spaces in this item were not being ignored,
3741 causing some misleading effects.
3743 5. When deliver_drop_privilege was set, single queue runner processes started
3744 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3745 same command in the process it spins off) were not dropping privilege.
3747 6. When the daemon running as "exim" started a queue runner, it always
3748 re-executed Exim in the spun-off process. This is a waste of effort when
3749 deliver_drop_privilege is set. The new process now just calls the
3750 queue-runner function directly.
3756 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3759 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3760 ignored. Though the use of -oP was forcing the writing of a pid file, it
3761 was always written to the default place.
3763 3. If the message "no IP address found for host xxxx" is generated during
3764 incoming verification, it is now followed by identification of the incoming
3765 connection (so you can more easily find what provoked it).
3767 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3769 5. Added some features to "harden" Exim a bit more against certain attacks:
3771 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3772 be put in Local/Makefile. This is like the never_users runtime option,
3773 but it cannot be overridden. The default setting is "root".
3775 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3776 prefix string with which any file named in a -C command line option
3779 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3780 is retained for -C and -D only if the caller of Exim is root. Without
3781 it, the exim user may also use -C and -D and retain privilege.
3783 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3784 command line option is disabled.
3786 6. Macro names set by the -D option must start with an upper case letter, just
3787 like macro names defined in the configuration file.
3789 7. Added "dereference=" facility to LDAP.
3791 8. Two instances of the typo "uknown" in the source files are fixed.
3793 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3794 the Configure-Makefile script screwed up while processing it.
3796 10. Incorporated PCRE 4.4.
3798 11. The SMTP synchronization check was not operating right at the start of an
3799 SMTP session. For example, it could not catch a HELO sent before the client
3800 waited for the greeting. There is now a check for outstanding input at the
3801 point when the greeting is written. Because of the duplex, asynchronous
3802 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3803 way, but not yet received, when the check is performed.
3805 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3806 on TCP/IP sockets, because this apparently causes some broken clients to
3809 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3810 unchanged) from the Cygwin maintainer.
3812 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3813 was supported instead of testing for mbx. Effectively a typo.
3815 15. The spa authenticator server code was not checking that the input it
3816 received was valid base64.
3818 16. The debug output line for the "set" modifier in ACLs was not showing the
3819 name of the variable that was being set.
3821 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3823 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3824 lookup on its own. Something else was needed to trigger the lookup. For
3825 example, a match in host_lookup or the need for a host name in a host list.
3826 Now, if $sender_host_name is referenced and the host name has not yet been
3827 looked up, a lookup is performed. If the lookup fails, the variable remains
3828 empty, and $host_lookup_failed is set to "1".
3830 19. Added "eqi" as a case-independent comparison operator.
3832 20. The saslauthd authentication condition could segfault if neither service
3833 nor realm was specified.
3835 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3836 error message that was logged was misleading, and incoming SMTP
3837 connections were dropped. The message is now more accurate, and temporary
3838 errors are given to SMTP connections.
3840 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3841 (except RSET) with a 421 error, until QUIT is received. However, it was
3842 failing to send a response to QUIT.
3844 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3845 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3846 ACL is now run after the helo_try_verify_hosts code.
3848 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3849 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3850 case-independent, so other case variants are also recognized). Apparently
3851 some systems use these upper case variants.
3853 25. If more than two messages were waiting for the same host, and a transport
3854 filter was specified for the transport, Exim sent two messages over the
3855 same TCP/IP connection, and then failed with "socket operation on non-
3856 socket" when it tried to send the third.
3858 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3860 27. The extern definition of crypt16() in expand.c was not being excluded when
3861 the OS had its own crypt16() function.
3863 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3864 as a preferred synonym for return_size_limit, both as an option and as an
3867 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3869 30. Changed the default configuration ACL to relax the local part checking rule
3870 for addresses that are not in any local domains. For these addresses,
3871 slashes and pipe symbols are allowed within local parts, but the sequence
3872 /../ is explicitly forbidden.
3874 31. SPA server authentication was not clearing the challenge buffer before
3877 32. log_message in a "warn" ACL statement was writing to the reject log as
3878 well as to the main log, which contradicts the documentation and doesn't
3879 seem right (because no rejection is happening). So I have stopped it.
3881 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3882 However, I am unable to do any testing of this.
3884 34. Fixed an infelicity in the appendfile transport. When checking directories
3885 for a mailbox, to see if any needed to be created, it was accidentally
3886 using path names with one or more superfluous leading slashes; tracing
3887 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3889 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3890 contents are added to the log line that is written for every discarded
3891 recipient. (Previously a log_message setting was ignored.)
3893 36. The ${quote: operator now quotes the string if it is empty.
3895 37. The install script runs exim in order to find its version number. If for
3896 some reason other than non-existence or emptiness, which it checks, it
3897 could not run './exim', it was installing it with an empty version number,
3898 i.e. as "exim-". This error state is now caught, and the installation is
3901 38. An argument was missing from the function that creates an error message
3902 when Exim fails to connect to the socket for saslauthd authentication.
3903 This could cause Exim to crash, or give a corrupted message.
3905 39. Added isip, isip4, and isip6 to ${if conditions.
3907 40. The ACL variables $acl_xx are now saved with the message, and can be
3908 accessed later in routers, transports, and filters.
3910 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3911 strings in the file are not string-expanded.
3913 42. If a MAIL command specified a SIZE value that was too large to fit into an
3914 int variable, the check against message_size_limit failed. Such values are
3915 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3916 day this will have to be increased, but I don't think I want to be around
3917 when emails are that large.
3924 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3925 iconv() is not standard in FreeBSD.
3927 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3928 IPv6 enabled. The observed symptom was a segmentation fault on return from
3929 the function os_common_find_running_interfaces() in src/os.c.
3931 3. In the check_special_case() function in daemon.c I had used "errno" as an
3932 argument name, which causes warnings on some systems. This was basically a
3933 typo, since it was named "eno" in the comments!
3935 4. The code that waits for the clock to tick (at a resolution of some fraction
3936 of a second) so as to ensure message-id uniqueness was always waiting for
3937 at least one whole tick, when it could have waited for less. [This is
3938 almost certainly not relevant at current processor speeds, where it is
3939 unlikely to ever wait at all. But we try to future-proof.]
3941 5. The function that sleeps for a time interval that includes fractions of a
3942 second contained a race. It did not block SIGALRM between setting the
3943 timer, and suspending (a couple of lines later). If the interval was short
3944 and the sigsuspend() was delayed until after it had expired, the suspension
3945 never ended. On busy systems this could lead to processes getting stuck for
3948 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3949 process, before it forks any delivery processes. The open lookup caching
3950 mechanism meant that the open file or database connection was passed into
3951 the delivery process. The problem was that delivery processes always tidy
3952 up cached lookup data. This could cause a problem for the next delivery
3953 process started by the queue runner, because the external queue runner
3954 process does not know about the closure. So the next delivery process
3955 still has data in the lookup cache. In the case of a file lookup, there was
3956 no problem because closing a file descriptor in a subprocess doesn't affect
3957 the parent. However, if the lookup was caching a connection to a database,
3958 the connection was closed, and the second delivery process was likely to
3959 see errors such as "PGSQL: query failed: server closed the connection
3960 unexpectedly". The problem has been fixed by closing all cached lookups
3961 in a queue runner before running a delivery process.
3963 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3964 seem to have the "const" qualifier which it has on other OS. I've
3967 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3968 *of the same type* (client or server) with the same public name that an
3969 error should be diagnosed.
3971 9. When Exim looked up a host name for an IP address, but failed to find the
3972 original IP address when looking up the host name (a safety check), it
3973 output the message "<ip address> does not match any IP for NULL", which was
3974 confusing, to say the least. The bug was that the host name should have
3975 appeared instead of "NULL".
3977 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3978 user that is built into the binary, and the -C or -D options is used, root
3979 privilege is dropped before the configuration file is read. In addition,
3980 logging is switched to stderr instead of the normal log files. If the
3981 configuration then re-defines the Exim user, the unprivileged environment
3982 is probably not what is expected, so Exim logs a panic warning message (but
3985 However, if deliver_drop_privilege is set, the unprivileged state may well
3986 be exactly what is intended, so the warning has been cut out in that case,
3987 and Exim is allowed to try to write to its normal log files.
3993 1. smtp_return_error_details was not giving details for temporary sender
3994 or receiver verification errors.
3996 2. Diagnose a configuration error if two authenticators have the same public
3999 3. Exim used not to create the message log file for a message until the first
4000 delivery attempt. This could be confusing when incoming messages were held
4001 for policy or load reasons. The message log file is now created at the time
4002 the message is received, and an initial "Received" line is written to it.
4004 4. The automatically generated man page for command line options had a minor
4005 bug that caused no ill effects; however, a more serious problem was that
4006 the procedure for building the man page automatically didn't always
4007 operate. Consequently, release 4.20 contains an out-of-date version. This
4008 shouldn't happen again.
4010 5. When building Exim with embedded Perl support, the script that builds the
4011 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4012 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4014 6. The freeze_tell option was not being used for messages that were frozen on
4015 arrival, either by an ACL or by local_scan().
4017 7. Added the smtp_incomplete_transaction log selector.
4019 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4020 was accepting AUTH without a new EHLO.
4022 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4023 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4026 10. Logging of TCP/IP connections (when configured) now happens in the main
4027 daemon process instead of the child process, so that the TCP/IP connection
4028 count is more accurate (but it can never be perfect).
4030 11. The use of "drop" in a nested ACL was not being handled correctly in the
4031 outer ACL. Now, if condition failure induced by the nested "drop" causes
4032 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4033 or "require"), the connection is dropped.
4035 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4036 that yield "discard" can now be used with an "accept" or a "discard" verb,
4037 but an error is generated for any others (because I can't see a useful way
4038 to define what should happen).
4040 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4041 are now processed in the same way as lines in the Exim configuration file.
4042 In particular, continuation lines are supported.
4044 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4046 15. Added -ti meaning -t -i.
4048 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4049 domains, and warn by logging if others are found.
4051 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4052 returned from GIFCONF when Exim is trying to find the list of interfaces on
4053 a host. The code in os.c has been modified to copy each ifreq to an aligned
4054 structure in all cases.
4056 Also, in some cases, the returned ifreq's were being copied to a 'struct
4057 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4058 means the last couple of bytes of an IPv6 address could be chopped if the
4059 ifreq contained only a normal sockaddr (14 bytes storage).
4061 18. Named domain lists were not supported in the hosts_treat_as_local option.
4062 An entry such as +xxxx was not recognized, and was treated as a literal
4065 19. Ensure that header lines added by a DATA ACL are included in the reject log
4066 if the ACL subsequently rejects the message.
4068 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4069 MD5 (which is deprecated).
4071 21. When testing a filter file using -bf, Exim was writing a message when it
4072 took the sender from a "From " line in the message, but it was not doing so
4073 when it took $return_path from a Return-Path: header line. It now does.
4075 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4076 with a valid header line field name (a series of printing characters
4077 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4079 23. Changed "disc" in the source to "disk" to conform to the documentation and
4080 the book and for uniformity.
4082 24. Ignore Sendmail's -Ooption=value command line item.
4084 25. When execve() failed while trying to run a command in a pipe transport,
4085 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4086 could be confused with a return value of 69 from the command itself. This
4087 has been changed to 127, the value the shell returns if it is asked to run
4088 a non-existent command. The wording for the related log line suggests a
4089 non-existent command as the problem.
4091 26. If received_header_text expands to an empty string, do not add a Received:
4092 header line to the message. (Well, it adds a token one on the spool, but
4093 marks it "old" so that it doesn't get used or transmitted.)
4095 27. Installed eximstats 1.28 (addition of -nt option).
4097 28. There was no check for failure on the call to getsockname() in the daemon
4098 code. This can fail if there is a shortage of resources on the system, with
4099 ENOMEM, for example. A temporary error is now given on failure.
4101 29. Contrary to the C standard, it seems that in some environments, the
4102 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4103 program. Exim now does this explicitly; it affects the formatting of
4104 timestamps using strftime().
4106 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4107 complaints. I've now initialized all the variables, to avoid this.
4109 32. Header lines added by a system filter were not being "seen" during
4110 transport-time rewrites.
4112 33. The info_callback() function passed to OpenSSL is set up with type void
4113 (*)(SSL *, int, int), as described somewhere. However, when calling the
4114 function (actually a macro) that sets it up, the type void(*)() is
4115 expected. I've put in a cast to prevent warnings from picky compilers.
4117 34. If a DNS black list lookup found a CNAME record, but there were no A
4118 records associated with the domain it pointed at, Exim crashed.
4120 35. If a DNS black list lookup returned more than one A record, Exim ignored
4121 all but the first. It now scans all returned addresses if a particular IP
4122 value is being sought. In this situation, the contents of the
4123 $dnslist_value variable are a list of all the addresses, separated by a
4126 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4127 to accept a host name and all its aliases if the forward lookup for any of
4128 them yielded the IP address of the incoming connection. Now it accepts only
4129 those names whose forward lookup yields the correct IP address. Any other
4130 names are discarded. This closes a loophole whereby a rogue DNS
4131 administrator could create reverse DNS records to break through a
4132 wildcarded host restriction in an ACL.
4134 37. If a user filter or a system filter that ran in a subprocess used any of
4135 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4136 the wrong values were passed to the pipe command ($thisaddress had the
4137 value of $0, $0 had the value of $1, etc). This bug was introduced by
4138 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4140 38. Improved the line breaking for long SMTP error messages from ACLs.
4141 Previously, if there was no break point between 40 and 75 characters, Exim
4142 left the rest of the message alone. Two changes have been made: (a) I've
4143 reduced the minimum length to 35 characters; (b) if it can't find a break
4144 point between 35 and 75 characters, it looks ahead and uses the first one
4145 that it finds. This may give the occasional overlong line, but at least the
4146 remaining text gets split now.
4148 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4149 file descriptors might be low, and that setting 1000 would always raise it.
4150 It turns out that in some environments, the limit is already over 1000 and
4151 that lowering it causes trouble. So now Exim takes care not to decrease it.
4153 40. When delivering a message, the value of $return_path is set to $sender_
4154 address at the start of routing (routers may change the value). By an
4155 oversight, this default was not being set up when an address was tested by
4156 -bt or -bv, which affected the outcome if any router or filter referred to
4159 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4160 log only when "message" or "log_message" are set. However, if one of the
4161 conditions was an address verification, or a call to a nested ACL, the
4162 messages generated by the underlying test were being passed through. This
4163 no longer happens. The underlying message is available in $acl_verify_
4164 message for both "message" and "log_message" expansions, so it can be
4165 passed through if needed.
4167 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4168 new expansion $bh_ to give the encoded byte string without charset
4169 translation. Translation happens only if iconv() is available; HAVE_ICONV
4170 indicates this at build time. HEADERS_CHARSET gives the charset to
4171 translate to; headers_charset can change it in the configuration, and
4172 "headers charset" can change it in an individual filter file.
4174 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4175 that creates RFC 2047 encoded "words" labels them as that charset instead
4176 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4177 expansion operator; (ii) when Exim creates a From: line for a local
4178 message; (iii) when a header line is rewritten to include a "phrase" part.
4180 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4181 buggy, causing it to skip the first lines of messages whose message ID
4182 ended in 'D'. This would not have bitten before Exim release 4.14, because
4183 message IDs were unlikely to end in 'D' before then. The effect was to have
4184 incorrect size information for certain domains.
4186 45. #include "config.h" was missing at the start of the crypt16.c module. This
4187 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4190 46. If there was a timeout during a "random" callout check, Exim treated it as
4191 a failure of the random address, and carried on sending RSET and the real
4192 address. If the delay was just some slowness somewhere, the response to the
4193 original RCPT would be taken as a response to RSET and so on, causing
4194 mayhem of various kinds.
4196 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4197 when I implemented it. It didn't allow for the fact that some option values
4198 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4199 the right test for positive values.
4201 48. Domain names in DNS records are case-independent. Exim always looks them up
4202 in lower case. Some resolvers return domain names in exactly the case they
4203 appear in the zone file, that is, they may contain uppercase letters. Not
4204 all resolvers do this - some return always lower case. Exim was treating a
4205 change of case by a resolver as a change of domain, similar to a widening
4206 of a domain abbreviation. This triggered its re-routing code and so it was
4207 trying to route what was effectively the same domain again. This normally
4208 caused routing to fail (because the router wouldn't handle the domain
4209 twice). Now Exim checks for this case specially, and just changes the
4210 casing of the domain that it ultimately uses when it transmits the message
4213 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4216 50. If a filter generated a file delivery with a non-absolute name (possible if
4217 no home directory exists for the router), the forbid_file option was not
4220 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4221 the existing equality matching.
4223 52. Exim was using ints instead of ino_t variables in some places where it was
4224 dealing with inode numbers.
4226 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4227 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4228 called TMPDIR, and if it finds it is different, it changes its value.
4230 54. The smtp_printf() function is now made available to local_scan() so
4231 additional output lines can be written before returning. There is also an
4232 smtp_fflush() function to enable the detection of a dropped connection.
4233 The variables smtp_input and smtp_batched_input are exported to
4236 55. Changed the default runtime configuration: the message "Unknown user"
4237 has been removed from the ACL, and instead placed on the localuser router,
4238 using the cannot_route_message feature. This means that any verification
4239 failures that generate their own messages won't get overridden. Similarly,
4240 the "Unrouteable address" message that was in the ACL for unverifiable
4241 relay addresses has also been removed.
4243 56. Added hosts_avoid_esmtp to the smtp transport.
4245 57. The exicyclog script was not checking for the esoteric option
4246 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4247 will work only if exicyclog is run under the appropriate euid.
4249 58. Following a discussion on the list, the rules by which Exim recognises line
4250 endings on incoming messages have been changed. The -dropcr and drop_cr
4251 options are now no-ops, retained only for backwards compatibility. The
4252 following line terminators are recognized: LF CRLF CR. However, special
4253 processing applies to CR:
4255 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4256 nor a local message in the state where . is a terminator.
4258 (ii) If a bare CR is encountered in a header line, an extra space is added
4259 after the line terminator so as not to end the header. The reasoning
4260 behind this is that bare CRs in header lines are most likely either
4261 to be mistakes, or people trying to play silly games.
4263 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4264 was being incorrectly given as 18 bytes larger than it should have been.
4265 This is a VOB (very old bug).
4267 60. This may never have affected anything current, but just in case it has:
4268 When the local host is found other than at the start of a list of hosts,
4269 the local host, those with the same MX, and any that follow, are discarded.
4270 When the list in question was part of a longer list of hosts, the following
4271 hosts (not currently being processed) were also being discarded. This no
4272 longer happens. I'm not sure if this situation could ever has previously
4275 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4278 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4279 header. This is something that is recommended in a new Internet Draft, and
4280 is something that is documented as being done by Sendmail. There are two
4281 possible values. For messages generated by the autoreply transport, Exim
4284 Auto-Submitted: auto-replied
4286 whereas for all other generated messages (e.g. bounces) it adds
4288 Auto-Submitted: auto-generated
4290 63. The "personal" condition in filters now includes a test for the
4291 Auto-Submitted: header. If it contains the string "auto-" the message it
4292 not considered personal.
4294 64. Added rcpt_include_affixes as a generic transport option.
4296 65. Added queue_only_override (default true).
4298 66. Added the syslog_duplication option.
4300 67. If what should have been the first header line of a message consisted of
4301 a space followed by a colon, Exim was mis-interpreting it as a header line.
4302 It isn't of course - it is syntactically invalid and should therefore be
4303 treated as the start of the message body. The misbehaviour could have
4304 caused a number of strange effects, including loss of data in subsequent
4305 header lines, and spool format errors.
4307 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4308 client host had authenticated. This control can now be exercised by an ACL
4309 for more flexibility.
4311 69. By default, callouts do not happen when testing with -bh. There is now a
4312 variant, -bhc, which does actually run the callout code, including
4313 consulting and updating the callout cache.
4315 70. Added support for saslauthd authentication, courtesy of Alexander
4318 71. If statvfs() failed on the spool or log directories while checking their
4319 size for availability, Exim confusingly gave the error "space shortage".
4320 Furthermore, in debugging mode it crashed with a floating point exception.
4321 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4322 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4323 serious problem, Exim now writes to the main and panic logs when this
4324 happens, with details of the failure. It then refuses to accept the
4325 incoming message, giving the message "spool directory problem" or "log
4326 directory problem" with a 421 code for SMTP messages.
4328 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4329 0, 1, and 2 exist, because some OS complain for execs without them (see
4330 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4331 descriptors. However, the code omitted to check that the open succeeded,
4332 causing mysterious errors if for some reason the permissions on /dev/null
4333 got screwed. Now Exim writes a message to the main and panic logs, and
4334 bombs out if it can't open /dev/null.
4336 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4337 interact so that it is all more flexible. It is supposed to remain
4338 backwards compatible. Also added extra_local_interfaces.
4340 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4341 to bomb out with an assertion failure - to the client this appears as a
4342 connection drop. This problem occurs in the part of the code that was taken
4343 from the Samba project. Fortunately, the assertion is in a very simple
4344 function, so I have fixed this by reproducing the function inline in the
4345 one place where it is called, and arranging for authentication to fail
4346 instead of killing the process with assert().
4348 75. The SPA client code was not working when the server requested OEM rather
4349 than Unicode encoding.
4351 76. Added code to make require_files with a specific uid setting more usable in
4352 the case where statting the file as root fails - usually a non-root-mounted
4353 NFS file system. When this happens and the failure is EACCES, Exim now
4354 forks a subprocess and does the per-uid checking as the relevant uid.
4356 77. Added process_log_path.
4358 78. If log_file_path was not explicitly set, a setting of check_log_space or
4359 check_log_inodes was ignored.
4361 79. If a space check for the spool or log partitions fails, the incident is now
4362 logged. Of course, in the latter case the data may get lost...
4364 80. Added the %p formatting code to string_format() so that it can be used to
4365 print addresses in debug_print(). Adjusted all the address printing in the
4366 debugging in store.c to use %p rather than %d.
4368 81. There was a concern that a line of code in smtp_in.c could overflow a
4369 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4370 initially expressed, the concern was not well-founded, because trailing
4371 spaces are removed early. However, if the trailing spaces were followed by
4372 a NULL, they did not get removed, so the overflow was possible. Two fixes
4375 (a) I re-wrote the offending code in a cleaner fashion.
4376 (b) If an incoming SMTP command contains a NULL character, it is rejected
4379 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4380 runs initgroups(), so that if the Exim user is in any additional groups,
4381 they will be used during message reception.
4387 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob