1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.426 2006/11/07 11:28:05 ph10 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
8 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
9 leftover -K file (the existence of which was triggered by #402).
10 While we were at it, introduced process PID as part of the -K
11 filename. This should rule out race conditions when creating
14 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
15 processing considerably. Previous code took too long for large mails,
16 triggering a timeout which in turn triggers #401.
18 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
19 in the DK code in transports.c. sendfile() is not really portable,
20 hence the _LINUX specificness.
22 TF/01 In the add_headers option to the mail command in an Exim filter,
23 there was a bug that Exim would claim a syntax error in any
24 header after the first one which had an odd number of characters
27 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
28 callout verification, Exim cached a "reject" for the entire domain. This
29 is correct for most verifications, but it is not correct for a recipient
30 verification with use_sender or use_postmaster set, because in that case
31 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
32 case of MAIL FROM:<> rejection from other early rejections (e.g.
33 rejection of HELO). When verifying a recipient using a non-null MAIL
34 address, the cache is ignored if it shows MAIL FROM:<> rejection.
35 Whatever the result of the callout, the value of the domain cache is
36 left unchanged (for any other kind of callout, getting as far as trying
37 RCPT means that the domain itself is ok).
39 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
42 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
43 manifest itself as EPIPE rather than ECONNECT. When tidying away a
44 session, the daemon ignores ECONNECT errors and logs others; it now
45 ignores EPIPE as well.
47 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
48 (quoted-printable decoding).
50 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
51 later the small subsequent patch to fix an introduced bug.
53 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
55 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
57 PH/08 An error is now given if message_size_limit is specified negative.
59 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
60 to be given (somewhat) arbitrary names.
62 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
65 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
66 miscellaneous code fixes
68 PH/10 Added the log_reject_target ACL modifier to specify where to log
71 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
72 hostname. This is wrong, because it relates to the incoming message (and
73 probably the interface on which it is arriving) and not to the outgoing
74 callout (which could be using a different interface). This has been
75 changed to use the value of the helo_data option from the smtp transport
76 instead - this is what is used when a message is actually being sent. If
77 there is no remote transport (possible with a router that sets up host
78 addresses), $smtp_active_hostname is used.
80 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
81 tweaks were necessary in order to get it to work (see also 21 below):
82 (a) The code assumed that strncpy() returns a negative number on buffer
83 overflow, which isn't the case. Replaced with Exim's string_format()
85 (b) There were several signed/unsigned issues. I just did the minimum
86 hacking in of casts. There is scope for a larger refactoring.
87 (c) The code used strcasecmp() which is not a standard C function.
88 Replaced with Exim's strcmpic() function.
89 (d) The code set only $1; it now sets $auth1 as well.
90 (e) A simple test gave the error "authentication client didn't specify
91 service in request". It would seem that Dovecot has changed its
92 interface. Fortunately there's a specification; I followed it and
93 changed what the client sends and it appears to be working now.
95 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
98 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
99 address A is aliased to B and C, where B exists and C does not. Without
100 -v the output is "A verified" because verification stops after a
101 successful redirection if more than one address is generated. However,
102 with -v the child addresses are also verified. Exim was outputting "A
103 failed to verify" and then showing the successful verification for C,
104 with its parentage. It now outputs "B failed to verify", showing B's
105 parentage before showing the successful verification of C.
107 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
108 look up a TXT record in a specific list after matching in a combined
111 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
112 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
113 they consult the DNS. I had assumed they would set it the way they
114 wanted; and indeed my experiments on Linux seem to show that in some
115 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
116 To be on the safe side, however, I have now made the interface to
117 host_find_byname() similar to host_find_bydns(), with an argument
118 containing the DNS resolver options. The host_find_byname() function now
119 sets these options at its start, just as host_find_bydns() does. The smtp
120 transport options dns_qualify_single and dns_search_parents are passed to
121 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
122 of host_find_byname() use the default settings of RES_DEFNAMES
123 (qualify_single) but not RES_DNSRCH (search_parents).
125 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
126 spool_read_header() do less string testing, by means of a preliminary
127 switch on the second character of optional "-foo" lines. (This is
128 overdue, caused by the large number of possibilities that now exist.
129 Originally there were few.) While I was there, I also converted the
130 str(n)cmp tests so they don't re-test the leading "-" and the first
131 character, in the hope this might squeeze out yet more improvement.
133 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
134 flag allowing group syntax was set by the header_syntax check but not
135 turned off, possible causing trouble later; (2) The flag was not being
136 set at all for the header_verify test, causing "group"-style headers to
137 be rejected. I have now set it in this case, and also caused header_
138 verify to ignore an empty address taken from a group. While doing this, I
139 came across some other cases where the code for allowing group syntax
140 while scanning a header line wasn't quite right (mostly, not resetting
141 the flag correctly in the right place). These bugs could have caused
142 trouble for malformed header lines. I hope it is now all correct.
144 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
145 with the "reply" argument non-NULL. The code, however (which originally
146 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
147 but it didn't always do it. This confused somebody who was copying the
148 code for some other use. I have removed all the tests.
150 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
151 feature that was used to support insecure browsers during the U.S. crypto
152 embargo. It requires special client support, and Exim is probably the
153 only MTA that supported it -- and would never use it because real RSA is
154 always available. This code has been removed, because it had the bad
155 effect of slowing Exim down by computing (never used) parameters for the
156 RSA_EXPORT functionality.
158 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
159 authenticator to fail if there's a tab character in the incoming data
160 (there should never be unless someone is messing about, as it's supposed
161 to be base64-encoded). Also added, on Timo's advice, the "secured" option
162 if the connection is using TLS or if the remote IP is the same as the
163 local IP, and the "valid-client-cert option" if a client certificate has
166 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
167 authenticators. This can be used for authorization after authentication
168 succeeds. (In the case of plaintext, it servers for both authentication
171 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
172 if any retry times were supplied.
174 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
175 connection, where there is no client IP address to check. In this
176 situation, the verify now always succeeds.
178 PH/25 Applied John Jetmore's -Mset patch.
180 PH/26 Added -bem to be like -Mset, but loading a message from a file.
182 PH/27 In a string expansion for a processed (not raw) header when multiple
183 headers of the same name were present, leading whitespace was being
184 removed from all of them, but trailing whitespace was being removed only
185 from the last one. Now trailing whitespace is removed from each header
186 before concatenation. Completely empty headers in a concatenation (as
189 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
190 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
192 PH/29 [Removed. This was a change that I later backed out, and forgot to
193 correct the ChangeLog entry (that I had efficiently created) before
194 committing the later change.]
196 PH/30 Exim was sometimes attempting to deliver messages that had suffered
197 address errors (4xx response to RCPT) over the same connection as other
198 messages routed to the same hosts. Such deliveries are always "forced",
199 so retry times are not inspected. This resulted in far too many retries
200 for the affected addresses. The effect occurred only when there were more
201 hosts than the hosts_max_try setting in the smtp transport when it had
202 the 4xx errors. Those hosts that it had tried were not added to the list
203 of hosts for which the message was waiting, so if all were tried, there
204 was no problem. Two fixes have been applied:
206 (i) If there are any address or message errors in an SMTP delivery, none
207 of the hosts (tried or untried) are now added to the list of hosts
208 for which the message is waiting, so the message should not be a
209 candidate for sending over the same connection that was used for a
210 successful delivery of some other message. This seems entirely
211 reasonable: after all the message is NOT "waiting for some host".
212 This is so "obvious" that I'm not sure why it wasn't done
213 previously. Hope I haven't missed anything, but it can't do any
214 harm, as the worst effect is to miss an optimization.
216 (ii) If, despite (i), such a delivery is accidentally attempted, the
217 routing retry time is respected, so at least it doesn't keep
218 hammering the server.
220 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
221 in ${readsocket because some servers need this prod.
223 PH/32 Added some extra debug output when updating a wait-xxx database.
225 PH/33 The hint "could be header name not terminated by colon", which has been
226 given for certain expansion errors for a long time, was not being given
227 for the ${if def:h_colon_omitted{... case.
229 PH/34 The spec says: "With one important exception, whenever a domain list is
230 being scanned, $domain contains the subject domain." There was at least
231 one case where this was not true.
233 PH/35 The error "getsockname() failed: connection reset by peer" was being
234 written to the panic log as well as the main log, but it isn't really
235 panic-worthy as it just means the connection died rather early on. I have
236 removed the panic log writing for the ECONNRESET error when getsockname()
239 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
240 runs only) independently of the message's sender address. This meant
241 that, if the 4xx error was in fact related to the sender, a different
242 message to the same recipient with a different sender could confuse
243 things. In particualar, this can happen when sending to a greylisting
244 server, but other circumstances could also provoke similar problems.
245 I have changed the default so that the retry time for these errors is now
246 based a combination of the sender and recipient addresses. This change
247 can be overridden by setting address_retry_include_sender=false in the
250 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
251 remote server are returned as part of bounce messages. This was not
252 happening for LMTP over a pipe (the lmtp transport), but now it is the
253 same for both kinds of LMTP.
259 SC/01 Use a glob alias rather than an array ref in eximstats generated
260 parser. This improves both readability and performance.
262 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
263 Don't display local sender or destination tables in eximstats unless
264 there is data to show.
265 Added average volumes into the eximstats top table text output.
267 SC/03 Collect data on the number of addresses (recipients) as well
268 as the number of messages in eximstats.
270 TF/01 Correct an error in the documentation for the redirect router. Exim
271 does (usually) call initgroups() when daemonizing.
273 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
274 with consistent privilege compared to when running as a daemon.
276 TF/03 Note in the spec that $authenticated_id is not set for local
277 submissions from trusted users.
279 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
280 Thanks to Dean Brooks <dean@iglou.com> for the patch.
282 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
283 by adding some example configuration directives to the default
284 configuration file. A little bit of work is required to uncomment the
285 directives and define how usernames and passwords are checked, but
286 there is now a framework to start from.
288 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
289 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
290 without this. I don't know how relevant this is to other LDAP libraries.
292 PH/02 Add the verb name to the "unknown ACL verb" error.
294 PH/03 Magnus Holmgren's patch for filter_prepend_home.
296 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
298 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
299 directory not expanded when it should be if an expanded home directory
300 was set for the address (which is overridden by the transport).
302 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
305 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
306 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
307 because it is too late at that time, and has no effect.
309 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
310 security issue with \' (bugzilla #107). I could not use the
311 PQescapeStringConn() function, because it needs a PGconn value as one of
314 PH/08 When testing addresses using -bt, indicate those final addresses that
315 are duplicates that would not cause an additional delivery. At least one
316 person was confused, thinking that -bt output corresponded to deliveries.
317 (Suppressing duplicates isn't a good idea as you lose the information
318 about possibly different redirections that led to the duplicates.)
320 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
321 systems where poll() doesn't work, in particular OS X.
323 PH/10 Added more information to debugging output for retry time not reached.
325 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
326 operations in malware.c.
328 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
331 PH/13 If write_rejectlog was set false when logging was sent to syslog with
332 syslog_duplication set false, log lines that would normally be written
333 both the the main log and to the reject log were not written to syslog at
336 PH/14 In the default configuration, change the use of "message" in ACL warn
337 statements to "add_header".
339 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
340 not followed by a command (e.g. "seen endif").
342 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
343 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
346 PH/17 Added extra conditions to the default value of delay_warning_condition
350 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
351 { match{$h_precedence:}{(?i)bulk|list|junk} } \
352 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
355 The Auto-Submitted: and various List- headers are standardised, whereas I
356 don't think Precedence: ever was.
358 PH/18 Refactored debugging code in route_finduser() to show more information,
359 in particular, the error code if getpwnam() issues one.
361 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
362 This is apparently needed in addition to the PH/07 change above to avoid
363 any possible encoding problems.
365 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
366 but not after initializing Perl.
368 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
369 output them only if debugging. By default they are written stderr,
370 apparently, which is not desirable.
372 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
375 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
378 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
380 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
381 authenticated or an ident call has been made. Suppress the default
382 values for $authenticated_id and $authenticated_sender (but permit -oMai
383 and -oMas) when testing with -bh.
385 PH/24 Re-jigged the order of the tests in the default configuration so that the
386 tests for valid domains and recipients precede the DNS black list and CSA
387 tests, on the grounds that those ones are more expensive.
389 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
390 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
391 This bug exists in every version of Exim that I still have, right back to
394 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
395 However, an attempt to turn on case-sensitivity in a regex key by
396 including (?-i) didn't work because the subject string was already
397 lowercased, and the effects were non-intuitive. It turns out that a
398 one-line patch can be used to allow (?-i) to work as expected.
404 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
405 other effects) broke the use of negated acl sub-conditions.
407 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
410 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
411 "Deny" causes Exim to reject the incoming connection with a 554 error.
412 Unfortunately, if there is a major crisis, such as a disk failure,
413 tcp-wrappers gives "deny", whereas what one would like would be some
414 kind of temporary error. A kludge has been added to help with this.
415 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
416 554 error is used if errno is still zero or contains ENOENT (which occurs
417 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
420 PH/03 Add -lutil to the default FreeBSD LIBS setting.
422 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
423 errors. Otherwise a message that provokes a temporary error (when other
424 messages do not) can cause a whole host to time out.
426 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
427 addresses were routed directly to files or pipes from a redirect router.
428 File deliveries just didn't batch; pipe deliveries might have suffered
431 PH/06 A failure to get a lock for a hints database would erroneously always say
432 "Failed to get write lock", even when it was really a read lock.
434 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
435 of 0600. This has been changed to use the value of the lockfile_mode
436 option (which defaults to 0600).
438 PH/08 Applied small patch from the Sieve maintainer.
440 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
441 folder from quota calculations, a direct delivery into this folder messed
442 up the contents of the maildirsize file. This was because the regex was
443 used only to exclude .Trash (or whatever) when the size of the mailbox
444 was calculated. There was no check that a delivery was happening into an
445 excluded directory. This bug has been fixed by ignoring all quota
446 processing for deliveries into excluded directories.
448 PH/10 Added the maildirfolder_create_regex option to appendfile.
454 PH/01 The code for finding all the local interface addresses on a FreeBSD
455 system running IPv6 was broken. This may well have applied to all BSD
456 systems, as well as to others that have similar system calls. The broken
457 code found IPv4 interfaces correctly, but gave incorrect values for the
458 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
459 that it would not match correctly against @[] and not recognize the IPv6
462 PH/02 The ipliteral router was not recognizing addresses of the form user@
463 [ipv6:....] because it didn't know about the "ipv6:" prefix.
465 PH/03 Added disable_ipv6.
467 PH/04 Changed $reply_address to use the raw form of the headers instead of the
468 decoded form, because it is most often used to construct To: headers
469 lines in autoreplies, and the decoded form may well be syntactically
470 invalid. However, $reply_address has leading white space removed, and all
471 newlines turned into spaces so that the autoreply transport does not
474 PH/05 If group was specified without a user on a router, and no group or user
475 was specified on a transport, the group from the router was ignored.
477 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
478 for visible compile-time settings that can be used to change these
479 numbers, for those that want even more. Backwards compatibility with old
480 spool files has been maintained. However, going back to a previous Exim
481 release will lost any variables that are in spool files.
483 PH/07 Two small changes when running in the test harness: increase delay when
484 passing a TCP/IP connection to a new process, in case the original
485 process has to generate a bounce, and remove special handling of
486 127.0.0.2 (sic), which is no longer necessary.
488 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
489 be the same on different OS.
491 PH/09 Moved a debug statement in filter processing to avoid a race problem when
494 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
495 whether --show-vars was specified or not
497 JJ/02 exipick: Added support for new ACL variable spool format introduced
500 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
501 syntactically invalid From: or Reply-to: line, and a filter used this to
502 generate an autoreply, and therefore failed to obtain an address for the
503 autoreply, Exim could try to deliver to a non-existent relative file
504 name, causing unrelated and misleading errors. What now happens is that
505 it logs this as a hard delivery error, but does not attempt to create a
508 PH/11 The exinext utility has a -C option for testing purposes, but although
509 the given file was scanned by exinext itself; it wasn't being passed on
510 when Exim was called.
512 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
513 an end-of-file indication when reading a command response.
515 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
516 compiled. In many other places in Exim, IPv6 addresses are always
517 recognized, so I have changed this. It also means that IPv4 domain
518 literals of the form [IPV4:n.n.n.n] are now always recognized.
520 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
521 used if the router is not running as root, for example, when verifying at
522 ACL time, or when using -bh. The debugging output from this situation was
523 non-existent - all you got was a failure to exec. I have made two
526 (a) Failures to set uid/gid, the current directory, or a process leader
527 in a subprocess such as that created by queryprogram now generate
528 suitable debugging ouput when -d is set.
530 (b) The queryprogram router detects when it is not running as root,
531 outputs suitable debugging information if -d is set, and then runs
532 the subprocess without attempting to change uid/gid.
534 PH/15 Minor change to Makefile for building test_host (undocumented testing
537 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
538 additional section of a DNS packet that returns MX or SRV records.
539 Instead, it always explicitly searches for A/AAAA records. This avoids
540 major problems that occur when a DNS server includes only records of one
541 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
542 fixed another bug: if SRV records were looked up and the corresponding
543 address records were *not* found in the additional section, the port
544 values from the SRV records were lost.
546 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
547 using the correct key (the original address) when searching the retry
548 rules in order to find which one to use for generating the retry hint.
550 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
551 adding the default one. Similarly, if it contains a Reply-To: header, the
552 errors_reply_to option, if set, is not used.
554 PH/19 When calculating a retry time, Exim used to measure the "time since
555 failure" by looking at the "first failed" field in the retry record. Now
556 it does not use this if it is later than than the arrival time of the
557 message. Instead it uses the arrival time. This makes for better
558 behaviour in cases where some deliveries succeed, thus re-setting the
559 "first failed" field. An example is a quota failure for a huge message
560 when small messages continue to be delivered. Without this change, the
561 "time since failure" will always be short, possible causing more frequent
562 delivery attempts for the huge message than are intended.
563 [Note: This change was subsequently modified - see PH/04 for 4.62.]
565 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
566 $1, $2, $3) because the numerical variables can be reset during some
567 expansion items (e.g. "match"), thereby losing the authentication data.
569 PH/21 Make -bV show the size of off_t variables so that the test suite can
570 decide whether to run tests for quotas > 2G.
572 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
573 mailbox_size, and mailbox_filecount in the appendfile transport. If a
574 filecount value is greater than 2G or if a quota value is greater than 2G
575 on a system where the size of off_t is not greater than 4, a panic error
578 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
579 never match. The debug and -bh output now contains an explicit error
580 message indicating a malformed IPv4 address or mask.
582 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
583 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
586 PH/25 Do not write to syslog when running in the test harness. The only
587 occasion when this arises is a failure to open the main or panic logs
588 (for which there is an explicit test).
590 PH/26 Added the /no_tell option to "control=freeze".
592 PH/27 If a host name lookup failed very early in a connection, for example, if
593 the IP address matched host_lookup and the reverse lookup yielded a name
594 that did not have a forward lookup, an error message of the form "no IP
595 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
596 could be logged. Now it outputs the IP address instead of "NULL".
598 PH/28 An enabling patch from MH: add new function child_open_exim2() which
599 allows the sender and the authenticated sender to be set when
600 submitting a message from within Exim. Since child_open_exim() is
601 documented for local_scan(), the new function should be too.
603 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
604 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
605 results in an empty string is now treated as unset.
607 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
609 PH/31 Added sender_verify_fail logging option.
611 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
612 needed by RFC 822 but not by RFC 2822 was commented out. I have now
613 tidied the source and removed it altogether.
615 PH/33 When a queue run was abandoned because the load average was too high, a
616 log line was always written; now it is written only if the queue_run log
617 selector is set. In addition, the log line for abandonment now contains
618 information about the queue run such as the pid. This is always present
619 in "start" and "stop" lines but was omitted from the "abandon" line.
621 PH/34 Omit spaces between a header name and the colon in the error message that
622 is given when verify = headers_syntax fails (if there are lots of them,
623 the message gets confusing).
625 PH/35 Change the default for dns_check_names_pattern to allow slashes within
626 names, as there are now some PTR records that contain slashes. This check
627 is only to protect against broken name servers that fall over on strange
628 characters, so the fact that it applies to all lookups doesn't matter.
630 PH/36 Now that the new test suite is complete, we can remove some of the
631 special code in Exim that was needed for the old test suite. For example,
632 sorting DNS records because real resolvers return them in an arbitrary
633 order. The new test suite's fake resolver always returns records in the
636 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
637 bounces) except when queue_only is set, to avoid logging races between
638 the different processes.
640 PH/38 Panic-die if .include specifies a non-absolute path.
642 PH/39 A tweak to the "H" retry rule from its user.
644 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
645 a label. They prevented compilation on older perls.
647 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
648 a warning to be raised on newish perls.
650 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
651 on queue. Changes to match documented behaviour of showing count of
652 messages matching specified criteria.
654 PH/40 Changed the default ident timeout from 30s to 5s.
656 PH/41 Added support for the use of login_cap features, on those BSD systems
657 that have them, for controlling the resources used by pipe deliveries.
659 PH/42 The content-scanning code uses fopen() to create files in which to put
660 message data. Previously it was not paying any attention to the mode of
661 the files. Exim runs with umask(0) because the rest of the code creates
662 files with open(), and sets the required mode explicitly. Thus, these
663 files were ending up world-writeable. This was not a big issue, because,
664 being within the spool directory, they were not world-accessible. I have
665 created a function called modefopen, which takes an additional mode
666 argument. It sets umask(777), creates the file, chmods it to the required
667 mode, then resets the umask. All the relevant calls to fopen() in the
668 content scanning code have been changed to use this function.
670 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
671 to 24 hours. This avoids potential overflow problems when processing G
672 and H retry rules. I suspect nobody ever tinkers with this value.
674 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
676 PH/45 When the plaintext authenticator is running as a client, the server's
677 challenges are checked to ensure they are valid base64 strings. By
678 default, the authentication attempt is cancelled if an invalid string is
679 received. Setting client_ignore_invalid_base64 true ignores these errors.
680 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
681 they are received. Thus, the responses can be made to depend on the
682 challenges. If an invalid string is ignored, an empty string is placed in
685 PH/46 Messages that are created by the autoreply transport now contains a
686 References: header, in accordance with RFCs 2822 and 3834.
688 PH/47 Added authenticated_sender_force to the smtp transport.
690 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
692 PH/49 Installed latest patch from the Sieve maintainer.
694 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
695 was also set, the appendfile transport was unnecessarily scanning a
696 directory of message files (e.g. for maildir delivery) to find the count
697 of files (along with the size), even though it did not need this
698 information. It now does the scan only if it needs to find either the
699 size of the count of files.
701 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
703 PH/52 Two bugs concerned with error handling when the smtp transport is
706 (i) Exim was not creating retry information for temporary errors given
707 for individual recipients after the DATA command when the smtp transport
708 was used in LMTP mode. This meant that they could be retried too
709 frequently, and not timed out correctly.
711 (ii) Exim was setting the flag that allows error details to be returned
712 for LMTP errors on RCPT commands, but not for LMTP errors for individual
713 recipients that were returned after the DATA command.
715 PH/53 This is related to PH/52, but is more general: for any failing address,
716 when detailed error information was permitted to be returned to the
717 sender, but the error was temporary, then after the final timeout, only
718 "retry timeout exceeded" was returned. Now it returns the full error as
719 well as "retry timeout exceeded".
721 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
722 do this, and (what is worse) MTAs that accept it.
724 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
725 will now be deprecated.
727 PH/56 New os.c-cygwin from the Cygwin maintainer.
729 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
730 formats (previously only available in exim formats via -bpr, -bpru,
731 and -bpra. Now also available in native and exiqgrep formats)
733 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
734 with very large, slow to parse queues
736 JJ/08 exipick: added ! as generic prefix to negate any criteria format
738 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
740 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
741 responses to authentication challenges, though it was showing the
742 challenges; (ii) I've removed the CR characters from the debug output for
745 PH/58 Allow for the insertion of a newline as well as a space when a string
746 is turned into more than one encoded-word during RFC 2047 encoding. The
747 Sieve code now uses this.
749 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
750 data_4xx, lost_connection, tls_required.
752 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
753 message was being sent as an SMTP response.
755 PH/61 Add -l and -k options to exicyclog.
757 PH/62 When verifying, if an address was redirected to one new address, so that
758 verification continued, and the new address failed or deferred after
759 having set something in $address_data, the value of $address_data was not
760 passed back to the ACL. This was different to the case when no
761 redirection occurred. The value is now passed back in both cases.
763 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
764 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
765 use login_cap.h, so on its own it isn't the distinguishing feature. The
766 new name refers directly to the setclassresources() function.
768 PH/65 Added configuration files for NetBSD3.
770 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
772 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
773 is preferred over IPv4.
775 PH/68 The bounce_return_message and bounce_return_body options were not being
776 honoured for bounces generated during the reception of non-SMTP messages.
777 In particular, this applied to messages rejected by the ACL. This bug has
778 been fixed. However, if bounce_return_message is true and bounce_return_
779 body is false, the headers that are returned for a non-SMTP message
780 include only those that have been read before the error was detected.
781 (In the case of an ACL rejection, they have all been read.)
783 PH/69 The HTML version of the specification is now built in a directory called
784 spec_html instead of spec.html, because the latter looks like a path with
785 a MIME-type, and this confuses some software.
787 PH/70 Catch two compiler warnings in sieve.c.
789 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
790 function verify_get_ident() calls ip_connect() to connect a socket, but
791 if the "connect()" function timed out, ip_connect() used to close the
792 socket. However, verify_get_ident() also closes the socket later, and in
793 between Exim writes to the log, which may get opened at this point. When
794 the socket was closed in ip_connect(), the log could get the same file
795 descriptor number as the socket. This naturally causes chaos. The fix is
796 not to close the socket in ip_connect(); the socket should be closed by
797 the function that creates it. There was only one place in the code where
798 this was missing, in the iplookup router, which I don't think anybody now
799 uses, but I've fixed it anyway.
801 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
802 well as to direct DNS lookups. Otherwise the handling of names in host
803 lists is inconsistent and therefore confusing.
809 PH/01 Two changes to the default runtime configuration:
811 (1) Move the checks for relay_from_hosts and authenticated clients from
812 after to before the (commented out) DNS black list checks.
814 (2) Add control=submission to the relay_from_hosts and authenticated
815 clients checks, on the grounds that messages accepted by these
816 statements are most likely to be submissions.
818 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
820 (1) Generate an error if the third argument for the ${prvs expansion is
823 (2) Treat a missing third argument of ${prvscheck as if it were an empty
826 (3) Reset the variables that are obtained from the first argument of
827 ${prvscheck and used in the second argument before leaving the code,
828 because their memory is reclaimed, so using them afterwards may do
831 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
832 one (it's much easier than Tom thought :-).
834 (5) Because of (4), we can now allow for the use of $prvscheck_result
835 inside the third argument.
837 PH/03 For some reason, the default setting of PATH when running a command from
838 a pipe transport was just "/usr/bin". I have changed it to
841 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
842 anything to be listed in the output from -bV.
844 PH/05 When a filter generated an autoreply, the entire To: header line was
845 quoted in the delivery log line, like this:
847 => >A.N.Other <ano@some.domain> <original@ddress> ...
849 This has been changed so that it extracts the operative address. There
850 may be more than one such address. If so, they are comma-separated, like
853 => >ano@some.domain,ona@other.domain <original@ddress> ...
855 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
856 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
857 not being looked up in the rDNS to get a host name, Exim was showing the
858 IP address twice in Received: lines, even though the IP addresses were
859 identical. For example:
861 Received: from [1.2.3.4] (helo=[1.2.3.4])
863 However, if the real host name was known, it was omitting the HELO data
864 if it matched the actual IP address. This has been tidied up so that it
865 doesn't show the same IP address twice.
867 PH/07 When both +timestamp and +memory debugging was on, the value given by
868 $tod_xxx expansions could be wrong, because the tod_stamp() function was
869 called by the debug printing, thereby overwriting the timestamp buffer.
870 Debugging no longer uses the tod_stamp() function when +timestamp is set.
872 PH/08 When the original message was included in an autoreply transport, it
873 always said "this is a copy of the message, including all the headers",
874 even if body_only or headers_only was set. It now gives an appropriate
877 PH/09 Applied a patch from the Sieve maintainer which:
879 o fixes some comments
880 o adds the (disabled) notify extension core
881 o adds some debug output for the result of if/elsif tests
882 o points to the current vacation draft in the documentation
883 and documents the missing references header update
887 o fixes a bug in processing the envelope test (when testing
888 multiple envelope elements, the last element determinted the
891 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
892 Electronic Mail") by including:
894 Auto-submitted: auto-generated
896 in the messages that it generates (bounce messages and others, such as
897 warnings). In the case of bounce messages for non-SMTP mesages, there was
898 also a typo: it was using "Auto_submitted" (underscore instead of
899 hyphen). Since every message generated by Exim is necessarily in response
900 to another message, thes have all been changed to:
902 Auto-Submitted: auto-replied
904 in accordance with these statements in the RFC:
906 The auto-replied keyword:
908 - SHOULD be used on messages sent in direct response to another
909 message by an automatic process,
911 - MUST NOT be used on manually-generated messages,
913 - MAY be used on Delivery Status Notifications (DSNs) and Message
914 Disposition Notifications (MDNs),
916 - MUST NOT be used on messages generated by automatic or periodic
917 processes, except for messages which are automatic responses to
920 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
921 to the default Received: header definition.
923 PH/12 Added log selector acl_warn_skipped (default on).
925 PH/13 After a successful wildlsearch lookup, discard the values of numeric
926 variables because (a) they are in the wrong storage pool and (b) even if
927 they were copied, it wouldn't work properly because of the caching.
929 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
930 checking when decoding. Apparently there are clients that generate
931 overlong encoded strings. Why am I not surprised?
933 PH/15 If the first argument of "${if match_address" was not empty, but did not
934 contain an "@" character, Exim crashed. Now it writes a panic log message
935 and treats the condition as false.
937 PH/16 In autoreply, treat an empty string for "once" the same as unset.
939 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
940 extension "envelope-auth". The code is finished and in agreement with
941 other implementations, but there is no documentation so far and in fact,
942 nobody wrote the draft yet. This extension is currently #undef'ed, thus
943 not changing the active code.
945 Print executed "if" and "elsif" statements when debugging is used. This
946 helps a great deal to understand what a filter does.
948 Document more things not specified clearly in RFC3028. I had all this
949 sorted out, when out of a sudden new issues came to my mind. Oops."
951 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
954 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
955 canonical form (as documented). However, after a host name lookup from
956 the IP address, check_host() was doing a simple string comparison with
957 addresses acquired from the DNS when checking that the found name did
958 have the original IP as one of its addresses. Since any found IPv6
959 addresses are likely to be in abbreviated form, the comparison could
960 fail. Luckily, there already exists a function for doing the comparison
961 by converting both addresses to binary, so now that is used instead of
964 PH/20 There was another similar case to PH/19, when a complete host name was
965 given in a host list; looking up its IP address could give an abbreviated
966 form, whereas the current host's name might or might not be abbreviated.
967 The same fix has been applied.
973 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
974 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
977 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
978 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
980 PH/03 Typo: missing ".o" in src/pcre/Makefile.
982 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
983 header line, restrict the check to what is listed in RFCs 2369 and 2929.
984 Also, for "Auto-Submitted", treat anything other than "no" as
985 non-personal, in accordance with RFC 3834. (Previously it treated
986 anything starting "auto-" as non-personal.)
988 TF/01 The control=submission/name=... option had a problem with syntax
989 errors if the name included a slash character. The /name= option
990 now slurps the rest of the string, so it can include any characters
991 but it must come last in the list of options (after /sender_retain
994 PH/05 Some modifications to the interface to the fake nameserver for the new
1002 TK/01 Added the "success_on_redirect" address verification option. See
1003 NewStuff for rationale and an example.
1005 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1007 PH/02 Patch to exigrep to allow it to work on syslog lines.
1009 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1010 fread() to skip over the body file's header line, because in Cygwin the
1011 header line is locked and is inaccessible.
1013 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1014 co-exist for some time) to make it clear that it is the Exim ID that is
1015 referenced, not the Message-ID: header line.
1017 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1018 string_format() function, because snprintf() does not exist on all
1021 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1022 ${stat: expansion item.
1024 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1025 protocol synchronization error", to keep the pedants happy.
1027 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1028 well as for IRIX systems, when gcc is being used. See the host.c source
1031 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1033 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1036 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1037 transport and to the smtp transport in LMTP mode.
1039 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1041 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1042 run a filter in a subprocess. This could lead to confusion in subsequent
1043 lookups in the parent process. There should also be a search_tidyup() at
1044 the end of the subprocess.
1046 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1047 only if the host matched helo_try_verify_hosts, which caused the
1048 verification to occur when the EHLO/HELO command was issued. The ACL just
1049 tested the remembered result. Now, if a previous verification attempt has
1050 not happened, "verify = helo" does it there and then.
1052 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1054 TK/03 Fix log output including CR from clamd.
1056 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1057 exist provoked a memory error which could cause a segfault.
1059 PH/15 Installed PCRE 6.2
1061 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1063 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1064 of the problem. Specifically, suggested +O2 rather than +O1 for the
1067 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1069 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1070 record, the port was not correctly logged when the outgoing_port log
1071 selector was set (it logged the transort's default port).
1073 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1074 fallback_hosts, and "hosts" in the smtp transport.
1076 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1077 host errors such as "Connection refused".
1079 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1080 authentication with radiusclient 0.4.9:
1082 - Error returned from rc_read_config was caught wrongly
1083 - Username/password not passed on to radius server due to wrong length.
1085 The presumption is that some radiusclient API changes for 4.51/PH/17
1086 were not taken care of correctly. The code is still untested by me (my
1087 Linux distribution still has 0.3.2 of radiusclient), but it was
1088 contributed by a Radius user.
1090 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1091 expanding the "port" option of the smtp transport.
1093 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1094 while reading a MIME header. Thanks to Tom Hughes for a patch.
1096 PH/24 Include config.h inside local_scan.h so that configuration settings are
1099 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1100 that in an ACL for RCPT (for example), you can examine exactly what was
1103 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1104 commands, but it was not correctly comparing the address with the actual
1105 client host address. Thus, it would show the EHLO address in Received:
1106 header lines when this was not necessary.
1108 PH/27 Added the % operator to ${eval:}.
1110 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1111 it should be ignoring failures (because with -C, for example, it has lost
1112 privilege). It wasn't ignoring creation failures other than "already
1115 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1118 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1119 because an input line was too long, either on its own, or by virtue of
1120 too many continuations, the temporary file was not being removed, and the
1121 return code was incorrect.
1123 PH/31 Missing "BOOL" in function definition in filtertest.c.
1125 PH/32 Applied Sieve patches from the maintainer.
1127 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1129 PH/33 Added "verify = not_blind".
1131 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1132 Local/Makefile (with some defaults set). These are used in built scripts
1133 such as exicyclog, but they have never been used in the exim_install
1134 script (though there are many overriding facilities there). I have
1135 arranged that the exim_install script now takes note of these two
1138 PH/35 Installed configuration files for Dragonfly.
1140 PH/36 When a locally submitted message by a trusted user did not contain a
1141 From: header, and the sender address was obtained from -f or from an SMTP
1142 MAIL command, and the trusted user did not use -F to supply a sender
1143 name, $originator_name was incorrectly used when constructing a From:
1144 header. Furthermore, $originator_name was used for submission mode
1145 messages from external hosts without From: headers in a similar way,
1146 which is clearly wrong.
1148 PH/37 Added control=suppress_local_fixups.
1150 PH/38 When log_selector = +received_sender was set, and the addition of the
1151 sender made the log line's construction buffer exactly full, or one byte
1152 less than full, an overflow happened when the terminating "\n" was
1155 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1156 when the result of a list match is failure because a DNS lookup failed.
1158 PH/40 RM_COMMAND is now used in the building process.
1160 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1161 the "build-* directories that it finds.
1163 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1164 address in a domain literal was a prefix of an interface address.
1166 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1167 when verifying a sender address, unless rewrite_headers is false.
1169 PH/44 Wrote a long comment about why errors_to addresses are verified as
1170 recipients, not senders.
1172 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1173 the ratelimit ACL was added.
1175 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1177 PH/46 Added extra information about PostgreSQL errors to the error string.
1179 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1180 avoiding the need to install special zones in a real server. This is
1181 backwards compatible; if it can't find the fake resolver, it drops back.
1182 Thus, both old and new test suites can be run.
1184 TF/02 Added util/ratelimit.pl
1186 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1187 clock is set back in time.
1189 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1190 Candler <B.Candler@pobox.com>.
1192 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1193 OK for addresses that are the result of redirections.
1195 PH/48 A number of further additions for the benefit of the new test suite,
1196 including a fake gethostbyname() that interfaces to the fake DNS resolver
1199 TF/06 The fix for widen_domains has also been applied to qualify_single and
1200 search_parents which are the other dnslookup options that can cause
1203 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1206 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1208 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1209 service" if it encountered an MX record with an empty target hostname.
1210 The message is now "an MX or SRV record indicated no SMTP service".
1212 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1213 if the DNS of the sending site is misconfigured. This is quite a
1214 common situation. This change restores the behaviour of treating a
1215 helo verification defer as a failure.
1217 PH/51 If self=fail was set on a router, the bounce message did not include the
1218 actual error message.
1224 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1226 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1227 command itself ended in error, the underlying message about the transport
1228 filter timeout was being overwritten with the pipe command error. Now the
1229 underlying error message should be appended to the second error message.
1231 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1233 PH/02 Reduce the amount of output that "make" produces by default. Full output
1234 can still be requested.
1236 PH/03 The warning log line about a condition test deferring for a "warn" verb
1237 was being output only once per connection, rather than after each
1238 occurrence (because it was using the same function as for successful
1239 "warn" verbs). This seems wrong, so I have changed it.
1241 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1242 it should not have, which might have caused a crash in the right
1243 circumstances, but probably never did.
1245 PH/04 Installed a modified version of Tony Finch's patch to make submission
1246 mode fix the return path as well as the Sender: header line, and to
1247 add a /name= option so that you can make the user's friendly name appear
1250 TF/03 Added the control = fakedefer ACL modifier.
1252 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1253 Mark Lowes for thorough testing.
1255 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1257 TK/03 Merged latest SRS patch from Miles Wilton.
1259 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1260 with the definition in sysexits.h (which is #included earlier).
1261 Fortunately, Exim does not actually use EX_OK. The code used to try to
1262 preserve the sysexits.h value, by assumimg that macro definitions were
1263 scanned for macro replacements. I have been disabused of this notion,
1264 so now the code just undefines EX_OK before #including unistd.h.
1266 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1267 in the smtp transport. When a block could not be written in a single
1268 write() function, the timeout was being re-applied to each part-write.
1269 This seems wrong - if the receiver was accepting one byte at a time it
1270 would take for ever. The timeout is now adjusted when this happens. It
1271 doesn't have to be particularly precise.
1273 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1274 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1276 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1277 without a domain if the check to <postmaster@domain> fails.
1279 SC/01 Eximstats: added -xls and the ability to specify output files
1280 (patch written by Frank Heydlauf).
1282 SC/02 Eximstats: use FileHandles for outputing results.
1284 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1286 SC/04 Eximstats: fixed display of large numbers with -nvr option
1288 SC/05 Eximstats: fixed merging of reports with empty tables.
1290 SC/06 Eximstats: added the -include_original_destination flag
1292 SC/07 Eximstats: removed tabs and trailing whitespace.
1294 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1296 TK/06 MBOX spool code: Add real "From " MBOX separator line
1297 so the .eml file is really in mbox format (even though
1298 most programs do not really care). Patch from Alex Miller.
1300 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1301 The latter is generated from $received_to and is only set if the
1302 message has one envelope recipient. SA can use these headers,
1303 obviously out-of-the-box. Patch from Alex Miller.
1305 PH/08 The ${def test on a variable was returning false if the variable's
1306 value was "0", contrary to what the specification has always said!
1307 The result should be true unless the variable is empty.
1309 PH/09 The syntax error of a character other than { following "${if
1310 def:variable_name" (after optional whitespace) was not being diagnosed.
1311 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1312 accidental colon was present, for example, could give incorrect results.
1314 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1315 result is used (not including appendfile, where other changes are about
1318 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1319 This involved changing a lot of size variables from int to off_t. It
1320 should work with maildirs and everything.
1322 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1323 spamd dying while we are connected to it.
1325 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1328 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1329 function for BDB 4.3.
1331 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1333 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1334 the sources for PCRE have been re-arranged and more files are now
1337 PH/15 The code I had for printing potentially long long variables in PH/11
1338 above was not the best (it lost precision). The length of off_t variables
1339 is now inspected at build time, and an appropriate printing format (%ld
1340 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1341 to be "long long int" or "long int". This is needed for the internal
1342 formatting function string_vformat().
1344 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1345 the configuration file to be ":syslog", then the script "guesses" where
1346 the logs files are, rather than using the compiled in default. In our
1347 case the guess is not the same as the compiled default, so the script
1348 suddenly stopped working when I started to use syslog. The patch checks
1349 to see if log_file_path is "". If so, it attempts to read it from exim
1350 with no configuration file to get the compiled in version, before it
1351 falls back to the previous guessing code."
1353 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1354 implementing BATV in an Exim configuration. See NewStuff for the gory
1357 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1358 Makefile that are specific to HP-UX.
1360 PH/18 If the "use_postmaster" option was set for a recipient callout together
1361 with the "random" option, the postmaster address was used as the MAIL
1362 FROM address for the random test, but not for the subsequent recipient
1363 test. It is now used for both.
1365 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1366 patch removes a few documentation additions to RFC 3028, because the
1367 latest draft now contains them. It adds the new en;ascii-case comparator
1368 and a new error check for 8bit text in MIME parts. Comparator and
1369 require names are now matched exactly. I enabled the subaddress
1370 extension, but it is not well tested yet (read: it works for me)."
1372 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1373 rework some of the code of TK/09 above to avoid the hardwired use of
1374 "%lld" and "long long". Replaced the call to snprintf() with a call to
1377 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1378 records point to non-existent hosts", "retry timeout exceeded", and
1379 "retry time not reached for any host after a long failure period".
1381 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1382 experimental DomainKeys support:
1384 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1385 (2) On an error such as an illegally used "control", the wrong name for
1386 the control was given.
1388 These problems did NOT occur unless DomainKeys support was compiled.
1390 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1392 PH/24 Added ${if match_ip condition.
1394 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1395 databases so that it will be absolutely obvious if a crash occurs in the
1396 DB library. This is a regular occurrence (often caused by mis-matched
1399 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1400 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1401 picked up on a user's system that detects such things. There doesn't seem
1402 to be a gcc warning option for this - only an attribute that has to be
1403 put on the function's prototype. It seems that in Fedora Core 4 they have
1404 set this on a number of new functions. No doubt there will be more in due
1407 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1408 specify a transport. However, if an address that was verified by such a
1409 router was the subject of a callout, Exim crashed because it tried to
1410 read the rcpt_include_affixes from the non-existent transport. Now it
1411 just assumes that the setting of that option is false. This bug was
1412 introduced by 4.51/PH/31.
1414 PH/28 Changed -d+all to exclude +memory, because that information is very
1415 rarely of interest, but it makes the output a lot bigger. People tend to
1416 do -d+all out of habit.
1418 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1419 code in os-type was giving problems when libc.so lives in lib64, like on
1422 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1423 aren't the modern standard, and it seems that some systems' include files
1424 don't always have them. Exim was already checking for some of the newer
1425 ones like T_AAAA, and defining it itself. I've added checks for all the
1426 record types that Exim uses.
1428 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1429 not automatically generating a new one, as it is supposed to. This
1430 prevented TLS from working. If the file did exist, but contained invalid
1431 data, a new version was generated, as expected. It was only the case of a
1432 non-existent file that was broken.
1434 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1435 with a change in libdomainkeys > 0.64.
1437 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1438 from DNS. If the selector record carries the flag, it now has
1439 precedence over the domain-wide flag.
1441 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1443 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1444 the use of a port name that isn't defined in /etc/services) occurred, the
1445 message was deferred as in a normal delivery, and thus remained on the
1446 spool, instead of being failed because of the mua_wrapper setting. This
1447 is now fixed, and I tidied up some of the mua_wrapper messages at the
1450 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1451 the messages in a hash of arrays rather than using individual hashes.
1452 This is a bit cleaner and results in dramatic memory savings, albeit
1453 at a slight CPU cost.
1455 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1456 as requested by Marc Sherman.
1458 SC/10 Eximstats: added histograms for user specified patterns as requested
1461 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1463 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1464 fopen() in the content-scanning modules that did not already have it.
1470 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1471 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1473 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1475 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1477 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1478 to test Sieve filters that use "vacation".
1480 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1481 that changes the way the GnuTLS parameters are stored in the cache file.
1482 The new format can be generated externally. For backward compatibility,
1483 if the data in the cache doesn't make sense, Exim assumes it has read an
1484 old-format file, and it generates new data and writes a new file. This
1485 means that you can't go back to an older release without removing the
1488 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1489 work if there are any delivery delays because "one_time" forces the
1490 parent to be marked "delivered", so its unseen clone is never tried
1491 again. For this reason, Exim now forbids the simultaneous setting of
1494 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1495 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1496 entry if you want to know the details. The fix, however, neglected to
1497 consider the case when local delivery batching is involved. The test for
1498 "previously delivered" was not happening when checking to see if an
1499 address could be batched with a previous (undelivered) one; under
1500 certain circumstances this could lead to multiple deliveries to the same
1503 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1504 in its include files, and this causes problems building Exim.
1506 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1507 header_syntax) but Exim was just ignoring anything given after a slash.
1508 In particular, this caused confusion with an attempt to use "verify =
1509 reverse_host_lookup/defer_ok". An error is now given when options are
1510 supplied for verify items that do not have them. (Maybe reverse_host_
1511 lookup should have a defer_ok option, but that's a different point.)
1513 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1514 defined by RFC 821) to 2048, because there were problems with some AUTH
1515 commands, and RFC 1869 says the size should be increased for extended
1516 SMTP commands that take arguments.
1518 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1521 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1522 "unknown" error; now it says that the functionality isn't in the binary.
1524 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1525 an address' error message when a string expansion fails (syntax or
1526 whatever). Otherwise the password may appear in the log. Following change
1527 PH/42 below, there is no longer a chance of it appearing in a bounce
1530 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1532 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1533 of its addresses was ever tried. (Bugzilla bug #2.)
1535 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1536 the result incorrectly in the debug output. (It correctly added a newline
1537 to what was transported.)
1539 TF/01 Added $received_time.
1541 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1542 commented out examples of how to interface to a virus scanner and to
1543 SpamAssassin. Also added commented examples of av_scanner and
1544 spamd_address settings.
1546 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1547 and controls are allowed in which ACLs. There were a couple of minor
1548 errors. Some of the entries in the conditions table (which is a table of
1549 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1550 negation of where the condition IS allowed.
1552 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1554 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1555 header file does not have a version number, so I've had to invent a new
1556 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1557 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1558 radiusclient), but it was contributed by a Radius user.
1560 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1561 files or directories, for OpenSSL.
1563 PH/19 When an Exim process that is running as root has to create an Exim log
1564 file, it does so in a subprocess that runs as exim:exim so as to get the
1565 ownership right at creation (otherwise, other Exim processes might see
1566 the file with the wrong ownership). There was no test for failure of this
1567 fork() call, which would lead to the process getting stuck as it waited
1568 for a non-existent subprocess. Forks do occasionally fail when resources
1569 run out. I reviewed all the other calls to fork(); they all seem to check
1572 PH/20 When checking for unexpected SMTP input at connect time (before writing
1573 the banner), Exim was not dealing correctly with a non-positive return
1574 from the read() function. If the client had disconnected by this time,
1575 the result was a log entry for a synchronization error with an empty
1576 string after "input=" when read() returned zero. If read() returned -1
1577 (an event I could not check), uninitialized data bytes were printed.
1578 There were reports of junk text (parts of files, etc) appearing after
1581 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1583 PH/22 Added support for macro redefinition, and (re)definition in between
1584 driver and ACL definitions.
1586 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1587 forgetting to use the resulting value; it was using the unexpanded value.
1589 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1590 hadn't been configured. The fix is from Juergen Kreileder, who
1591 understands it better than I do:
1593 "Here's what I see happening with three configured cyrus_sasl
1594 authenticators configured (plain, login, cram-md5):
1596 On startup auth_cyrus_sasl_init() gets called for each of these.
1597 This means three calls to sasl_listmech() without a specified mech_list.
1598 => SASL tests which mechs of all available mechs actually work
1599 => three warnings about OTP not working
1600 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1602 With the patch, sasl_listmech() also gets called three times. But now
1603 SASL's mech_list option is set to the server_mech specified in the the
1604 authenticator. Or in other words, the answer from sasl_listmech()
1605 gets limited to just the mech you're testing for (which is different
1607 => the return list contains just 'plain' or 'login', 'cram-md5' or
1608 nothing depending on the value of ob->server_mech.
1610 I've just tested the patch: Authentication still works fine,
1611 unavailable mechs specified in the exim configuration are still
1612 caught, and the auth.log warnings about OTP are gone."
1614 PH/25 When debugging is enabled, the contents of the command line are added
1615 to the debugging output, even when log_selector=+arguments is not
1618 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1619 answer is "GNU", and only if the return is "GNU/something" is the answer
1622 PH/27 $acl_verify_message is now set immediately after the failure of a
1623 verification in an ACL, and so is available in subsequent modifiers. In
1624 particular, the message can be preserved by coding like this:
1626 warn !verify = sender
1627 set acl_m0 = $acl_verify_message
1629 Previously, $acl_verify_message was set only while expanding "message"
1630 and "log_message" when a very denied access.
1632 PH/28 Modified OS/os.c-Linux with
1634 -#ifndef OS_LOAD_AVERAGE
1635 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1637 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1638 nomenclature these days.)
1640 PH/29 Installed patch from the Sieve maintainer that adds the options
1641 sieve_useraddress and sieve_subaddress to the redirect router.
1643 PH/30 In these circumstances:
1644 . Two addresses routed to the same list of hosts;
1645 . First host does not offer TLS;
1646 . First host accepts first address;
1647 . First host gives temporary error to second address;
1648 . Second host offers TLS and a TLS session is established;
1649 . Second host accepts second address.
1650 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1651 and peerdn, if requested) that were in fact used only for the second
1654 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1655 attention to any local part prefix or suffix that was matched by the
1656 router that accepted the address. It now behaves in the same way as it
1657 does for delivery: the affixes are removed from the local part unless
1658 rcpt_include_affixes is set on the transport.
1660 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1661 timeout during the DATA phase of an incoming message.
1663 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1664 applied a patch sanctioned by the Sieve maintainer.
1666 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1667 the uid or gid is negative. A case of a negative gid caused this to be
1668 noticed. The fix allows for either to be negative.
1670 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1671 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1674 PH/36 The change PH/12 above was broken. Fixed it.
1676 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1677 the grounds that routing the same address twice would always produce the
1678 same answer. This might have been true once, but it is certainly no
1679 longer true now. Routing a child address may depend on the previous
1680 routing that produced that child. Some complicated redirection strategies
1681 went wrong when messages had multiple recipients, and made Exim's
1682 behaviour dependent on the order in which the addresses were given.
1684 I have moved the duplicate checking until after the routing is complete.
1685 Exim scans the addresses that are assigned to local and remote
1686 transports, and removes any duplicates. This means that more work will be
1687 done, as duplicates will always all be routed, but duplicates are
1688 presumably rare, so I don't expect this is of any significance.
1690 For deliveries to pipes, files, and autoreplies, the duplicate checking
1691 still happens during the routing process, since they are not going to be
1694 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1695 It corrects a timeout issue with spamd. This is Ian's comment: "The
1696 background is that sometimes spamd either never reads data from a
1697 connection it has accepted, or it never writes response data. The exiscan
1698 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1699 blindly assumes that writes won't block so it may never time out."
1701 PH/39 Allow G after quota size as well as K and M.
1703 PH/40 The value set for $authenticated_id in an authenticator may not contain
1704 binary zeroes or newlines because the value is written to log lines and
1705 to spool files. There was no check on this. Now the value is run through
1706 the string_printing() function so that such characters are converted to
1707 printable escape sequences.
1709 PH/41 $message_linecount is a new variable that contains the total number of
1710 lines in the message. Compare $body_linecount, which is the count for the
1713 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1714 bounce and delay warning messages, except in certain special cases, which
1717 (a) An SMTP error message from a remote host;
1718 (b) A message specified in a :fail: redirection;
1719 (c) A message specified in a "fail" command in a system filter;
1720 (d) A message specified in a FAIL return from the queryprogram router;
1721 (e) A message specified by the cannot_route_message router option.
1723 In these cases only, Exim does include the error details in bounce and
1724 warning messages. There are also a few cases where bland messages such
1725 as "unrouteable address" or "local delivery error" are given.
1727 PH/43 $value is now also set for the "else" part of a ${run expansion.
1729 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1730 being worked on, but at least Exim now implements the latest version to
1733 PH/45 In a pipe transport, although a timeout while waiting for the pipe
1734 process to complete was treated as a delivery failure, a timeout while
1735 writing the message to the pipe was logged, but erroneously treated as a
1736 successful delivery. Such timeouts include transport filter timeouts. For
1737 consistency with the overall process timeout, these timeouts are now
1738 treated as errors, giving rise to delivery failures by default. However,
1739 there is now a new Boolean option for the pipe transport called
1740 timeout_defer, which, if set TRUE, converts the failures into defers for
1741 both kinds of timeout. A transport filter timeout is now identified in
1744 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1745 systems where "make" and "gmake" are different, calling "gmake" at top
1746 level broke things. I've arranged for the value of $(MAKE) to be passed
1747 from the Makefile to this script so that it can call the same version of
1751 A note about Exim versions 4.44 and 4.50
1752 ----------------------------------------
1754 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1755 changes of various kinds. As a consequence, a big documentation update was
1756 needed. This delayed the release for rather longer than seemed good, especially
1757 in the light of a couple of (minor) security issues. Therefore, the changes
1758 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1759 release. So 4.44 and 4.50 are in effect two different branches that both start
1762 I have left the 4.50 change log unchanged; it contains all the changes since
1763 4.43. The change log for 4.44 is below; many of its items are identical to
1764 those for 4.50. This seems to be the most sensible way to preserve the
1765 historical information.
1771 1. Minor wording change to the doc/README.SIEVE file.
1773 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1774 computation of the current number of files was incorrect.
1776 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1777 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1780 4. Give more explanation in the error message when the command for a transport
1781 filter fails to execute.
1783 5. There are several places where Exim runs a non-Exim command in a
1784 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1785 was being done only for the command run by the queryprogram router. It is
1786 now done for all such subprocesses. The other cases are: ${run, transport
1787 filters, and the commands run by the lmtp and pipe transports.
1789 6. Added CONFIGURE_GROUP build-time option.
1791 7. Some older OS have a limit of 256 on the maximum number of file
1792 descriptors. Exim was using setrlimit() to set 1000 as a large value
1793 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1794 systems. I've change it so that if it can't get 1000, it tries for 256.
1796 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1797 was an oversight, and furthermore, ever since the addition of extra
1798 controls (e.g. 4.43/32), the checks on when to allow different forms of
1799 "control" were broken. There should now be diagnostics for all cases when a
1800 control that does not make sense is encountered.
1802 9. Added the /retain_sender option to "control=submission".
1804 10. $recipients is now available in the predata ACL (oversight).
1806 11. Tidy the search cache before the fork to do a delivery from a message
1807 received from the command line. Otherwise the child will trigger a lookup
1808 failure and thereby defer the delivery if it tries to use (for example) a
1809 cached ldap connection that the parent has called unbind on.
1811 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1812 of $address_data from the recipient verification was clobbered by the
1813 sender verification.
1815 13. The value of address_data from a sender verification is now available in
1816 $sender_address_data in subsequent conditions in the ACL statement.
1818 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1820 15. Added a new option "connect=<time>" to callout options, to set a different
1823 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1824 was its contents. (It was OK if the option was not defined at all.)
1826 17. A "Completed" log line is now written for messages that are removed from
1827 the spool by the -Mrm option.
1829 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1830 information about exactly what failed.
1832 19. Added -dd to debug only the daemon process.
1834 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1835 handles timeouts, both on the server side and network timeouts. Renamed the
1836 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1838 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1839 It is now set to "smtps".
1841 22. $host_address is now set to the target address during the checking of
1842 ignore_target_hosts.
1844 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1845 being passed; this would have caused $sender_host_name to have been used if
1846 matching the list had actually called for a host name (not very likely,
1847 since this list is usually IP addresses). A host name is now passed as
1850 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1851 code that specifies a non-restarting handler (typically sigaction() in
1852 modern systems) in an attempt to fix a rare and obscure crash bug.
1854 25. Narrowed the window for a race in the daemon that could cause it to ignore
1855 SIGCHLD signals. This is not a major problem, because they are used only to
1856 wake it up if nothing else does.
1858 26. A malformed maildirsize file could cause Exim to calculate negative values
1859 for the mailbox size or file count. Odd effects could occur as a result.
1860 The maildirsize information is now recalculated if the size or filecount
1863 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1864 support for a long time. Removed HAVE_SYS_VFS_H.
1866 28. Installed the latest version of exipick from John Jetmore.
1868 29. In an address list, if the pattern was not a regular expression, an empty
1869 subject address (from a bounce message) matched only if the pattern was an
1870 empty string. Non-empty patterns were not even tested. This was the wrong
1871 because it is perfectly reasonable to use an empty address as part of a
1872 database query. An empty address is now tested by patterns that are
1873 lookups. However, all the other forms of pattern expect the subject to
1874 contain a local part and a domain, and therefore, for them, an empty
1875 address still always fails if the pattern is not itself empty.
1877 30. Exim went into a mad DNS loop when attempting to do a callout where the
1878 host was specified on an smtp transport, and looking it up yielded more
1879 than one IP address.
1881 31. Re-factored the code for checking spool and log partition space into a
1882 function that finds that data and another that does the check. The former
1883 is then used to implement four new variables: $spool_space, $log_space,
1884 $spool_inodes, and $log_inodes.
1886 32. The RFC2047 encoding function was originally intended for short strings
1887 such as real names; it was not keeping to the 75-character limit for
1888 encoded words that the RFC imposes. It now respects the limit, and
1889 generates multiple encoded words if necessary. To be on the safe side, I
1890 have increased the buffer size for the ${rfc2047: expansion operator from
1893 33. It is now permitted to omit both strings after an "if" condition; if the
1894 condition is true, the result is "true". As before, when the second string
1895 is omitted, a false condition yields an empty string. This makes it less
1896 cumbersome to write custom ACL and router conditions.
1898 34. Failure to deliver a bounce message always caused it to be frozen, even if
1899 there was an errors_to setting on the router. The errors_to setting is now
1902 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1903 canonical form (fully expanded) before being placed in
1904 $sender_host_address.
1906 36. The table in the code that translates DNS record types into text (T_A to
1907 "A" for instance) was missing entries for NS and CNAME. It is just possible
1908 that this could have caused confusion if both these types were looked up
1909 for the same domain, because the text type is used as part of Exim's
1910 per-process caching. But the chance of anyone hitting this buglet seems
1913 37. The dnsdb lookup has been extended in a number of ways.
1915 (1) There is a new type, "zns", which walks up the domain tree until it
1916 finds some nameserver records. It should be used with care.
1918 (2) There is a new type, "mxh", which is like "mx" except that it returns
1919 just the host names, not the priorities.
1921 (3) It is now possible to give a list of domains (or IP addresses) to be
1922 looked up. The behaviour when one of the lookups defers can be
1923 controlled by a keyword.
1925 (4) It is now possible to specify the separator character for use when
1926 multiple records are returned.
1928 38. The dnslists ACL condition has been extended: it is now possible to supply
1929 a list of IP addresses and/or domains to be looked up in a particular DNS
1932 39. Added log_selector=+queue_time_overall.
1934 40. When running the queue in the test harness, wait just a tad after forking a
1935 delivery process, to get repeatability of debugging output.
1937 41. Include certificate and key file names in error message when GnuTLS fails
1938 to set them up, because the GnuTLS error message doesn't include the name
1939 of the failing file when there is a problem reading it.
1941 42. Allow both -bf and -bF in the same test run.
1943 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1945 44. The "Exiscan patch" is now merged into the mainline Exim source.
1947 45. Sometimes the final signoff response after QUIT could fail to get
1948 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1949 < 0 before doing a fflush(). This bug looks as though it goes back to the
1950 introduction of TLS in release 3.20, but "sometimes" must have been rare
1951 because the tests only now provoked it.
1953 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1954 (this can affect the format of dates).
1956 47. exim_tidydb, when checking for the continued existence of a message for
1957 which it has found a message-specific retry record, was not finding
1958 messages that were in split spool directories. Consequently, it was
1959 deleting retry records that should have stayed in existence.
1961 48. Steve fixed some bugs in eximstats.
1963 49. The SPA authentication driver was not abandoning authentication and moving
1964 on to the next authenticator when an expansion was forced to fail,
1965 contradicting the general specification for all authenticators. Instead it
1966 was generating a temporary error. It now behaves as specified.
1968 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1969 (the order specifies the preference for clients). The order is now AES256,
1970 AES128, 3DES, ARCFOUR128.
1972 51. Small patch to Sieve code - explicitly set From: when generating an
1975 52. Exim crashed if a remote delivery caused a very long error message to be
1976 recorded - for instance if somebody sent an entire SpamAssassin report back
1977 as a large number of 550 error lines. This bug was coincidentally fixed by
1978 increasing the size of one of Exim's internal buffers (big_buffer) that
1979 happened as part of the Exiscan merge. However, to be on the safe side, I
1980 have made the code more robust (and fixed the comments that describe what
1983 53. Now that there can be additional text after "Completed" in log lines (if
1984 the queue_time_overall log selector is set), a one-byte patch to exigrep
1985 was needed to allow it to recognize "Completed" as not the last thing in
1988 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1989 patch that reportedly fixes this has been added. I am not expert enough to
1990 create a test for it. This is what the patch creator wrote:
1992 "I found a little strange behaviour of ldap code when working with
1993 Windows 2003 AD Domain, where users was placed in more than one
1994 Organization Units. When I tried to give exim partial DN, the exit code
1995 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1996 But simultaneously result of request was absolutely normal ldap result,
1997 so I produce this patch..."
1999 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2000 REFERENCE, so I have modified the code to exclude the patch when that macro
2003 55. Some experimental protocols are using DNS PTR records for new purposes. The
2004 keys for these records are domain names, not reversed IP addresses. The
2005 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2006 leaves it alone. Component reversal etc. now happens only for IP addresses.
2009 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2011 57. Double the size of the debug message buffer (to 2048) so that more of very
2012 long debug lines gets shown.
2014 58. The exicyclog utility now does better if the number of log files to keep
2015 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2017 59. Two changes related to the smtp_active_hostname option:
2019 (1) $smtp_active_hostname is now available as a variable.
2020 (2) The default for smtp_banner uses $smtp_active_hostname instead
2021 of $primary_hostname.
2023 60. The host_aton() function is supposed to be passed a string that is known
2024 to be a valid IP address. However, in the case of IPv6 addresses, it was
2025 not checking this. This is a hostage to fortune. Exim now panics and dies
2026 if the condition is not met. A case was found where this could be provoked
2027 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2028 components; fortuitously, this particular loophole had already been fixed
2029 by change 4.50/55 above.
2031 If there are any other similar loopholes, the new check in host_aton()
2032 itself should stop them being exploited. The report I received stated that
2033 data on the command line could provoke the exploit when Exim was running as
2034 exim, but did not say which command line option was involved. All I could
2035 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2036 running as the user.
2039 61. There was a buffer overflow vulnerability in the SPA authentication code
2040 (which came originally from the Samba project). I have added a test to the
2041 spa_base64_to_bits() function which I hope fixes it.
2044 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2045 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2047 63. The daemon start-up calls getloadavg() while still root for those OS that
2048 need the first call to be done as root, but it missed one case: when
2049 deliver_queue_load_max is set with deliver_drop_privilege. This is
2050 necessary for the benefit of the queue runner, because there is no re-exec
2051 when deliver_drop_privilege is set.
2053 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2054 This has been fixed.
2056 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2057 was in use, was not putting the data itself into the right store pool;
2058 consequently, it could be overwritten for a subsequent message in the same
2059 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2062 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2064 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2065 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2066 the function were treating the return as a boolean value, which happened to
2067 work because 0=false and not-0=true, but is not correct code.
2069 68. The host_aton() function was not handling scoped IPv6 addresses (those
2070 with, for example, "%eth0" on the end) correctly.
2072 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2073 negated items (that is, ~something) in unsigned ints. Some compilers
2074 apparently mutter when there is no cast.
2076 70. If an address verification called from an ACL failed, and did not produce a
2077 user-specific message (i.e. there was only a "system" message), nothing was
2078 put in $acl_verify_message. In this situation, it now puts the system
2081 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2082 session; change 4.31/43 added the unwanted input to the log line - except
2083 that it did not do this in the start of session case. It now does.
2085 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2086 This is wrong and can cause the other end to generate a synchronization
2087 error if it is another Exim or anything else that does the synchronization
2088 check. A QUIT command is no longer sent after a timeout.
2090 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2091 during host lookups.
2093 74. The defer_ok option of callout verification was not working if it was used
2094 when verifying addresses in header lines, that is, for this case:
2096 verify = header_sender/callout=defer_ok
2098 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2099 those file descriptors could be used for SMTP connections. If anything
2100 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2101 could be sent to the SMTP client, causing chaos. The daemon now opens
2102 stdin, stdout, and stderr to /dev/null when it puts itself into the
2105 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2106 log by default. The user can override this with suitable Perl magic.
2108 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2109 the log message when discard triggers, was not working for the DATA ACL or
2110 for the non-SMTP ACL.
2112 78. Error message wording change in sieve.c.
2114 79. If smtp_accept_max_per_host was set, the number of connections could be
2115 restricted to fewer than expected, because the daemon was trying to set up
2116 a new connection before checking whether the processes handling previous
2117 connections had finished. The check for completed processes is now done
2118 earlier. On busy systems, this bug wouldn't be noticed because something
2119 else would have woken the daemon, and it would have reaped the completed
2122 80. If a message was submitted locally by a user whose login name contained one
2123 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2124 It caused a spool format error. I have fixed the spool reading code. A
2125 related problem was that the "from" clause in the Received: line became
2126 illegal because of the space(s). It is now covered by ${quote_local_part.
2128 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2131 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2133 83. Patch from Sieve maintainer to fix unterminated string problem in
2134 "vacation" handling.
2136 84. Some minor changes to the Linux configuration files to help with other
2137 OS variants using glibc.
2139 85. One more patch for Sieve to update vacation handling to latest spec.
2142 ----------------------------------------------------
2143 See the note above about the 4.44 and 4.50 releases.
2144 ----------------------------------------------------
2150 1. Change 4.43/35 introduced a bug that caused file counts to be
2151 incorrectly computed when quota_filecount was set in an appendfile
2154 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2155 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2158 3. Give more explanation in the error message when the command for a transport
2159 filter fails to execute.
2161 4. There are several places where Exim runs a non-Exim command in a
2162 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2163 was being done only for the command run by the queryprogram router. It is
2164 now done for all such subprocesses. The other cases are: ${run, transport
2165 filters, and the commands run by the lmtp and pipe transports.
2167 5. Some older OS have a limit of 256 on the maximum number of file
2168 descriptors. Exim was using setrlimit() to set 1000 as a large value
2169 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2170 systems. I've change it so that if it can't get 1000, it tries for 256.
2172 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2173 was an oversight, and furthermore, ever since the addition of extra
2174 controls (e.g. 4.43/32), the checks on when to allow different forms of
2175 "control" were broken. There should now be diagnostics for all cases when a
2176 control that does not make sense is encountered.
2178 7. $recipients is now available in the predata ACL (oversight).
2180 8. Tidy the search cache before the fork to do a delivery from a message
2181 received from the command line. Otherwise the child will trigger a lookup
2182 failure and thereby defer the delivery if it tries to use (for example) a
2183 cached ldap connection that the parent has called unbind on.
2185 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2186 of $address_data from the recipient verification was clobbered by the
2187 sender verification.
2189 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2190 was its contents. (It was OK if the option was not defined at all.)
2192 11. A "Completed" log line is now written for messages that are removed from
2193 the spool by the -Mrm option.
2195 12. $host_address is now set to the target address during the checking of
2196 ignore_target_hosts.
2198 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2199 being passed; this would have caused $sender_host_name to have been used if
2200 matching the list had actually called for a host name (not very likely,
2201 since this list is usually IP addresses). A host name is now passed as
2204 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2205 code that specifies a non-restarting handler (typically sigaction() in
2206 modern systems) in an attempt to fix a rare and obscure crash bug.
2208 15. Narrowed the window for a race in the daemon that could cause it to ignore
2209 SIGCHLD signals. This is not a major problem, because they are used only to
2210 wake it up if nothing else does.
2212 16. A malformed maildirsize file could cause Exim to calculate negative values
2213 for the mailbox size or file count. Odd effects could occur as a result.
2214 The maildirsize information is now recalculated if the size or filecount
2217 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2218 support for a long time. Removed HAVE_SYS_VFS_H.
2220 18. Updated exipick to current release from John Jetmore.
2222 19. Allow an empty sender to be matched against a lookup in an address list.
2223 Previously the only cases considered were a regular expression, or an
2226 20. Exim went into a mad DNS lookup loop when doing a callout where the
2227 host was specified on the transport, if the DNS lookup yielded more than
2230 21. The RFC2047 encoding function was originally intended for short strings
2231 such as real names; it was not keeping to the 75-character limit for
2232 encoded words that the RFC imposes. It now respects the limit, and
2233 generates multiple encoded words if necessary. To be on the safe side, I
2234 have increased the buffer size for the ${rfc2047: expansion operator from
2237 22. Failure to deliver a bounce message always caused it to be frozen, even if
2238 there was an errors_to setting on the router. The errors_to setting is now
2241 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2242 canonical form (fully expanded) before being placed in
2243 $sender_host_address.
2245 24. Updated eximstats to version 1.33
2247 25. Include certificate and key file names in error message when GnuTLS fails
2248 to set them up, because the GnuTLS error message doesn't include the name
2249 of the failing file when there is a problem reading it.
2251 26. Expand error message when OpenSSL has problems setting up cert/key files.
2254 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2255 (this can affect the format of dates).
2257 28. exim_tidydb, when checking for the continued existence of a message for
2258 which it has found a message-specific retry record, was not finding
2259 messages that were in split spool directories. Consequently, it was
2260 deleting retry records that should have stayed in existence.
2262 29. eximstats updated to version 1.35
2263 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2264 1.35 - bugfix such that pie charts by volume are generated correctly
2266 30. The SPA authentication driver was not abandoning authentication and moving
2267 on to the next authenticator when an expansion was forced to fail,
2268 contradicting the general specification for all authenticators. Instead it
2269 was generating a temporary error. It now behaves as specified.
2271 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2272 (the order specifies the preference for clients). The order is now AES256,
2273 AES128, 3DES, ARCFOUR128.
2275 31. Small patch to Sieve code - explicitly set From: when generating an
2278 32. Exim crashed if a remote delivery caused a very long error message to be
2279 recorded - for instance if somebody sent an entire SpamAssassin report back
2280 as a large number of 550 error lines. This bug was coincidentally fixed by
2281 increasing the size of one of Exim's internal buffers (big_buffer) that
2282 happened as part of the Exiscan merge. However, to be on the safe side, I
2283 have made the code more robust (and fixed the comments that describe what
2286 33. Some experimental protocols are using DNS PTR records for new purposes. The
2287 keys for these records are domain names, not reversed IP addresses. The
2288 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2289 leaves it alone. Component reversal etc. now happens only for IP addresses.
2292 34. The host_aton() function is supposed to be passed a string that is known
2293 to be a valid IP address. However, in the case of IPv6 addresses, it was
2294 not checking this. This is a hostage to fortune. Exim now panics and dies
2295 if the condition is not met. A case was found where this could be provoked
2296 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2297 components; fortuitously, this particular loophole had already been fixed
2298 by change 4.50/55 or 4.44/33 above.
2300 If there are any other similar loopholes, the new check in host_aton()
2301 itself should stop them being exploited. The report I received stated that
2302 data on the command line could provoke the exploit when Exim was running as
2303 exim, but did not say which command line option was involved. All I could
2304 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2305 running as the user.
2308 35. There was a buffer overflow vulnerability in the SPA authentication code
2309 (which came originally from the Samba project). I have added a test to the
2310 spa_base64_to_bits() function which I hope fixes it.
2313 36. The daemon start-up calls getloadavg() while still root for those OS that
2314 need the first call to be done as root, but it missed one case: when
2315 deliver_queue_load_max is set with deliver_drop_privilege. This is
2316 necessary for the benefit of the queue runner, because there is no re-exec
2317 when deliver_drop_privilege is set.
2319 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2320 was in use, was not putting the data itself into the right store pool;
2321 consequently, it could be overwritten for a subsequent message in the same
2322 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2325 38. Sometimes the final signoff response after QUIT could fail to get
2326 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2327 < 0 before doing a fflush(). This bug looks as though it goes back to the
2328 introduction of TLS in release 3.20, but "sometimes" must have been rare
2329 because the tests only now provoked it.
2335 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2336 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2337 Now it returns an integer. A number of calls were still expecting a T/F
2338 return. Fortuitously, in all cases, the tests worked in OK situations,
2339 which is the norm. However, things would have gone wrong on any write
2340 failures on the smtp file descriptor. This function is used when sending
2341 messages over SMTP and also when doing verify callouts.
2343 2. When Exim is called to do synchronous delivery of a locally submitted
2344 message (the -odf or -odi options), it no longer closes stderr before doing
2347 3. Implemented the mua_wrapper option.
2349 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2351 5. Implemented the functions header_remove(), header_testname(),
2352 header_add_at_position(), and receive_remove_recipient(), and exported them
2355 6. If an ACL "warn" statement specified the addition of headers, Exim already
2356 inserted X-ACL-Warn: at the start if there was no header name. However, it
2357 was not making this test for the second and subsequent header lines if
2358 there were newlines in the string. This meant that an invalid header could
2359 be inserted if Exim was badly configured.
2361 7. Allow an ACL "warn" statement to add header lines at the start or after all
2362 the Received: headers, as well as at the end.
2364 8. Added the rcpt_4xx retry error code.
2366 9. Added postmaster_mailfrom=xxx to callout verification option.
2368 10. Added mailfrom=xxxx to the callout verification option, for verify=
2371 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2372 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2374 12. Inserted some casts to stop certain compilers warning when using pointer
2375 differences as field lengths or precisions in printf-type calls (mostly
2376 affecting debugging statements).
2378 13. Added optional readline() support for -be (dynamically loaded).
2380 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2381 same clock tick as a message's arrival, so that its received time was the
2382 same as the "first fail" time on the retry record, and that message
2383 remained on the queue past the ultimate address timeout, every queue runner
2384 would try a delivery (because it was past the ultimate address timeout) but
2385 after another failure, the ultimate address timeout, which should have then
2386 bounced the address, did not kick in. This was a "< instead of <=" error;
2387 in most cases the first failure would have been in the next clock tick
2388 after the received time, and all would be well.
2390 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2391 being recognized when the domain list was tested by the match_domain
2392 condition in an expansion string.
2394 16. Added the ${str2b64: operator.
2396 17. Exim was always calling setrlimit() to set a large limit for the number of
2397 processes, without checking whether the existing limit was already
2398 adequate. (It did check for the limit on file descriptors.) Furthermore,
2399 errors from getrlimit() and setrlimit() were being ignored. Now they are
2400 logged to the main and panic logs, but Exim does carry on, to try to do its
2401 job under whatever limits there are.
2403 18. Imported PCRE 5.0.
2405 19. Trivial typo in log message " temporarily refused connection" (the leading
2408 20. If the log selector return_path_on_delivery was set and an address was
2409 redirected to /dev/null, the delivery process crashed because it assumed
2410 that a return path would always be set for a "successful" delivery. In this
2411 case, the whole delivery is bypassed as an optimization, and therefore no
2414 21. Internal re-arrangement: the function for sending a challenge and reading
2415 a response while authentication was assuming a zero-terminated challenge
2416 string. It's now changed to take a pointer and a length, to allow for
2417 binary data in such strings.
2419 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2421 23. Exim was not respecting finduser_retries when seeking the login of the
2422 uid under which it was called; it was always trying 10 times. (The default
2423 setting of finduser_retries is zero.) Also, it was sleeping after the final
2424 failure, which is pointless.
2426 24. Implemented tls_on_connect_ports.
2428 25. Implemented acl_smtp_predata.
2430 26. If the domain in control=submission is set empty, Exim assumes that the
2431 authenticated id is a complete email address when it generates From: or
2432 Sender: header lines.
2434 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2435 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2436 chown and chgrp in /bin and hostname in /usr/bin.
2438 28. Exim was keeping the "process log" file open after each use, just as it
2439 does for the main log. This opens the possibility of it remaining open for
2440 long periods when the USR1 signal hits a daemon. Occasional processlog
2441 errors were reported, that could have been caused by this. Anyway, it seems
2442 much more sensible not to leave this file open at all, so that is what now
2445 29. The long-running daemon process does not normally write to the log once it
2446 has entered its main loop, and it closes the log before doing so. This is
2447 so that log files can straightforwardly be renamed and moved. However,
2448 there are a couple of unusual error situations where the daemon does write
2449 log entries, and I had neglected to close the log afterwards.
2451 30. The text of an SMTP error response that was received during a remote
2452 delivery was being truncated at 512 bytes. This is too short for some of
2453 the long messages that one sometimes sees. I've increased the limit to
2456 31. It is now possible to make retry rules that apply only when a message has a
2457 specific sender, in particular, an empty sender.
2459 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2460 it possible to be selective about when SMTP synchronization is enforced.
2462 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2464 32. Implemented hosts_connection_nolog.
2466 33. Added an ACL for QUIT.
2468 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2471 35. Added mailbox_size and mailbox_filecount to appendfile.
2473 36. Added control = no_multiline_responses to ACLs.
2475 37. There was a bug in the logic of the code that waits for the clock to tick
2476 in the case where the clock went backwards by a substantial amount such
2477 that the microsecond fraction of "now" was more than the microsecond
2478 fraction of "then" (but the whole seconds number was less).
2480 38. Added support for the libradius Radius client library this is found on
2481 FreeBSD (previously only the radiusclient library was supported).
2487 1. When certain lookups returned multiple values in the form name=value, the
2488 quoting of the values was not always being done properly. Specifically:
2489 (a) If the value started with a double quote, but contained no whitespace,
2491 (b) If the value contained whitespace other than a space character (i.e.
2492 tabs or newlines or carriage returns) it was not quoted.
2493 This fix has been applied to the mysql and pgsql lookups by writing a
2494 separate quoting function and calling it from the lookup code. The fix
2495 should probably also be applied to nisplus, ibase and oracle lookups, but
2496 since I cannot test any of those, I have not disturbed their existing code.
2498 2. A hit in the callout cache for a specific address caused a log line with no
2499 reason for rejecting RCPT. Now it says "Previous (cached) callout
2500 verification failure".
2502 3. There was an off-by-one bug in the queryprogram router. An over-long
2503 return line was truncated at 256 instead of 255 characters, thereby
2504 overflowing its buffer with the terminating zero. As well as fixing this, I
2505 have increased the buffer size to 1024 (and made a note to document this).
2507 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2508 when Exim is waiting for an SMTP response from a remote server, Exim
2509 restarts its select() call on the socket, thereby resetting its timeout.
2510 This is not a problem when such interrupts are rare. Somebody set up a cron
2511 job to run exiwhat every 2 minutes, which is less than the normal select()
2512 timeout (5 or 10 minutes). This meant that the select() timeout never
2513 kicked in because it was always reset. I have fixed this by comparing the
2514 time when an interrupt arrives with the time at the start of the first call
2515 to select(). If more time than the timeout has elapsed, the interrupt is
2516 treated as a timeout.
2518 5. Some internal re-factoring in preparation for the addition of Sieve
2519 extensions (by MH). In particular, the "personal" test is moved to a
2520 separate function, and given an option for scanning Cc: and Bcc: (which is
2521 not set for Exim filters).
2523 6. When Exim created an email address using the login of the caller as the
2524 local part (e.g. when creating a From: or Sender: header line), it was not
2525 quoting the local part when it contained special characters such as @.
2527 7. Installed new OpenBSD configuration files.
2529 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2530 try to make them clearer.
2532 9. Callout options, other than the timeout value, were being ignored when
2533 verifying sender addresses in header lines. For example, when using
2535 verify = header_sender/callout=no_cache
2537 the cache was (incorrectly) being used.
2539 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2540 only the Cygwin environment.
2542 11. When return_path_on_delivery was set as a log selector, if different remote
2543 addresses in the same message used different return paths and parallel
2544 remote delivery occurred, the wrong values would sometimes be logged.
2545 (Whenever a remote delivery process finished, the return path value from
2546 the most recently started remote delivery process was logged.)
2548 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2549 lines when AUTH and/or TLS are in use. This is the "received protocol"
2550 field. Exim used to use "asmtp" for authenticated SMTP, without any
2551 indication (in the protocol name) for TLS use. Now it follows the RFC and
2552 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2553 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2554 names appear in log lines as well as in Received: header lines.
2556 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2557 extensions, and comparison tests, and to fix some bugs.
2559 14. Changes to the "personal" filter test:
2561 (1) The test was buggy in that it was just doing the equivalent of
2562 "contains" tests on header lines. For example, if a user's address was
2563 anne@some.where, the "personal" test would incorrectly be true for
2565 To: susanne@some.where
2567 This test is now done by extracting each address from the header in turn,
2568 and checking the entire address. Other tests that are part of "personal"
2569 are now done using regular expressions (for example, to check local parts
2570 of addresses in From: header lines).
2572 (2) The list of non-personal local parts in From: addresses has been
2573 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2574 taken from the Sieve specification recommendations.
2576 (3) If the message contains any header line starting with "List-" it is
2577 treated as non-personal.
2579 (4) The test for "circular" in the Subject: header line has been removed
2580 because it now seems ill-conceived.
2582 15. Minor typos in src/EDITME comments corrected.
2584 16. Installed latest exipick from John Jetmore.
2586 17. If headers_add on a router specified a text string that was too long for
2587 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2588 of string_sprintf() is now avoided.
2590 18. $message_body_size was not set (it was always zero) when running the DATA
2591 ACL and the local_scan() function.
2593 19. For the "mail" command in an Exim filter, no default was being set for
2594 the once_repeat time, causing a random time value to be used if "once" was
2595 specified. (If the value happened to be <= 0, no repeat happened.) The
2596 default is now 0s, meaning "never repeat". The "vacation" command was OK
2597 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2598 (I found it when inspecting the code).
2600 20. There is now an overall timeout for performing a callout verification. It
2601 defaults to 4 times the callout timeout, which applies to individual SMTP
2602 commands during the callout. The overall timeout applies when there is more
2603 than one host that can be tried. The timeout is checked before trying the
2604 next host. This prevents very long delays if there are a large number of
2605 hosts and all are timing out (e.g. when the network connections are timing
2606 out). The value of the overall timeout can be changed by specifying an
2607 additional sub-option for "callout", called "maxwait". For example:
2609 verify = sender/callout=5s,maxwait=20s
2611 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2612 to the end before writing, but this should make it even safer).
2614 22. Exim was forgetting that it had advertised PIPELINING for the second and
2615 subsequent messages on an SMTP connection. It was also not resetting its
2616 memory on STARTTLS and an internal HELO.
2618 23. When Exim logs an SMTP synchronization error within a session, it now
2619 records whether PIPELINING has been advertised or not.
2621 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2622 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2623 rather than long int.
2625 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2627 26. Added the never_mail option to autoreply.
2633 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2634 crash if the getsockname() call failed; this can happen if a connection is
2635 closed very soon after it is established. The problem was simply in the
2636 order in which certain operations were done, causing Exim to try to write
2637 to the SMTP stream before it had set up the file descriptor. The bug has
2638 been fixed by making things happen in the correct order.
2644 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2645 before the connection was closed, thus losing the rejection response.
2647 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2648 some early Solaris releases, but causes trouble in current releases where
2649 socklen_t is defined.
2651 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2654 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2657 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2658 that releasing the top of it at the end releases what was used for sub-
2659 expansions (unless the block got too big). However, discard this block if
2660 the first thing is a variable or header, so that we can use its block when
2661 it is dynamic (useful for very large $message_headers, for example).
2663 6. Lookups now cache *every* query, not just the most recent. A new, separate
2664 store pool is used for this. It can be recovered when all lookup caches are
2665 flushed. Lookups now release memory at the end of their result strings.
2666 This has involved some general refactoring of the lookup sources.
2668 7. Some code has been added to the store_xxx() functions to reduce the amount
2669 of flapping under certain conditions.
2671 8. log_incoming_interface used to affect only the <= reception log lines. Now
2672 it causes the local interface and port to be added to several more SMTP log
2673 lines, for example "SMTP connection from", and rejection lines.
2675 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2677 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2679 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2680 could be overwritten at the end of the current message (or the start of a
2681 new message if it was set in a HELO ACL). The value is now preserved for
2682 the duration of the SMTP connection.
2684 12. If a transport had a headers_rewrite setting, and a matching header line
2685 contained an unqualified address, that address was qualified, even if it
2686 did not match any rewriting rules. The underlying bug was that the values
2687 of the flags that permit the existence of unqualified sender and recipient
2688 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2689 non-local messages, and by -bnq for local messages) were not being
2690 preserved with the message after it was received.
2692 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2693 "next input=" as part of the text comprising the host identity instead of
2694 the correct text. The code was using the same buffer for two different
2695 strings. However, depending on which order the printing function evaluated
2696 its arguments, the bug did not always show up. Under Linux, for example, my
2697 test suite worked just fine.
2699 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2700 doesn't work with some older versions of Perl. It has been changed to "my",
2701 which in any case is probably the better facility to use.
2703 15. A really picky compiler found some instances of statements for creating
2704 error messages that either had too many or two few arguments for the format
2707 16. The size of the buffer for calls to the DNS resolver has been increased
2708 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2709 for addresses that have a lot of PTR records. This alleviates a problem; it
2710 does not fully solve it.
2712 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2713 buffer now truncates the list and logs the incident, which is the same
2714 action as happens when Exim is looking up a host name and its aliases.
2715 Previously in this situation something unpredictable would happen;
2716 sometimes it was "internal error: store_reset failed".
2718 18. If a server dropped the connection unexpectedly when an Exim client was
2719 using GnuTLS and trying to read a response, the client delivery process
2720 crashed while trying to generate an error log message.
2722 19. If a "warn" verb in an ACL added multiple headers to a message in a single
2723 string, for example:
2725 warn message = H1: something\nH2: something
2727 the text was added as a single header line from Exim's point of view
2728 though it ended up OK in the delivered message. However, searching for the
2729 second and subsequent header lines using $h_h2: did not work. This has been
2730 fixed. Similarly, if a system filter added multiple headers in this way,
2731 the routers could not see them.
2733 20. Expanded the error message when iplsearch is called with an invalid key to
2734 suggest using net-iplsearch in a host list.
2736 21. When running tests using -bh, any delays imposed by "delay" modifiers in
2737 ACLs are no longer actually imposed (and a message to that effect is
2740 22. If a "gecos" field in a passwd entry contained escaped characters, in
2741 particular, if it contained a \" sequence, Exim got it wrong when building
2742 a From: or a Sender: header from that name. A second bug also caused
2743 incorrect handling when an unquoted " was present following a character
2744 that needed quoting.
2746 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2747 was not being matched caselessly.
2749 24. Arranged for all hyphens in the exim.8 source to be escaped with
2752 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2753 properly. Recipient callout cache records were still being keyed to include
2754 the sender, even when use_sender was set false. This led to far more
2755 callouts that were necessary. The sender is no longer included in the key
2756 when use_sender is false.
2758 26. Added "control = submission" modifier to ACLs.
2760 27. Added the ${base62d: operator to decode base 62 numbers.
2762 28. dnsdb lookups can now access SRV records.
2764 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2765 the configuration file.
2767 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2768 (-v) mode. This makes the output for a verbose queue run more intelligible.
2770 31. Added a use_postmaster feature to recipient callouts.
2772 32. Added the $body_zerocount variable, containing the number of binary zero
2773 bytes in the message body.
2775 33. The time of last modification of the "new" subdirectory is now used as the
2776 "mailbox time last read" when there is a quota error for a maildir
2779 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2781 35. Added +ignore_unknown as a special item in host lists.
2783 36. Code for decoding IPv6 addresses in host lists is now included, even if
2784 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2785 address was recognized as an IP address, but was then not correctly decoded
2786 into binary, causing unexpected and incorrect effects when compared with
2793 1. Very minor rewording of debugging text in manualroute to say "list of
2794 hosts" instead of "hostlist".
2796 2. If verify=header_syntax was set, and a header line with an unqualified
2797 address (no domain) and a large number of spaces between the end of the
2798 name and the colon was received, the reception process suffered a buffer
2799 overflow, and (when I tested it) crashed. This was caused by some obsolete
2800 code that should have been removed. The fix is to remove it!
2802 3. When running in the test harness, delay a bit after writing a bounce
2803 message to get a bit more predictability in the log output.
2805 4. Added a call to search_tidyup() just before forking a reception process. In
2806 theory, someone could use a lookup in the expansion of smtp_accept_max_
2807 per_host which, without the tidyup, could leave open a database connection.
2809 5. Added the variables $recipient_data and $sender_data which get set from a
2810 lookup success in an ACL "recipients" or "senders" condition, or a router
2811 "senders" option, similar to $domain_data and $local_part_data.
2813 6. Moved the writing of debug_print from before to after the "senders" test
2816 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2817 problems for message scanning, either using a data ACL, or using
2818 local_scan() because the Received: header was not generated till after they
2819 were called (in order to set the time as the time of reception completion).
2820 I have revised the way this works. The header is now generated after the
2821 body is received, but before the ACL or local_scan() are called. After they
2822 are run, the timestamp in the header is updated.
2828 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2829 before starting a queue runner without re-exec. This happened only when
2830 deliver_drop_privilege was set or when the Exim user was set to root. The
2831 effect of the bug was that timeouts during subsequent deliveries caused
2832 crashes instead of being properly handled. The handler is now left at its
2833 default (and expected) setting.
2835 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2836 message, again when deliver_drop_privilege is set or Exim is run as root.
2837 The bug described in (1) was not present in this case, but the tidying up
2838 of the other signals was missing. I have made the two cases consistent.
2840 3. The ignore_target_hosts setting on a manualroute router was being ignored
2841 for hosts that were looked up using the /MX notation.
2843 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2846 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2847 operated on the sender address. After changing the $sender_address to <>
2848 for the sender address verify, Exim was re-instated it as the original
2849 (before rewriting) address, but remembering that it had rewritten it, so it
2850 wasn't rewriting it again. This bug also had the effect of breaking the
2851 sender address verification caching when the sender address was rewritten.
2853 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2854 This has been changed so that if the ip literal address matches
2855 ignore_target_hosts, the router declines.
2857 7. Added expansion conditions match_domain, match_address, and match_local_
2858 part (NOT match_host).
2860 8. The placeholder for the Received: header didn't have a length field set.
2862 9. Added code to Exim itself and to exim_lock to test for a specific race
2863 condition that could lead to file corruption when using MBX delivery. The
2864 issue is with the lockfile that is created in /tmp. If this file is removed
2865 after a process has opened it but before that process has acquired a lock,
2866 there is the potential for a second process to recreate the file and also
2867 acquire a lock. This could lead to two Exim processes writing to the file
2868 at the same time. The added code performs the same test as UW imapd; it
2869 checks after acquiring the lock that its file descriptor still refers to
2870 the same named file.
2872 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2873 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2876 11. Added the smtp_active_hostname option. If used, this will typically be made
2877 to depend on the incoming interface address. Because $interface_address is
2878 not set up until the daemon has forked a reception process, error responses
2879 that can happen earlier (such as "too many connections") no longer contain
2882 12. If an expansion in a condition on a "warn" statement fails because a lookup
2883 defers, the "warn" statement is abandoned, and the next ACL statement is
2884 processed. Previously this caused the whole ACL to be aborted.
2886 13. Added the iplsearch lookup type.
2888 14. Added ident_timeout as a log selector.
2890 15. Added tls_certificate_verified as a log selector.
2892 16. Added a global option tls_require_ciphers (compare the smtp transport
2893 option of the same name). This controls incoming TLS connections.
2895 17. I finally figured out how to make tls_require_ciphers do a similar thing
2896 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2897 before starting the TLS session.
2899 18. Tabs are now shown as \t in -bP output.
2901 19. If the log selector return_path_on_delivery was set, Exim crashed when
2902 bouncing a message because it had too many Received: header lines.
2904 20. If two routers both had headers_remove settings, and the first one included
2905 a superfluous trailing colon, the final name in the first list and the
2906 first name in the second list were incorrectly joined into one item (with a
2907 colon in the middle).
2913 1. Added -C and -D options to the exinext utility, mainly to make it easier
2914 to include in the automated testing, but these could be helpful when
2915 multiple configurations are in use.
2917 2. The exinext utility was not formatting the output nicely when there was
2918 an alternate port involved in the retry record key, nor when there was a
2919 message id as well (for retries that were specific to a specific message
2920 and a specific host). It was also confused by IPv6 addresses, because of
2921 the additional colons they contain. I have fixed the IPv4 problem, and
2922 patched it up to do a reasonable job for IPv6.
2924 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2925 delivery, the log line now contains "pipelined" if PIPELINING was used.
2927 4. An SMTP transport process used to panic and die if the bind() call to set
2928 an explicit outgoing interface failed. This has been changed; it is now
2929 treated in the same way as a connect() failure.
2931 5. A reference to $sender_host_name in the part of a conditional expansion
2932 that was being skipped was still causing a DNS lookup. This no longer
2935 6. The def: expansion condition was not recognizing references to header lines
2936 that used bh_ and bheader_.
2938 7. Added the _cache feature to named lists.
2940 8. The code for checking quota_filecount in the appendfile transport was
2941 allowing one more file than it should have been.
2943 9. For compatibility with Sendmail, the command line option
2951 and sets the incoming protocol and host name (for trusted callers). The
2952 host name and its colon can be omitted when only the protocol is to be set.
2953 Note the Exim already has two private options, -pd and -ps, that refer to
2954 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2955 "s", but I don't think that's a major issue.
2957 10. A number of refactoring changes to the code, none of which should affect
2960 (a) The number of logging options was getting close to filling up the
2961 32-bit word that was used as a bit map. I have split them into two classes:
2962 those that are passed in the argument to log_write(), and those that are
2963 only ever tested independently outside of that function. These are now in
2964 separate 32-bit words, so there is plenty of room for expansion again.
2965 There is no change in the user interface or the logging behaviour.
2967 (b) When building, for example, log lines, the code previously used a
2968 macro that called string_cat() twice, in order to add two strings. This is
2969 not really sufficiently general. Furthermore, there was one instance where
2970 it was actually wrong because one of the argument was used twice, and in
2971 one call a function was used. (As it happened, calling the function twice
2972 did not affect the overall behaviour.) The macro has been replaced by a
2973 function that can join an arbitrary number of extra strings onto a growing
2976 (c) The code for expansion conditions now uses a table and a binary chop
2977 instead of a serial search (which was left over from when there were very
2978 few conditions). Also, it now recognizes conditions like "pam" even when
2979 the relevant support is not compiled in: a suitably worded error message is
2980 given if an attempt is made to use such a condition.
2982 11. Added ${time_interval:xxxxx}.
2984 12. A bug was causing one of the ddress fields not to be passed back correctly
2985 from remote delivery subprocesses. The field in question was not being
2986 subsequently used, so this caused to problems in practice.
2988 13. Added new log selectors queue_time and deliver_time.
2990 14. Might have fixed a bug in maildirsizefile handling that threw up
2991 "unexpected character" debug warnings, and recalculated the data
2992 unnecessarily. In any case, I expanded the warning message to give more
2995 15. Added the message "Restricted characters in address" to the statements in
2996 the default ACL that block characters like @ and % in local parts.
2998 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2999 Three changes have been made:
3001 (a) There was a serious bug; a negative response to MAIL caused the whole
3002 recipient domain to be cached as invalid, thereby blocking all messages
3003 to all local parts at the same domain, from all senders. This bug has
3004 been fixed. The domain is no longer cached after a negative response to
3005 MAIL if the sender used is not empty.
3007 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3010 (c) A new callout option, "use_sender" has been added for people who want
3011 the modified behaviour.
3017 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3020 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3021 indeed breaks things for older releases.
3023 3. Added additional logging to the case where there is a problem reading data
3024 from a filter that is running in a subprocess using a pipe, in order to
3025 try to track down a specific problem.
3027 4. Testing facility fudge: when running in the test harness and attempting
3028 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3029 getting "No route to host". Convert this to a timeout.
3031 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3034 6. Some OS don't have socklen_t but use size_t instead. This affects the
3035 fifth argument of getsockopt() amongst other things. This is now
3036 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3037 can be set for individual OS. I have set it for SunOS5, OSF1, and
3038 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3039 some earlier ones do not.
3041 7. Change 4.30/15 was not doing the test caselessly.
3043 8. The standard form for an IPv6 address literal was being rejected by address
3044 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3045 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3046 this, as well as the form without the "IPv6" on the front (but only when
3047 address literals are enabled, of course).
3049 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3051 10. Exim crashed if a message with an empty sender address specified by -f
3052 encountered a router with an errors_to setting. This could be provoked only
3053 by a command such as
3057 where an empty string was supplied; "<>" did not hit this bug.
3059 11. Installed PCRE release 4.5.
3061 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3062 remained set. It is now erased.
3064 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3065 times from message ids (which are base 36 rather than the normal 62).
3067 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3068 were being counted as actual protocol errors, and logged if the log
3069 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3070 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3071 and DATA following a set of rejected RCPTs do not count as protocol errors.
3072 In other words, Exim assumes they were pipelined, though this may not
3073 actually be the case. Of course, in all cases the client gets an
3074 appropriate error code.
3076 15. If a lookup fails in an ACL condition, a message about the failure may
3077 be available; it is used if testing the ACL cannot continue, because most
3078 such messages specify what the cause of the deferral is. However, some
3079 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3080 that caused an old message to be retained and used if a later statement
3081 caused a defer, replacing the real cause of the deferral.
3083 16. If an IP address had so many PTR records that the DNS lookup buffer
3084 was not large enough to hold them, Exim could crash while trying to process
3085 the truncated data. It now detects and logs this case.
3087 17. Further to 4.21/58, another change has been made: if (and only if) the
3088 first line of a message (the first header line) ends with CRLF, a bare LF
3089 in a subsequent header line has a space inserted after it, so as not to
3090 terminate the header.
3092 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3093 unnecessarily, used atoi() instead of strtol(), and didn't check the
3094 termination when getting file sizes from file names by regex.
3096 19. Completely re-implemented the support for maildirsize files, in the light
3097 of a number of problems with the previous contributed implementation
3098 (4.30/29). In particular:
3100 . If the quota is zero, the maildirsize file is maintained, but no quota is
3103 . If the maildir directory does not exist, it is created before any attempt
3104 to write a maildirsize file.
3106 . The quota value in the file is just a cache; if the quota is changed in
3107 the transport, the new value overrides.
3109 . A regular expression is available for excluding directories from the
3112 20. The autoreply transport checks the characters in options that define the
3113 message's headers; it allows continued headers, but it was checking with
3114 isspace() after an embedded newline instead of explicitly looking for a
3117 21. If all the "regular" hosts to which an address was routed had passed their
3118 expiry times, and had not reached their retry times, the address was
3119 bounced, even if fallback hosts were defined. Now Exim should go on to try
3122 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3123 equivalent code in the SMTP transport. Some hosts send humungous responses
3124 to HELO/EHLO, more than 1024 it seems.
3126 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3127 gives compiler warnings in some environments. I've now done it "properly",
3130 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3131 (because of problems with the built-in one) was declared to return uschar *
3132 instead of char *, causing compiler failure.
3134 25. Fixed a file descriptor leak when processing alias/forward files.
3136 26. Fixed a minor format string issue in dbfn.c.
3138 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3140 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3141 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3144 29. When a sender address is verified, it is cached, to save repeating the test
3145 when there is more than one recipient in a message. However, when the
3146 verification involves a callout, it is possible for different callout
3147 options to be set for different recipients. It is too complicated to keep
3148 track of this in the cache, so now Exim always runs a verification when a
3149 callout is required, relying on the callout cache for the optimization.
3150 The overhead is duplication of the address routing, but this should not be
3153 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3154 to be verified with callout=postmaster, and the main callout worked but the
3155 postmaster check failed, the verification correctly failed. However, if a
3156 subsequent RCPT command asked for sender verification *without* the
3157 postmaster check, incorrect caching caused this verification also to fail,
3160 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3161 it was not caching the DNS options (qualify_single, search_parents) that
3162 were used when the lookup failed. A subsequent lookup with different
3163 options therefore always gave the same answer, though there were cases
3164 where it should not have. (Example: a "domains = !$mx_any" option on a
3165 dnslookup router: the "domains" option is always processed without any
3166 widening, but the router might have qualify_single set.) Now Exim uses the
3167 cached value only when the same options are set.
3169 32. Added John Jetmore's "exipick" utility to the distribution.
3171 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3172 than a timeout (e.g. a certificate is required, and is not provided), an
3173 Exim server now closes the connection immediately. Previously it waited for
3174 the client to close - but if the client is SSL, it seems that they each
3175 wait for each other, leading to a delay before one of them times out.
3177 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3178 maintained 0.8.x compatibility because I don't think many are using it, and
3179 it is clearly obsolete.
3181 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3184 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3185 client certificate was expired. A simple patch fixes this, though I don't
3186 understand the full logic of why the verify callback is called multiple
3189 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3190 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3191 which causes problems with some clients (such as the Certicom SSL Plus
3192 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3193 disables the coutermeasure allowing Eudora to connect."
3195 38. Exim was not checking that a write() to a log file succeeded. This could
3196 lead to Bad Things if a log got too big, in particular if it hit a file
3197 size limit. Exim now panics and dies if it cannot write to a log file, just
3198 as it does if it cannot open a log file.
3200 39. Modified OS/Makefile-Linux so that it now contains
3202 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3204 The two -D definitions ensure that Exim is compiled with large file
3205 support, which makes it possible to handle log files that are bigger than
3208 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3209 instance) a domain was checked against a named list that involved a lookup,
3210 causing $domain_data to be set, then another domain was checked against the
3211 same list, then the first domain was re-checked, the value of $domain_data
3212 after the final check could be wrong. In particular, if the second check
3213 failed, it could be set empty. This bug probably also applied to
3216 41. The strip_trailing_dot option was not being applied to the address given
3217 with the -f command-line option.
3219 42. The code for reading a message's header from the spool was incrementing
3220 $received_count, but never initializing it. This meant that the value was
3221 incorrect (doubled) while delivering a message in the same process in which
3222 it was received. In the most common configuration of Exim, this never
3223 happens - a fresh exec is done - but it can happen when
3224 deliver_drop_privilege is set.
3226 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3227 it now includes up to 150 characters of the unexpected data in the log
3230 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3231 and building data strings. The size of both of these buffers was 10 000
3232 bytes - far larger than anybody would *ever* want, thought I. Needless to
3233 say, somebody hit the limit. I have increased the maximum line length to
3234 20 000 and the maximum data length of concatenated lines to 100 000. I have
3235 also fixed two bugs, because there was no checking on these buffers. Tsk,
3236 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3237 buffer is too small.
3239 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3240 lsearch lookups. Now it does.
3242 46. When parsing a route_list item in a manualroute router, a fixed-length
3243 buffer was used for the list of hosts. I made this 1024 bytes long,
3244 thinking that nobody would ever have a list of hosts that long. Wrong.
3245 Somebody had a whole pile of complicated expansion conditions, and the
3246 string was silently truncated, leading to an expansion error. It turns out
3247 that it is easier to change to an unlimited length (owing to other changes
3248 that have happened since this code was originally written) than to build
3249 structure for giving a limitation error. The length of the item that
3250 expands into the list of hosts is now unlimited.
3252 47. The lsearch lookup could not handle data where the length of text line was
3253 more than 4095 characters. Such lines were truncated, leading to shortened
3254 data being returned. It should now handle lines of any length.
3256 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3257 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3260 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3261 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3262 no longer happens in comment lines. A list of replacements is now placed
3263 at the head of all of the source files, except those whose only change is
3264 to replace PERL_COMMAND in the very first #! line.
3266 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3267 messages on the queue, with a bottom-up merge sort, using code contributed
3268 by Michael Haardt. This should make operations like -bp somewhat faster on
3269 large queues. It won't affect queue runners, except when queue_run_in_order
3272 51. Installed eximstats 1.31 in the distribution.
3274 52. Added support for SRV lookups to the dnslookup router.
3276 53. If an ACL referred to $message_body or $message_body_end, the value was not
3277 reset for any messages that followed in the same SMTP session.
3279 54. The store-handling optimization for building very long strings was not
3280 differentiating between the different store pools. I don't think this
3281 actually made any difference in practice, but I've tidied it.
3283 55. While running the routers to verify a sender address, $sender_address
3284 was still set to the sender address. This is wrong, because when routing to
3285 send a bounce to the sender, it would be empty. Therefore, I have changed
3286 it so that, while verifying a sender address, $sender_address is set to <>.
3287 (There is no change to what happens when verifying a recipient address.)
3289 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3290 target A or AAAA records (if not already returned) without resetting the
3291 qualify_single or search_parents options of the DNS resolver. These are
3292 inappropriate in this case because the targets of MX and SRV records must
3293 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3294 target that, when qualified, matched something in the local domain. These
3295 two options are now turned off when doing these lookups.
3297 57. It seems that at least some releases of Reiserfs (which does not have the
3298 concept of a fixed number of inodes) returns zero and not -1 for the
3299 number of available inodes. This interacted badly with check_spool_inodes,
3300 which assumed that -1 was the "no such thing" setting. What I have done is
3301 to check that the total number of inodes is greater than zero before doing
3302 the test of how many are available.
3304 58. When a "warn" ACL statement has a log_message modifier, the message is
3305 remembered, and not repeated. This is to avoid a lot of repetition when a
3306 message has many recipients that cause the same warning to be written.
3307 Howewer, Exim was preserving the list of already written lines for an
3308 entire SMTP session, which doesn't seem right. The memory is now reset if a
3309 new message is started.
3311 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3312 debugging output unless log_rewrite was also set.
3314 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3315 of (int)(handle) when we know that handle contains (void *)(-1).
3317 61. The Exim daemon panic-logs an error return when it closes the incoming
3318 connection. However "connection reset by peer" seems to be common, and
3319 isn't really an error worthy of noting specially, so that particular error
3322 62. When Exim is trying to find all the local interfaces, it used to panic and
3323 die if the ioctl to get the interface flags failed. However, it seems that
3324 on at least one OS (Solaris 9) it is possible to have an interface that is
3325 included in the list of interfaces, but for which you get a failure error
3326 for this call. This happens when the interface is not "plumbed" into a
3327 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3328 failure of the "get flags" call assumes that the interface is down.
3330 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3331 makes life easier for people who are doing arithmetic on fields extracted
3332 from dates, where you often get leading zeros that should not be
3333 interpreted as octal.
3335 64. Added qualify_domain to the redirect router, to override the global
3338 65. If a pathologically long header line contained very many addresses (the
3339 report of this problem mentioned 10 000) and each of them was rewritten,
3340 Exim could use up a very large amount of memory. (It kept on making new
3341 copies of the header line as it rewrote, and never released the old ones.)
3342 At the expense of a bit more processing, the header rewriting function has
3343 been changed so that it no longer eats memory in this way.
3345 66. The generation of the Received: header has been moved from the time that a
3346 message starts to be received, to the time that it finishes. The timestamp
3347 in the Received: header should now be very close to that of the <= log
3348 line. There are two side-effects of this change:
3350 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3351 logged header lines no longer include the local Received: line, because
3352 it has not yet been created. The same applies to a copy of the message
3353 that is returned to a non-SMTP sender when a message is rejected.
3355 (b) When a filter file is tested using -bf, no additional Received: header
3356 is added to the test message. After some thought, I decided that this
3359 This change does not affect the value of $received_for. It is still set
3360 after address rewriting, but before local_scan() is called.
3362 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3364 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3365 gave an unhelpful panic error message, and a defer error. I have managed to
3366 change this behaviour so that it now rejects any supplied certificate,
3367 which seems right, as the list of acceptable certificates is empty.
3369 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3370 gave an unhelpful defer error. I have not managed to make this reject any
3371 supplied certificates, but the error message it gives is "no certificate
3372 supplied", which is not helpful.
3374 70. exigrep's output now also includes lines that are not associated with any
3375 message, but which match the given pattern. Implemented by a patch from
3376 Martin Sluka, which also tidied up the Perl a bit.
3378 71. Recipient callout verification, like sender verification, was using <> in
3379 the MAIL FROM command. This isn't really the right thing, since the actual
3380 sender may affect whether the remote host accepts the recipient or not. I
3381 have changed it to use the actual sender in the callout; this means that
3382 the cache record is now keyed on a recipient/sender pair, not just the
3383 recipient address. There doesn't seem to be a real danger of callout loops,
3384 since a callout by the remote host to check the sender would use <>.
3385 [SEE ABOVE: changed after hitting problems.]
3387 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3388 temporary errors. However, in the case of such a code being given after
3389 the end of a data transmission (i.e. after ".") Exim was failing to write
3390 a retry record for the message. (Yes, there was some broken host that was
3391 actually sending 8xx at this point.)
3393 73. An unknown lookup type in a host list could cause Exim to panic-die when
3394 the list was checked. (An example that provoked this was putting <; in the
3395 middle of a list instead of at the start.) If this happened during a DATA
3396 ACL check, a -D file could be left lying around. This kind of configuration
3397 error no longer causes Exim to die; instead it causes a defer errror. The
3398 incident is still logged to the main and panic logs.
3400 74. Buglet left over from Exim 3 conversion. The message "too many messages
3401 in one connection" was written to the rejectlog but not the mainlog, except
3402 when address rewriting (yes!) was being logged.
3404 75. Added write_rejectlog option.
3406 76. When a system filter was run not as root (that is, when system_filter_user
3407 was set), the values of the $n variables were not being returned to the
3408 main process; thus, they were not subsequently available in the $sn
3411 77. Added +return_path_on_delivery log selector.
3413 78. A connection timeout was being treated differently from recipients deferred
3414 when testing hosts_max_try with a message that was older than the host's
3415 retry timeout. (The host should not be counted, thus allowing all hosts to
3416 be tried at least once before bouncing.) This may have been the cause of an
3417 occasionally reported bug whereby a message would remain on the queue
3418 longer than the retry timeout, but would be bounced if a delivery was
3419 forced. I say "may" because I never totally pinned down the problem;
3420 setting up timeout/retry tests is difficult. See also the next item.
3422 79. The ultimate address timeout was not being applied to errors that involved
3423 a combination of host plus message (for example, a timeout on a MAIL
3424 command). When an address resolved to a number of possible hosts, and they
3425 were not all tried for each delivery (e.g. because of hosts_max_try), a
3426 message could remain on the queue longer than the retry timeout.
3428 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3431 81. Fixed an obscure SMTP outgoing bug which required at least the following
3432 conditions: (a) there was another message waiting for the same server;
3433 (b) the server returned 5xx to all RCPT commands in the first message so
3434 that the message was not completed; (c) the server dropped the connection
3435 or gave a negative response to the RSET that Exim sends to abort the
3436 transaction. The observed case was a dropped connection after DATA that had
3437 been sent in pipelining mode. That is, the server had advertised PIPELINING
3438 but was not implementing it correctly. The effect of the bug was incorrect
3439 behaviour, such as trying another host, and this could lead to a crash.
3445 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3446 and daemon.c were passed as pointers to ints; they should have been
3447 pointers to socklen_t variables (which are typically unsigned ints).
3449 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3452 3. Fixed a really odd bug that affected only the testing scheme; patching a
3453 certain fixed string in the binary changed the value of another string that
3454 happened to be identical to the end of the original first string.
3456 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3457 name", it returns that address as the IP address. On some operating
3458 systems (e.g. Solaris), it also passes back the IP address string as the
3459 "host name". However, on others (e.g. Linux), it passes back an empty
3460 string. Exim wasn't checking for this, and was changing the host name to an
3461 empty string, assuming it had been canonicized.
3463 5. Although rare, it is permitted to have more than one PTR record for a given
3464 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3465 all the names associated with an address, because they do in Solaris.
3466 However, it seems that they do not in Linux for data that comes from the
3467 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3468 I found this out when I moved to a new Linux workstation and tried to run
3469 the Exim test suite.
3471 To get round this problem I have changed the code so that it now does its
3472 own call to the DNS to look up PTR records when searching for a host name.
3473 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3474 addresses that are only in /etc/hosts are still found.
3476 This behaviour is, however, controlled by an option called host_lookup_
3477 order, which defaults to "bydns:byaddr". If people want to use the other
3478 order, or indeed, just use one or the other means of lookup, they can
3479 specify it in this variable.
3481 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3482 some operating systems, this comes back from gethostbyaddr() as an empty
3483 string, and this is what Exim used to test for. However, it seems that in
3484 other systems, "." is yielded. Exim now tests for this case too.
3486 7. The values of check_spool_space and check_log_space are now held internally
3487 as a number of kilobytes instead of an absolute number of bytes. If a
3488 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3489 kilobyte. This means that much larger values can be stored.
3491 8. Exim monitor: an attempt to get the action menu when not actually pointing
3492 at a message produces an empty menu entitled "No message selected". This
3493 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3494 no entries in it ("Shell widget menu has zero width and/or height"). So I
3495 have added a single, blank menu entry in this case.
3497 9. Added ${quote_local_part.
3499 10. MIME decoding is now applied to the contents of Subject: header lines when
3502 11. Now that a reference to $sender_host_address automatically causes a reverse
3503 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3504 host lookup before query-style lookups in lists that might use this
3505 variable. This has therefore been abolished, and the "net-" prefix is no
3506 longer necessary for query-style lookups.
3508 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3509 have been a typo for LFLAGS, so it has been changed.
3511 13. The install script calls Exim with "-C /dev/null" in order to find the
3512 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3513 to be output. Howeve, since Exim outputs its version number before the
3514 error, it didn't break the script. It just looked ugly. I fixed this by
3515 always allowing "-C /dev/null" if the caller is root.
3517 14. Ignore overlarge ACL variable number when reading spool file - insurance
3518 against a later release with more variables having written the file.
3520 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3521 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3522 this, as well as the form without the "IPv6" on the front.
3524 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3525 OS/Makefile-Darwin file.
3527 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3528 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3530 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3531 variable. However, some OS use ints for inodes. I've added cast to long int
3532 to get rid of the compiler warning.
3534 19. I had forgotten to lock out "/../" in configuration file names when
3535 ALT_CONFIG_PREFIX was set.
3537 20. Routers used for verification do not need to specify transports. However,
3538 if such a router generated a host list, and callout was configured, Exim
3539 crashed, because it could not find a port number from the (non-existent)
3540 transport. It now assumes port 25 in this circumstance.
3542 21. Added the -t option to exigrep.
3544 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3545 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3546 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3547 them is set, LOOKUP_LSEARCH is forced.
3549 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3551 24. Added sender and host information to the "rejected by local_scan()" log
3552 line; previously there was no indication of these.
3554 25. Added .include_if_exists.
3556 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3557 Linux. It turns out that not all file systems support this. Apparently some
3558 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3559 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3560 not supported on the file descriptor, is now ignored when Exim is trying to
3561 sync a directory. This applies only to Linux.
3563 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3565 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3566 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3569 29. Added support for maildirsize files from supplied patch (modified a bit).
3571 30. The use of :fail: followed by an empty string could lead Exim to respond to
3572 sender verification failures with (e.g.):
3574 550 Verification failed for <xxx>
3575 550 Sender verify failed
3577 where the first response line was missing the '-' that indicates it is not
3578 the final line of the response.
3580 31. The loop for finding the name of the user that called Exim had a hardwired
3581 limit of 10; it now uses the value of finduser_retries, which is used for
3582 all other user lookups.
3584 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3587 33. Exim was neglecting to zero errno before one call of strtol() when
3588 expanding a string and expecting an integer value. On some systems this
3589 resulted in spurious "integer overflow" errors. Also, it was casting the
3590 result into an int without checking.
3592 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3593 did not work. The code looks as if it has *never* worked, though it appears
3594 to have been documented since at least releast 1.62. I have made it work.
3596 35. The "timeout_DNS" error in retry rules, also documented since at least
3597 1.62, also never worked. As it isn't clear exactly what this means, and
3598 clearly it isn't a major issue, I have abolished the feature by treating it
3599 as "timeout", and writing a warning to the main and panic logs.
3601 36. The display of retry rules for -brt wasn't always showing the error code
3604 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3605 timeout_connect_A, timeout_connect_MX.
3607 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3608 to the empty sender.
3610 39. The daemon was not analysing the content of -oX till after it had closed
3611 stderr and disconnected from the controlling terminal. This meant that any
3612 syntax errors were only noted on the panic log, and the return code from
3613 the command was 0. By re-arranging the code a little, I've made the
3614 decoding happen first, so such errors now appear on stderr, and the return
3615 code is 1. However, the actual setting up of the sockets still happens in
3616 the disconnected process, so errors there are still only recorded on the
3619 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3620 connections (as happens on some IP stacks) was logged at start up time as
3621 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3622 it from "IPv6 and IPv4", which means that two separate sockets are being
3625 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3626 says whether AF_INET or AF_INET6 was passed as an argument.
3628 42. Exiwhat output was messed up when time zones were included in log
3631 43. Exiwhat now gives more information about the daemon's listening ports,
3632 and whether -tls-on-connect was used.
3634 44. The "port" option of the smtp transport is now expanded.
3636 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3637 silently ignored. Now an error message is written to the main and panic
3640 46. There's a new ACL modifier called "logwrite" which writes to a log file
3641 as soon as it is encountered.
3643 47. Added $local_user_uid and $local_user_gid at routing time.
3645 48. Exim crashed when trying to verify a sender address that was being
3648 49. Exim was recognizing only a space character after ".include". It now also
3649 recognizes a tab character.
3651 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3652 extracting the relevant information from the specification. The man page no
3653 longer contains scrambled data for the -d option, and I've added a section
3654 at the front about calling Exim under different names.
3656 51. Added "extra_headers" argument to the "mail" command in filter files.
3658 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3661 53. Installed eximstats 1.29.
3663 54. Added transport_filter_timeout as a generic transport option.
3665 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3666 Cc: header lines. This was required by RFC 822, but it not required by RFC
3669 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3670 incoming messages that did not have them. Now it does so only if the
3671 message originates locally, that is, if there is no associated remote host
3672 address. When Resent- header lines are present, this applies to the Resent-
3673 lines rather than the non-Resent- lines.
3675 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3676 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3678 58. Messages for configuration errors now include the name of the main
3679 configuration file - useful now that there may be more than one file in a
3680 list (.included file names were always shown).
3682 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3683 for those rare installations that do not start the daemon as root or run it
3684 setuid root. I've cut out the call to initgroups() if the daemon is not
3687 60. The Exim user and group can now be bound into the binary as text strings
3688 that are looked up at the start of Exim's processing.
3690 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3692 62. Added $mailstore_basename variable.
3694 63. Installed patch to sieve.c from Michael Haardt.
3696 64. When Exim failed to open the panic log after failing to open the main log,
3697 the original message it was trying to log was written to stderr and debug
3698 output, but if they were not available (the usual case in production), it
3699 was lost. Now it is written to syslog before the two lines that record the
3700 failures to open the logs.
3702 65. Users' Exim filters run in subprocesses under the user's uid. It is
3703 possible for a "deliver" command or an alias in a "personal" command to
3704 provoke an address rewrite. If logging of address rewriting is configured,
3705 this fails because the process is not running as root or exim. There may be
3706 a better way of dealing with this, but for the moment (because 4.30 needs
3707 to be released), I have disabled address rewrite logging when running a
3708 filter in a non-root, non-exim process.
3714 1. The buildconfig auxiliary program wasn't quoting the value set for
3715 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3716 not defined. This bug was masked in 4.22 by the effect that was fixed in
3719 2. Some messages that were rejected after a message id was allocated were
3720 shown as "incomplete" by exigrep. It no longer does this for messages that
3721 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3723 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3724 have allow_domain_literals set, the ID did not get logged in the <= line.
3725 Domain literals are now always recognized in Message-ID: header lines.
3727 4. The first argument for a ${extract expansion item is the key name or field
3728 number. Leading and trailing spaces in this item were not being ignored,
3729 causing some misleading effects.
3731 5. When deliver_drop_privilege was set, single queue runner processes started
3732 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3733 same command in the process it spins off) were not dropping privilege.
3735 6. When the daemon running as "exim" started a queue runner, it always
3736 re-executed Exim in the spun-off process. This is a waste of effort when
3737 deliver_drop_privilege is set. The new process now just calls the
3738 queue-runner function directly.
3744 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3747 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3748 ignored. Though the use of -oP was forcing the writing of a pid file, it
3749 was always written to the default place.
3751 3. If the message "no IP address found for host xxxx" is generated during
3752 incoming verification, it is now followed by identification of the incoming
3753 connection (so you can more easily find what provoked it).
3755 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3757 5. Added some features to "harden" Exim a bit more against certain attacks:
3759 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3760 be put in Local/Makefile. This is like the never_users runtime option,
3761 but it cannot be overridden. The default setting is "root".
3763 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3764 prefix string with which any file named in a -C command line option
3767 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3768 is retained for -C and -D only if the caller of Exim is root. Without
3769 it, the exim user may also use -C and -D and retain privilege.
3771 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3772 command line option is disabled.
3774 6. Macro names set by the -D option must start with an upper case letter, just
3775 like macro names defined in the configuration file.
3777 7. Added "dereference=" facility to LDAP.
3779 8. Two instances of the typo "uknown" in the source files are fixed.
3781 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3782 the Configure-Makefile script screwed up while processing it.
3784 10. Incorporated PCRE 4.4.
3786 11. The SMTP synchronization check was not operating right at the start of an
3787 SMTP session. For example, it could not catch a HELO sent before the client
3788 waited for the greeting. There is now a check for outstanding input at the
3789 point when the greeting is written. Because of the duplex, asynchronous
3790 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3791 way, but not yet received, when the check is performed.
3793 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3794 on TCP/IP sockets, because this apparently causes some broken clients to
3797 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3798 unchanged) from the Cygwin maintainer.
3800 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3801 was supported instead of testing for mbx. Effectively a typo.
3803 15. The spa authenticator server code was not checking that the input it
3804 received was valid base64.
3806 16. The debug output line for the "set" modifier in ACLs was not showing the
3807 name of the variable that was being set.
3809 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3811 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3812 lookup on its own. Something else was needed to trigger the lookup. For
3813 example, a match in host_lookup or the need for a host name in a host list.
3814 Now, if $sender_host_name is referenced and the host name has not yet been
3815 looked up, a lookup is performed. If the lookup fails, the variable remains
3816 empty, and $host_lookup_failed is set to "1".
3818 19. Added "eqi" as a case-independent comparison operator.
3820 20. The saslauthd authentication condition could segfault if neither service
3821 nor realm was specified.
3823 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3824 error message that was logged was misleading, and incoming SMTP
3825 connections were dropped. The message is now more accurate, and temporary
3826 errors are given to SMTP connections.
3828 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3829 (except RSET) with a 421 error, until QUIT is received. However, it was
3830 failing to send a response to QUIT.
3832 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3833 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3834 ACL is now run after the helo_try_verify_hosts code.
3836 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3837 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3838 case-independent, so other case variants are also recognized). Apparently
3839 some systems use these upper case variants.
3841 25. If more than two messages were waiting for the same host, and a transport
3842 filter was specified for the transport, Exim sent two messages over the
3843 same TCP/IP connection, and then failed with "socket operation on non-
3844 socket" when it tried to send the third.
3846 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3848 27. The extern definition of crypt16() in expand.c was not being excluded when
3849 the OS had its own crypt16() function.
3851 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3852 as a preferred synonym for return_size_limit, both as an option and as an
3855 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3857 30. Changed the default configuration ACL to relax the local part checking rule
3858 for addresses that are not in any local domains. For these addresses,
3859 slashes and pipe symbols are allowed within local parts, but the sequence
3860 /../ is explicitly forbidden.
3862 31. SPA server authentication was not clearing the challenge buffer before
3865 32. log_message in a "warn" ACL statement was writing to the reject log as
3866 well as to the main log, which contradicts the documentation and doesn't
3867 seem right (because no rejection is happening). So I have stopped it.
3869 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3870 However, I am unable to do any testing of this.
3872 34. Fixed an infelicity in the appendfile transport. When checking directories
3873 for a mailbox, to see if any needed to be created, it was accidentally
3874 using path names with one or more superfluous leading slashes; tracing
3875 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3877 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3878 contents are added to the log line that is written for every discarded
3879 recipient. (Previously a log_message setting was ignored.)
3881 36. The ${quote: operator now quotes the string if it is empty.
3883 37. The install script runs exim in order to find its version number. If for
3884 some reason other than non-existence or emptiness, which it checks, it
3885 could not run './exim', it was installing it with an empty version number,
3886 i.e. as "exim-". This error state is now caught, and the installation is
3889 38. An argument was missing from the function that creates an error message
3890 when Exim fails to connect to the socket for saslauthd authentication.
3891 This could cause Exim to crash, or give a corrupted message.
3893 39. Added isip, isip4, and isip6 to ${if conditions.
3895 40. The ACL variables $acl_xx are now saved with the message, and can be
3896 accessed later in routers, transports, and filters.
3898 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3899 strings in the file are not string-expanded.
3901 42. If a MAIL command specified a SIZE value that was too large to fit into an
3902 int variable, the check against message_size_limit failed. Such values are
3903 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3904 day this will have to be increased, but I don't think I want to be around
3905 when emails are that large.
3912 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3913 iconv() is not standard in FreeBSD.
3915 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3916 IPv6 enabled. The observed symptom was a segmentation fault on return from
3917 the function os_common_find_running_interfaces() in src/os.c.
3919 3. In the check_special_case() function in daemon.c I had used "errno" as an
3920 argument name, which causes warnings on some systems. This was basically a
3921 typo, since it was named "eno" in the comments!
3923 4. The code that waits for the clock to tick (at a resolution of some fraction
3924 of a second) so as to ensure message-id uniqueness was always waiting for
3925 at least one whole tick, when it could have waited for less. [This is
3926 almost certainly not relevant at current processor speeds, where it is
3927 unlikely to ever wait at all. But we try to future-proof.]
3929 5. The function that sleeps for a time interval that includes fractions of a
3930 second contained a race. It did not block SIGALRM between setting the
3931 timer, and suspending (a couple of lines later). If the interval was short
3932 and the sigsuspend() was delayed until after it had expired, the suspension
3933 never ended. On busy systems this could lead to processes getting stuck for
3936 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3937 process, before it forks any delivery processes. The open lookup caching
3938 mechanism meant that the open file or database connection was passed into
3939 the delivery process. The problem was that delivery processes always tidy
3940 up cached lookup data. This could cause a problem for the next delivery
3941 process started by the queue runner, because the external queue runner
3942 process does not know about the closure. So the next delivery process
3943 still has data in the lookup cache. In the case of a file lookup, there was
3944 no problem because closing a file descriptor in a subprocess doesn't affect
3945 the parent. However, if the lookup was caching a connection to a database,
3946 the connection was closed, and the second delivery process was likely to
3947 see errors such as "PGSQL: query failed: server closed the connection
3948 unexpectedly". The problem has been fixed by closing all cached lookups
3949 in a queue runner before running a delivery process.
3951 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3952 seem to have the "const" qualifier which it has on other OS. I've
3955 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3956 *of the same type* (client or server) with the same public name that an
3957 error should be diagnosed.
3959 9. When Exim looked up a host name for an IP address, but failed to find the
3960 original IP address when looking up the host name (a safety check), it
3961 output the message "<ip address> does not match any IP for NULL", which was
3962 confusing, to say the least. The bug was that the host name should have
3963 appeared instead of "NULL".
3965 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3966 user that is built into the binary, and the -C or -D options is used, root
3967 privilege is dropped before the configuration file is read. In addition,
3968 logging is switched to stderr instead of the normal log files. If the
3969 configuration then re-defines the Exim user, the unprivileged environment
3970 is probably not what is expected, so Exim logs a panic warning message (but
3973 However, if deliver_drop_privilege is set, the unprivileged state may well
3974 be exactly what is intended, so the warning has been cut out in that case,
3975 and Exim is allowed to try to write to its normal log files.
3981 1. smtp_return_error_details was not giving details for temporary sender
3982 or receiver verification errors.
3984 2. Diagnose a configuration error if two authenticators have the same public
3987 3. Exim used not to create the message log file for a message until the first
3988 delivery attempt. This could be confusing when incoming messages were held
3989 for policy or load reasons. The message log file is now created at the time
3990 the message is received, and an initial "Received" line is written to it.
3992 4. The automatically generated man page for command line options had a minor
3993 bug that caused no ill effects; however, a more serious problem was that
3994 the procedure for building the man page automatically didn't always
3995 operate. Consequently, release 4.20 contains an out-of-date version. This
3996 shouldn't happen again.
3998 5. When building Exim with embedded Perl support, the script that builds the
3999 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4000 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4002 6. The freeze_tell option was not being used for messages that were frozen on
4003 arrival, either by an ACL or by local_scan().
4005 7. Added the smtp_incomplete_transaction log selector.
4007 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4008 was accepting AUTH without a new EHLO.
4010 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4011 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4014 10. Logging of TCP/IP connections (when configured) now happens in the main
4015 daemon process instead of the child process, so that the TCP/IP connection
4016 count is more accurate (but it can never be perfect).
4018 11. The use of "drop" in a nested ACL was not being handled correctly in the
4019 outer ACL. Now, if condition failure induced by the nested "drop" causes
4020 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4021 or "require"), the connection is dropped.
4023 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4024 that yield "discard" can now be used with an "accept" or a "discard" verb,
4025 but an error is generated for any others (because I can't see a useful way
4026 to define what should happen).
4028 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4029 are now processed in the same way as lines in the Exim configuration file.
4030 In particular, continuation lines are supported.
4032 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4034 15. Added -ti meaning -t -i.
4036 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4037 domains, and warn by logging if others are found.
4039 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4040 returned from GIFCONF when Exim is trying to find the list of interfaces on
4041 a host. The code in os.c has been modified to copy each ifreq to an aligned
4042 structure in all cases.
4044 Also, in some cases, the returned ifreq's were being copied to a 'struct
4045 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4046 means the last couple of bytes of an IPv6 address could be chopped if the
4047 ifreq contained only a normal sockaddr (14 bytes storage).
4049 18. Named domain lists were not supported in the hosts_treat_as_local option.
4050 An entry such as +xxxx was not recognized, and was treated as a literal
4053 19. Ensure that header lines added by a DATA ACL are included in the reject log
4054 if the ACL subsequently rejects the message.
4056 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4057 MD5 (which is deprecated).
4059 21. When testing a filter file using -bf, Exim was writing a message when it
4060 took the sender from a "From " line in the message, but it was not doing so
4061 when it took $return_path from a Return-Path: header line. It now does.
4063 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4064 with a valid header line field name (a series of printing characters
4065 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4067 23. Changed "disc" in the source to "disk" to conform to the documentation and
4068 the book and for uniformity.
4070 24. Ignore Sendmail's -Ooption=value command line item.
4072 25. When execve() failed while trying to run a command in a pipe transport,
4073 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4074 could be confused with a return value of 69 from the command itself. This
4075 has been changed to 127, the value the shell returns if it is asked to run
4076 a non-existent command. The wording for the related log line suggests a
4077 non-existent command as the problem.
4079 26. If received_header_text expands to an empty string, do not add a Received:
4080 header line to the message. (Well, it adds a token one on the spool, but
4081 marks it "old" so that it doesn't get used or transmitted.)
4083 27. Installed eximstats 1.28 (addition of -nt option).
4085 28. There was no check for failure on the call to getsockname() in the daemon
4086 code. This can fail if there is a shortage of resources on the system, with
4087 ENOMEM, for example. A temporary error is now given on failure.
4089 29. Contrary to the C standard, it seems that in some environments, the
4090 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4091 program. Exim now does this explicitly; it affects the formatting of
4092 timestamps using strftime().
4094 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4095 complaints. I've now initialized all the variables, to avoid this.
4097 32. Header lines added by a system filter were not being "seen" during
4098 transport-time rewrites.
4100 33. The info_callback() function passed to OpenSSL is set up with type void
4101 (*)(SSL *, int, int), as described somewhere. However, when calling the
4102 function (actually a macro) that sets it up, the type void(*)() is
4103 expected. I've put in a cast to prevent warnings from picky compilers.
4105 34. If a DNS black list lookup found a CNAME record, but there were no A
4106 records associated with the domain it pointed at, Exim crashed.
4108 35. If a DNS black list lookup returned more than one A record, Exim ignored
4109 all but the first. It now scans all returned addresses if a particular IP
4110 value is being sought. In this situation, the contents of the
4111 $dnslist_value variable are a list of all the addresses, separated by a
4114 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4115 to accept a host name and all its aliases if the forward lookup for any of
4116 them yielded the IP address of the incoming connection. Now it accepts only
4117 those names whose forward lookup yields the correct IP address. Any other
4118 names are discarded. This closes a loophole whereby a rogue DNS
4119 administrator could create reverse DNS records to break through a
4120 wildcarded host restriction in an ACL.
4122 37. If a user filter or a system filter that ran in a subprocess used any of
4123 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4124 the wrong values were passed to the pipe command ($thisaddress had the
4125 value of $0, $0 had the value of $1, etc). This bug was introduced by
4126 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4128 38. Improved the line breaking for long SMTP error messages from ACLs.
4129 Previously, if there was no break point between 40 and 75 characters, Exim
4130 left the rest of the message alone. Two changes have been made: (a) I've
4131 reduced the minimum length to 35 characters; (b) if it can't find a break
4132 point between 35 and 75 characters, it looks ahead and uses the first one
4133 that it finds. This may give the occasional overlong line, but at least the
4134 remaining text gets split now.
4136 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4137 file descriptors might be low, and that setting 1000 would always raise it.
4138 It turns out that in some environments, the limit is already over 1000 and
4139 that lowering it causes trouble. So now Exim takes care not to decrease it.
4141 40. When delivering a message, the value of $return_path is set to $sender_
4142 address at the start of routing (routers may change the value). By an
4143 oversight, this default was not being set up when an address was tested by
4144 -bt or -bv, which affected the outcome if any router or filter referred to
4147 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4148 log only when "message" or "log_message" are set. However, if one of the
4149 conditions was an address verification, or a call to a nested ACL, the
4150 messages generated by the underlying test were being passed through. This
4151 no longer happens. The underlying message is available in $acl_verify_
4152 message for both "message" and "log_message" expansions, so it can be
4153 passed through if needed.
4155 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4156 new expansion $bh_ to give the encoded byte string without charset
4157 translation. Translation happens only if iconv() is available; HAVE_ICONV
4158 indicates this at build time. HEADERS_CHARSET gives the charset to
4159 translate to; headers_charset can change it in the configuration, and
4160 "headers charset" can change it in an individual filter file.
4162 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4163 that creates RFC 2047 encoded "words" labels them as that charset instead
4164 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4165 expansion operator; (ii) when Exim creates a From: line for a local
4166 message; (iii) when a header line is rewritten to include a "phrase" part.
4168 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4169 buggy, causing it to skip the first lines of messages whose message ID
4170 ended in 'D'. This would not have bitten before Exim release 4.14, because
4171 message IDs were unlikely to end in 'D' before then. The effect was to have
4172 incorrect size information for certain domains.
4174 45. #include "config.h" was missing at the start of the crypt16.c module. This
4175 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4178 46. If there was a timeout during a "random" callout check, Exim treated it as
4179 a failure of the random address, and carried on sending RSET and the real
4180 address. If the delay was just some slowness somewhere, the response to the
4181 original RCPT would be taken as a response to RSET and so on, causing
4182 mayhem of various kinds.
4184 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4185 when I implemented it. It didn't allow for the fact that some option values
4186 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4187 the right test for positive values.
4189 48. Domain names in DNS records are case-independent. Exim always looks them up
4190 in lower case. Some resolvers return domain names in exactly the case they
4191 appear in the zone file, that is, they may contain uppercase letters. Not
4192 all resolvers do this - some return always lower case. Exim was treating a
4193 change of case by a resolver as a change of domain, similar to a widening
4194 of a domain abbreviation. This triggered its re-routing code and so it was
4195 trying to route what was effectively the same domain again. This normally
4196 caused routing to fail (because the router wouldn't handle the domain
4197 twice). Now Exim checks for this case specially, and just changes the
4198 casing of the domain that it ultimately uses when it transmits the message
4201 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4204 50. If a filter generated a file delivery with a non-absolute name (possible if
4205 no home directory exists for the router), the forbid_file option was not
4208 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4209 the existing equality matching.
4211 52. Exim was using ints instead of ino_t variables in some places where it was
4212 dealing with inode numbers.
4214 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4215 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4216 called TMPDIR, and if it finds it is different, it changes its value.
4218 54. The smtp_printf() function is now made available to local_scan() so
4219 additional output lines can be written before returning. There is also an
4220 smtp_fflush() function to enable the detection of a dropped connection.
4221 The variables smtp_input and smtp_batched_input are exported to
4224 55. Changed the default runtime configuration: the message "Unknown user"
4225 has been removed from the ACL, and instead placed on the localuser router,
4226 using the cannot_route_message feature. This means that any verification
4227 failures that generate their own messages won't get overridden. Similarly,
4228 the "Unrouteable address" message that was in the ACL for unverifiable
4229 relay addresses has also been removed.
4231 56. Added hosts_avoid_esmtp to the smtp transport.
4233 57. The exicyclog script was not checking for the esoteric option
4234 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4235 will work only if exicyclog is run under the appropriate euid.
4237 58. Following a discussion on the list, the rules by which Exim recognises line
4238 endings on incoming messages have been changed. The -dropcr and drop_cr
4239 options are now no-ops, retained only for backwards compatibility. The
4240 following line terminators are recognized: LF CRLF CR. However, special
4241 processing applies to CR:
4243 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4244 nor a local message in the state where . is a terminator.
4246 (ii) If a bare CR is encountered in a header line, an extra space is added
4247 after the line terminator so as not to end the header. The reasoning
4248 behind this is that bare CRs in header lines are most likely either
4249 to be mistakes, or people trying to play silly games.
4251 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4252 was being incorrectly given as 18 bytes larger than it should have been.
4253 This is a VOB (very old bug).
4255 60. This may never have affected anything current, but just in case it has:
4256 When the local host is found other than at the start of a list of hosts,
4257 the local host, those with the same MX, and any that follow, are discarded.
4258 When the list in question was part of a longer list of hosts, the following
4259 hosts (not currently being processed) were also being discarded. This no
4260 longer happens. I'm not sure if this situation could ever has previously
4263 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4266 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4267 header. This is something that is recommended in a new Internet Draft, and
4268 is something that is documented as being done by Sendmail. There are two
4269 possible values. For messages generated by the autoreply transport, Exim
4272 Auto-Submitted: auto-replied
4274 whereas for all other generated messages (e.g. bounces) it adds
4276 Auto-Submitted: auto-generated
4278 63. The "personal" condition in filters now includes a test for the
4279 Auto-Submitted: header. If it contains the string "auto-" the message it
4280 not considered personal.
4282 64. Added rcpt_include_affixes as a generic transport option.
4284 65. Added queue_only_override (default true).
4286 66. Added the syslog_duplication option.
4288 67. If what should have been the first header line of a message consisted of
4289 a space followed by a colon, Exim was mis-interpreting it as a header line.
4290 It isn't of course - it is syntactically invalid and should therefore be
4291 treated as the start of the message body. The misbehaviour could have
4292 caused a number of strange effects, including loss of data in subsequent
4293 header lines, and spool format errors.
4295 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4296 client host had authenticated. This control can now be exercised by an ACL
4297 for more flexibility.
4299 69. By default, callouts do not happen when testing with -bh. There is now a
4300 variant, -bhc, which does actually run the callout code, including
4301 consulting and updating the callout cache.
4303 70. Added support for saslauthd authentication, courtesy of Alexander
4306 71. If statvfs() failed on the spool or log directories while checking their
4307 size for availability, Exim confusingly gave the error "space shortage".
4308 Furthermore, in debugging mode it crashed with a floating point exception.
4309 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4310 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4311 serious problem, Exim now writes to the main and panic logs when this
4312 happens, with details of the failure. It then refuses to accept the
4313 incoming message, giving the message "spool directory problem" or "log
4314 directory problem" with a 421 code for SMTP messages.
4316 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4317 0, 1, and 2 exist, because some OS complain for execs without them (see
4318 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4319 descriptors. However, the code omitted to check that the open succeeded,
4320 causing mysterious errors if for some reason the permissions on /dev/null
4321 got screwed. Now Exim writes a message to the main and panic logs, and
4322 bombs out if it can't open /dev/null.
4324 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4325 interact so that it is all more flexible. It is supposed to remain
4326 backwards compatible. Also added extra_local_interfaces.
4328 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4329 to bomb out with an assertion failure - to the client this appears as a
4330 connection drop. This problem occurs in the part of the code that was taken
4331 from the Samba project. Fortunately, the assertion is in a very simple
4332 function, so I have fixed this by reproducing the function inline in the
4333 one place where it is called, and arranging for authentication to fail
4334 instead of killing the process with assert().
4336 75. The SPA client code was not working when the server requested OEM rather
4337 than Unicode encoding.
4339 76. Added code to make require_files with a specific uid setting more usable in
4340 the case where statting the file as root fails - usually a non-root-mounted
4341 NFS file system. When this happens and the failure is EACCES, Exim now
4342 forks a subprocess and does the per-uid checking as the relevant uid.
4344 77. Added process_log_path.
4346 78. If log_file_path was not explicitly set, a setting of check_log_space or
4347 check_log_inodes was ignored.
4349 79. If a space check for the spool or log partitions fails, the incident is now
4350 logged. Of course, in the latter case the data may get lost...
4352 80. Added the %p formatting code to string_format() so that it can be used to
4353 print addresses in debug_print(). Adjusted all the address printing in the
4354 debugging in store.c to use %p rather than %d.
4356 81. There was a concern that a line of code in smtp_in.c could overflow a
4357 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4358 initially expressed, the concern was not well-founded, because trailing
4359 spaces are removed early. However, if the trailing spaces were followed by
4360 a NULL, they did not get removed, so the overflow was possible. Two fixes
4363 (a) I re-wrote the offending code in a cleaner fashion.
4364 (b) If an incoming SMTP command contains a NULL character, it is rejected
4367 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4368 runs initgroups(), so that if the Exim user is in any additional groups,
4369 they will be used during message reception.
4375 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob