1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.417 2006/10/30 16:41:04 ph10 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 TF/01 In the add_headers option to the mail command in an Exim filter,
10 there was a bug that Exim would claim a syntax error in any
11 header after the first one which had an odd number of characters
14 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
15 callout verification, Exim cached a "reject" for the entire domain. This
16 is correct for most verifications, but it is not correct for a recipient
17 verification with use_sender or use_postmaster set, because in that case
18 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
19 case of MAIL FROM:<> rejection from other early rejections (e.g.
20 rejection of HELO). When verifying a recipient using a non-null MAIL
21 address, the cache is ignored if it shows MAIL FROM:<> rejection.
22 Whatever the result of the callout, the value of the domain cache is
23 left unchanged (for any other kind of callout, getting as far as trying
24 RCPT means that the domain itself is ok).
26 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
29 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
30 manifest itself as EPIPE rather than ECONNECT. When tidying away a
31 session, the daemon ignores ECONNECT errors and logs others; it now
32 ignores EPIPE as well.
34 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
35 (quoted-printable decoding).
37 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
38 later the small subsequent patch to fix an introduced bug.
40 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
42 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
44 PH/08 An error is now given if message_size_limit is specified negative.
46 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
47 to be given (somewhat) arbitrary names.
49 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
52 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
53 miscellaneous code fixes
55 PH/10 Added the log_reject_target ACL modifier to specify where to log
58 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
59 hostname. This is wrong, because it relates to the incoming message (and
60 probably the interface on which it is arriving) and not to the outgoing
61 callout (which could be using a different interface). This has been
62 changed to use the value of the helo_data option from the smtp transport
63 instead - this is what is used when a message is actually being sent. If
64 there is no remote transport (possible with a router that sets up host
65 addresses), $smtp_active_hostname is used.
67 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
68 tweaks were necessary in order to get it to work (see also 21 below):
69 (a) The code assumed that strncpy() returns a negative number on buffer
70 overflow, which isn't the case. Replaced with Exim's string_format()
72 (b) There were several signed/unsigned issues. I just did the minimum
73 hacking in of casts. There is scope for a larger refactoring.
74 (c) The code used strcasecmp() which is not a standard C function.
75 Replaced with Exim's strcmpic() function.
76 (d) The code set only $1; it now sets $auth1 as well.
77 (e) A simple test gave the error "authentication client didn't specify
78 service in request". It would seem that Dovecot has changed its
79 interface. Fortunately there's a specification; I followed it and
80 changed what the client sends and it appears to be working now.
82 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
85 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
86 address A is aliased to B and C, where B exists and C does not. Without
87 -v the output is "A verified" because verification stops after a
88 successful redirection if more than one address is generated. However,
89 with -v the child addresses are also verified. Exim was outputting "A
90 failed to verify" and then showing the successful verification for C,
91 with its parentage. It now outputs "B failed to verify", showing B's
92 parentage before showing the successful verification of C.
94 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
95 look up a TXT record in a specific list after matching in a combined
98 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
99 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
100 they consult the DNS. I had assumed they would set it the way they
101 wanted; and indeed my experiments on Linux seem to show that in some
102 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
103 To be on the safe side, however, I have now made the interface to
104 host_find_byname() similar to host_find_bydns(), with an argument
105 containing the DNS resolver options. The host_find_byname() function now
106 sets these options at its start, just as host_find_bydns() does. The smtp
107 transport options dns_qualify_single and dns_search_parents are passed to
108 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
109 of host_find_byname() use the default settings of RES_DEFNAMES
110 (qualify_single) but not RES_DNSRCH (search_parents).
112 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
113 spool_read_header() do less string testing, by means of a preliminary
114 switch on the second character of optional "-foo" lines. (This is
115 overdue, caused by the large number of possibilities that now exist.
116 Originally there were few.) While I was there, I also converted the
117 str(n)cmp tests so they don't re-test the leading "-" and the first
118 character, in the hope this might squeeze out yet more improvement.
120 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
121 flag allowing group syntax was set by the header_syntax check but not
122 turned off, possible causing trouble later; (2) The flag was not being
123 set at all for the header_verify test, causing "group"-style headers to
124 be rejected. I have now set it in this case, and also caused header_
125 verify to ignore an empty address taken from a group. While doing this, I
126 came across some other cases where the code for allowing group syntax
127 while scanning a header line wasn't quite right (mostly, not resetting
128 the flag correctly in the right place). These bugs could have caused
129 trouble for malformed header lines. I hope it is now all correct.
131 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
132 with the "reply" argument non-NULL. The code, however (which originally
133 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
134 but it didn't always do it. This confused somebody who was copying the
135 code for some other use. I have removed all the tests.
137 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
138 feature that was used to support insecure browsers during the U.S. crypto
139 embargo. It requires special client support, and Exim is probably the
140 only MTA that supported it -- and would never use it because real RSA is
141 always available. This code has been removed, because it had the bad
142 effect of slowing Exim down by computing (never used) parameters for the
143 RSA_EXPORT functionality.
145 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
146 authenticator to fail if there's a tab character in the incoming data
147 (there should never be unless someone is messing about, as it's supposed
148 to be base64-encoded). Also added, on Timo's advice, the "secured" option
149 if the connection is using TLS or if the remote IP is the same as the
150 local IP, and the "valid-client-cert option" if a client certificate has
153 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
154 authenticators. This can be used for authorization after authentication
155 succeeds. (In the case of plaintext, it servers for both authentication
158 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
159 if any retry times were supplied.
161 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
162 connection, where there is no client IP address to check. In this
163 situation, the verify now always succeeds.
165 PH/25 Applied John Jetmore's -Mset patch.
167 PH/26 Added -bem to be like -Mset, but loading a message from a file.
169 PH/27 In a string expansion for a processed (not raw) header when multiple
170 headers of the same name were present, leading whitespace was being
171 removed from all of them, but trailing whitespace was being removed only
172 from the last one. Now trailing whitespace is removed from each header
173 before concatenation. Completely empty headers in a concatenation (as
176 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
177 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
179 PH/29 After an address error (typically a 4xx response from a server), Exim
180 always tries the failing address if it appears in a new message, but
181 respects the retry time otherwise. This was implemented by checking for
182 being in a queue run, which isn't quite right. Now it checks the
183 "first_delivery" flag instead.
185 PH/30 Exim was sometimes attempting to deliver messages that had suffered
186 address errors (4xx response to RCPT) over the same connection as other
187 messages routed to the same hosts. Such deliveries are always "forced",
188 so retry times are not inspected. This resulted in far too many retries
189 for the affected addresses. The effect occurred only when there were more
190 hosts than the hosts_max_try setting in the smtp transport when it had
191 the 4xx errors. Those hosts that it had tried were not added to the list
192 of hosts for which the message was waiting, so if all were tried, there
193 was no problem. Two fixes have been applied:
195 (i) If there are any address or message errors in an SMTP delivery, none
196 of the hosts (tried or untried) are now added to the list of hosts
197 for which the message is waiting, so the message should not be a
198 candidate for sending over the same connection that was used for a
199 successful delivery of some other message. This seems entirely
200 reasonable: after all the message is NOT "waiting for some host".
201 This is so "obvious" that I'm not sure why it wasn't done
202 previously. Hope I haven't missed anything, but it can't do any
203 harm, as the worst effect is to miss an optimization.
205 (ii) If, despite (i), such a delivery is accidentally attempted, the
206 routing retry time is respected, so at least it doesn't keep
207 hammering the server.
213 SC/01 Use a glob alias rather than an array ref in eximstats generated
214 parser. This improves both readability and performance.
216 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
217 Don't display local sender or destination tables in eximstats unless
218 there is data to show.
219 Added average volumes into the eximstats top table text output.
221 SC/03 Collect data on the number of addresses (recipients) as well
222 as the number of messages in eximstats.
224 TF/01 Correct an error in the documentation for the redirect router. Exim
225 does (usually) call initgroups() when daemonizing.
227 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
228 with consistent privilege compared to when running as a daemon.
230 TF/03 Note in the spec that $authenticated_id is not set for local
231 submissions from trusted users.
233 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
234 Thanks to Dean Brooks <dean@iglou.com> for the patch.
236 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
237 by adding some example configuration directives to the default
238 configuration file. A little bit of work is required to uncomment the
239 directives and define how usernames and passwords are checked, but
240 there is now a framework to start from.
242 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
243 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
244 without this. I don't know how relevant this is to other LDAP libraries.
246 PH/02 Add the verb name to the "unknown ACL verb" error.
248 PH/03 Magnus Holmgren's patch for filter_prepend_home.
250 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
252 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
253 directory not expanded when it should be if an expanded home directory
254 was set for the address (which is overridden by the transport).
256 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
259 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
260 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
261 because it is too late at that time, and has no effect.
263 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
264 security issue with \' (bugzilla #107). I could not use the
265 PQescapeStringConn() function, because it needs a PGconn value as one of
268 PH/08 When testing addresses using -bt, indicate those final addresses that
269 are duplicates that would not cause an additional delivery. At least one
270 person was confused, thinking that -bt output corresponded to deliveries.
271 (Suppressing duplicates isn't a good idea as you lose the information
272 about possibly different redirections that led to the duplicates.)
274 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
275 systems where poll() doesn't work, in particular OS X.
277 PH/10 Added more information to debugging output for retry time not reached.
279 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
280 operations in malware.c.
282 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
285 PH/13 If write_rejectlog was set false when logging was sent to syslog with
286 syslog_duplication set false, log lines that would normally be written
287 both the the main log and to the reject log were not written to syslog at
290 PH/14 In the default configuration, change the use of "message" in ACL warn
291 statements to "add_header".
293 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
294 not followed by a command (e.g. "seen endif").
296 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
297 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
300 PH/17 Added extra conditions to the default value of delay_warning_condition
304 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
305 { match{$h_precedence:}{(?i)bulk|list|junk} } \
306 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
309 The Auto-Submitted: and various List- headers are standardised, whereas I
310 don't think Precedence: ever was.
312 PH/18 Refactored debugging code in route_finduser() to show more information,
313 in particular, the error code if getpwnam() issues one.
315 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
316 This is apparently needed in addition to the PH/07 change above to avoid
317 any possible encoding problems.
319 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
320 but not after initializing Perl.
322 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
323 output them only if debugging. By default they are written stderr,
324 apparently, which is not desirable.
326 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
329 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
332 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
334 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
335 authenticated or an ident call has been made. Suppress the default
336 values for $authenticated_id and $authenticated_sender (but permit -oMai
337 and -oMas) when testing with -bh.
339 PH/24 Re-jigged the order of the tests in the default configuration so that the
340 tests for valid domains and recipients precede the DNS black list and CSA
341 tests, on the grounds that those ones are more expensive.
343 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
344 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
345 This bug exists in every version of Exim that I still have, right back to
348 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
349 However, an attempt to turn on case-sensitivity in a regex key by
350 including (?-i) didn't work because the subject string was already
351 lowercased, and the effects were non-intuitive. It turns out that a
352 one-line patch can be used to allow (?-i) to work as expected.
358 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
359 other effects) broke the use of negated acl sub-conditions.
361 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
364 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
365 "Deny" causes Exim to reject the incoming connection with a 554 error.
366 Unfortunately, if there is a major crisis, such as a disk failure,
367 tcp-wrappers gives "deny", whereas what one would like would be some
368 kind of temporary error. A kludge has been added to help with this.
369 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
370 554 error is used if errno is still zero or contains ENOENT (which occurs
371 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
374 PH/03 Add -lutil to the default FreeBSD LIBS setting.
376 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
377 errors. Otherwise a message that provokes a temporary error (when other
378 messages do not) can cause a whole host to time out.
380 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
381 addresses were routed directly to files or pipes from a redirect router.
382 File deliveries just didn't batch; pipe deliveries might have suffered
385 PH/06 A failure to get a lock for a hints database would erroneously always say
386 "Failed to get write lock", even when it was really a read lock.
388 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
389 of 0600. This has been changed to use the value of the lockfile_mode
390 option (which defaults to 0600).
392 PH/08 Applied small patch from the Sieve maintainer.
394 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
395 folder from quota calculations, a direct delivery into this folder messed
396 up the contents of the maildirsize file. This was because the regex was
397 used only to exclude .Trash (or whatever) when the size of the mailbox
398 was calculated. There was no check that a delivery was happening into an
399 excluded directory. This bug has been fixed by ignoring all quota
400 processing for deliveries into excluded directories.
402 PH/10 Added the maildirfolder_create_regex option to appendfile.
408 PH/01 The code for finding all the local interface addresses on a FreeBSD
409 system running IPv6 was broken. This may well have applied to all BSD
410 systems, as well as to others that have similar system calls. The broken
411 code found IPv4 interfaces correctly, but gave incorrect values for the
412 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
413 that it would not match correctly against @[] and not recognize the IPv6
416 PH/02 The ipliteral router was not recognizing addresses of the form user@
417 [ipv6:....] because it didn't know about the "ipv6:" prefix.
419 PH/03 Added disable_ipv6.
421 PH/04 Changed $reply_address to use the raw form of the headers instead of the
422 decoded form, because it is most often used to construct To: headers
423 lines in autoreplies, and the decoded form may well be syntactically
424 invalid. However, $reply_address has leading white space removed, and all
425 newlines turned into spaces so that the autoreply transport does not
428 PH/05 If group was specified without a user on a router, and no group or user
429 was specified on a transport, the group from the router was ignored.
431 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
432 for visible compile-time settings that can be used to change these
433 numbers, for those that want even more. Backwards compatibility with old
434 spool files has been maintained. However, going back to a previous Exim
435 release will lost any variables that are in spool files.
437 PH/07 Two small changes when running in the test harness: increase delay when
438 passing a TCP/IP connection to a new process, in case the original
439 process has to generate a bounce, and remove special handling of
440 127.0.0.2 (sic), which is no longer necessary.
442 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
443 be the same on different OS.
445 PH/09 Moved a debug statement in filter processing to avoid a race problem when
448 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
449 whether --show-vars was specified or not
451 JJ/02 exipick: Added support for new ACL variable spool format introduced
454 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
455 syntactically invalid From: or Reply-to: line, and a filter used this to
456 generate an autoreply, and therefore failed to obtain an address for the
457 autoreply, Exim could try to deliver to a non-existent relative file
458 name, causing unrelated and misleading errors. What now happens is that
459 it logs this as a hard delivery error, but does not attempt to create a
462 PH/11 The exinext utility has a -C option for testing purposes, but although
463 the given file was scanned by exinext itself; it wasn't being passed on
464 when Exim was called.
466 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
467 an end-of-file indication when reading a command response.
469 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
470 compiled. In many other places in Exim, IPv6 addresses are always
471 recognized, so I have changed this. It also means that IPv4 domain
472 literals of the form [IPV4:n.n.n.n] are now always recognized.
474 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
475 used if the router is not running as root, for example, when verifying at
476 ACL time, or when using -bh. The debugging output from this situation was
477 non-existent - all you got was a failure to exec. I have made two
480 (a) Failures to set uid/gid, the current directory, or a process leader
481 in a subprocess such as that created by queryprogram now generate
482 suitable debugging ouput when -d is set.
484 (b) The queryprogram router detects when it is not running as root,
485 outputs suitable debugging information if -d is set, and then runs
486 the subprocess without attempting to change uid/gid.
488 PH/15 Minor change to Makefile for building test_host (undocumented testing
491 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
492 additional section of a DNS packet that returns MX or SRV records.
493 Instead, it always explicitly searches for A/AAAA records. This avoids
494 major problems that occur when a DNS server includes only records of one
495 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
496 fixed another bug: if SRV records were looked up and the corresponding
497 address records were *not* found in the additional section, the port
498 values from the SRV records were lost.
500 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
501 using the correct key (the original address) when searching the retry
502 rules in order to find which one to use for generating the retry hint.
504 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
505 adding the default one. Similarly, if it contains a Reply-To: header, the
506 errors_reply_to option, if set, is not used.
508 PH/19 When calculating a retry time, Exim used to measure the "time since
509 failure" by looking at the "first failed" field in the retry record. Now
510 it does not use this if it is later than than the arrival time of the
511 message. Instead it uses the arrival time. This makes for better
512 behaviour in cases where some deliveries succeed, thus re-setting the
513 "first failed" field. An example is a quota failure for a huge message
514 when small messages continue to be delivered. Without this change, the
515 "time since failure" will always be short, possible causing more frequent
516 delivery attempts for the huge message than are intended.
517 [Note: This change was subsequently modified - see PH/04 for 4.62.]
519 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
520 $1, $2, $3) because the numerical variables can be reset during some
521 expansion items (e.g. "match"), thereby losing the authentication data.
523 PH/21 Make -bV show the size of off_t variables so that the test suite can
524 decide whether to run tests for quotas > 2G.
526 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
527 mailbox_size, and mailbox_filecount in the appendfile transport. If a
528 filecount value is greater than 2G or if a quota value is greater than 2G
529 on a system where the size of off_t is not greater than 4, a panic error
532 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
533 never match. The debug and -bh output now contains an explicit error
534 message indicating a malformed IPv4 address or mask.
536 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
537 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
540 PH/25 Do not write to syslog when running in the test harness. The only
541 occasion when this arises is a failure to open the main or panic logs
542 (for which there is an explicit test).
544 PH/26 Added the /no_tell option to "control=freeze".
546 PH/27 If a host name lookup failed very early in a connection, for example, if
547 the IP address matched host_lookup and the reverse lookup yielded a name
548 that did not have a forward lookup, an error message of the form "no IP
549 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
550 could be logged. Now it outputs the IP address instead of "NULL".
552 PH/28 An enabling patch from MH: add new function child_open_exim2() which
553 allows the sender and the authenticated sender to be set when
554 submitting a message from within Exim. Since child_open_exim() is
555 documented for local_scan(), the new function should be too.
557 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
558 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
559 results in an empty string is now treated as unset.
561 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
563 PH/31 Added sender_verify_fail logging option.
565 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
566 needed by RFC 822 but not by RFC 2822 was commented out. I have now
567 tidied the source and removed it altogether.
569 PH/33 When a queue run was abandoned because the load average was too high, a
570 log line was always written; now it is written only if the queue_run log
571 selector is set. In addition, the log line for abandonment now contains
572 information about the queue run such as the pid. This is always present
573 in "start" and "stop" lines but was omitted from the "abandon" line.
575 PH/34 Omit spaces between a header name and the colon in the error message that
576 is given when verify = headers_syntax fails (if there are lots of them,
577 the message gets confusing).
579 PH/35 Change the default for dns_check_names_pattern to allow slashes within
580 names, as there are now some PTR records that contain slashes. This check
581 is only to protect against broken name servers that fall over on strange
582 characters, so the fact that it applies to all lookups doesn't matter.
584 PH/36 Now that the new test suite is complete, we can remove some of the
585 special code in Exim that was needed for the old test suite. For example,
586 sorting DNS records because real resolvers return them in an arbitrary
587 order. The new test suite's fake resolver always returns records in the
590 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
591 bounces) except when queue_only is set, to avoid logging races between
592 the different processes.
594 PH/38 Panic-die if .include specifies a non-absolute path.
596 PH/39 A tweak to the "H" retry rule from its user.
598 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
599 a label. They prevented compilation on older perls.
601 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
602 a warning to be raised on newish perls.
604 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
605 on queue. Changes to match documented behaviour of showing count of
606 messages matching specified criteria.
608 PH/40 Changed the default ident timeout from 30s to 5s.
610 PH/41 Added support for the use of login_cap features, on those BSD systems
611 that have them, for controlling the resources used by pipe deliveries.
613 PH/42 The content-scanning code uses fopen() to create files in which to put
614 message data. Previously it was not paying any attention to the mode of
615 the files. Exim runs with umask(0) because the rest of the code creates
616 files with open(), and sets the required mode explicitly. Thus, these
617 files were ending up world-writeable. This was not a big issue, because,
618 being within the spool directory, they were not world-accessible. I have
619 created a function called modefopen, which takes an additional mode
620 argument. It sets umask(777), creates the file, chmods it to the required
621 mode, then resets the umask. All the relevant calls to fopen() in the
622 content scanning code have been changed to use this function.
624 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
625 to 24 hours. This avoids potential overflow problems when processing G
626 and H retry rules. I suspect nobody ever tinkers with this value.
628 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
630 PH/45 When the plaintext authenticator is running as a client, the server's
631 challenges are checked to ensure they are valid base64 strings. By
632 default, the authentication attempt is cancelled if an invalid string is
633 received. Setting client_ignore_invalid_base64 true ignores these errors.
634 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
635 they are received. Thus, the responses can be made to depend on the
636 challenges. If an invalid string is ignored, an empty string is placed in
639 PH/46 Messages that are created by the autoreply transport now contains a
640 References: header, in accordance with RFCs 2822 and 3834.
642 PH/47 Added authenticated_sender_force to the smtp transport.
644 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
646 PH/49 Installed latest patch from the Sieve maintainer.
648 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
649 was also set, the appendfile transport was unnecessarily scanning a
650 directory of message files (e.g. for maildir delivery) to find the count
651 of files (along with the size), even though it did not need this
652 information. It now does the scan only if it needs to find either the
653 size of the count of files.
655 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
657 PH/52 Two bugs concerned with error handling when the smtp transport is
660 (i) Exim was not creating retry information for temporary errors given
661 for individual recipients after the DATA command when the smtp transport
662 was used in LMTP mode. This meant that they could be retried too
663 frequently, and not timed out correctly.
665 (ii) Exim was setting the flag that allows error details to be returned
666 for LMTP errors on RCPT commands, but not for LMTP errors for individual
667 recipients that were returned after the DATA command.
669 PH/53 This is related to PH/52, but is more general: for any failing address,
670 when detailed error information was permitted to be returned to the
671 sender, but the error was temporary, then after the final timeout, only
672 "retry timeout exceeded" was returned. Now it returns the full error as
673 well as "retry timeout exceeded".
675 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
676 do this, and (what is worse) MTAs that accept it.
678 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
679 will now be deprecated.
681 PH/56 New os.c-cygwin from the Cygwin maintainer.
683 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
684 formats (previously only available in exim formats via -bpr, -bpru,
685 and -bpra. Now also available in native and exiqgrep formats)
687 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
688 with very large, slow to parse queues
690 JJ/08 exipick: added ! as generic prefix to negate any criteria format
692 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
694 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
695 responses to authentication challenges, though it was showing the
696 challenges; (ii) I've removed the CR characters from the debug output for
699 PH/58 Allow for the insertion of a newline as well as a space when a string
700 is turned into more than one encoded-word during RFC 2047 encoding. The
701 Sieve code now uses this.
703 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
704 data_4xx, lost_connection, tls_required.
706 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
707 message was being sent as an SMTP response.
709 PH/61 Add -l and -k options to exicyclog.
711 PH/62 When verifying, if an address was redirected to one new address, so that
712 verification continued, and the new address failed or deferred after
713 having set something in $address_data, the value of $address_data was not
714 passed back to the ACL. This was different to the case when no
715 redirection occurred. The value is now passed back in both cases.
717 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
718 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
719 use login_cap.h, so on its own it isn't the distinguishing feature. The
720 new name refers directly to the setclassresources() function.
722 PH/65 Added configuration files for NetBSD3.
724 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
726 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
727 is preferred over IPv4.
729 PH/68 The bounce_return_message and bounce_return_body options were not being
730 honoured for bounces generated during the reception of non-SMTP messages.
731 In particular, this applied to messages rejected by the ACL. This bug has
732 been fixed. However, if bounce_return_message is true and bounce_return_
733 body is false, the headers that are returned for a non-SMTP message
734 include only those that have been read before the error was detected.
735 (In the case of an ACL rejection, they have all been read.)
737 PH/69 The HTML version of the specification is now built in a directory called
738 spec_html instead of spec.html, because the latter looks like a path with
739 a MIME-type, and this confuses some software.
741 PH/70 Catch two compiler warnings in sieve.c.
743 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
744 function verify_get_ident() calls ip_connect() to connect a socket, but
745 if the "connect()" function timed out, ip_connect() used to close the
746 socket. However, verify_get_ident() also closes the socket later, and in
747 between Exim writes to the log, which may get opened at this point. When
748 the socket was closed in ip_connect(), the log could get the same file
749 descriptor number as the socket. This naturally causes chaos. The fix is
750 not to close the socket in ip_connect(); the socket should be closed by
751 the function that creates it. There was only one place in the code where
752 this was missing, in the iplookup router, which I don't think anybody now
753 uses, but I've fixed it anyway.
755 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
756 well as to direct DNS lookups. Otherwise the handling of names in host
757 lists is inconsistent and therefore confusing.
763 PH/01 Two changes to the default runtime configuration:
765 (1) Move the checks for relay_from_hosts and authenticated clients from
766 after to before the (commented out) DNS black list checks.
768 (2) Add control=submission to the relay_from_hosts and authenticated
769 clients checks, on the grounds that messages accepted by these
770 statements are most likely to be submissions.
772 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
774 (1) Generate an error if the third argument for the ${prvs expansion is
777 (2) Treat a missing third argument of ${prvscheck as if it were an empty
780 (3) Reset the variables that are obtained from the first argument of
781 ${prvscheck and used in the second argument before leaving the code,
782 because their memory is reclaimed, so using them afterwards may do
785 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
786 one (it's much easier than Tom thought :-).
788 (5) Because of (4), we can now allow for the use of $prvscheck_result
789 inside the third argument.
791 PH/03 For some reason, the default setting of PATH when running a command from
792 a pipe transport was just "/usr/bin". I have changed it to
795 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
796 anything to be listed in the output from -bV.
798 PH/05 When a filter generated an autoreply, the entire To: header line was
799 quoted in the delivery log line, like this:
801 => >A.N.Other <ano@some.domain> <original@ddress> ...
803 This has been changed so that it extracts the operative address. There
804 may be more than one such address. If so, they are comma-separated, like
807 => >ano@some.domain,ona@other.domain <original@ddress> ...
809 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
810 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
811 not being looked up in the rDNS to get a host name, Exim was showing the
812 IP address twice in Received: lines, even though the IP addresses were
813 identical. For example:
815 Received: from [1.2.3.4] (helo=[1.2.3.4])
817 However, if the real host name was known, it was omitting the HELO data
818 if it matched the actual IP address. This has been tidied up so that it
819 doesn't show the same IP address twice.
821 PH/07 When both +timestamp and +memory debugging was on, the value given by
822 $tod_xxx expansions could be wrong, because the tod_stamp() function was
823 called by the debug printing, thereby overwriting the timestamp buffer.
824 Debugging no longer uses the tod_stamp() function when +timestamp is set.
826 PH/08 When the original message was included in an autoreply transport, it
827 always said "this is a copy of the message, including all the headers",
828 even if body_only or headers_only was set. It now gives an appropriate
831 PH/09 Applied a patch from the Sieve maintainer which:
833 o fixes some comments
834 o adds the (disabled) notify extension core
835 o adds some debug output for the result of if/elsif tests
836 o points to the current vacation draft in the documentation
837 and documents the missing references header update
841 o fixes a bug in processing the envelope test (when testing
842 multiple envelope elements, the last element determinted the
845 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
846 Electronic Mail") by including:
848 Auto-submitted: auto-generated
850 in the messages that it generates (bounce messages and others, such as
851 warnings). In the case of bounce messages for non-SMTP mesages, there was
852 also a typo: it was using "Auto_submitted" (underscore instead of
853 hyphen). Since every message generated by Exim is necessarily in response
854 to another message, thes have all been changed to:
856 Auto-Submitted: auto-replied
858 in accordance with these statements in the RFC:
860 The auto-replied keyword:
862 - SHOULD be used on messages sent in direct response to another
863 message by an automatic process,
865 - MUST NOT be used on manually-generated messages,
867 - MAY be used on Delivery Status Notifications (DSNs) and Message
868 Disposition Notifications (MDNs),
870 - MUST NOT be used on messages generated by automatic or periodic
871 processes, except for messages which are automatic responses to
874 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
875 to the default Received: header definition.
877 PH/12 Added log selector acl_warn_skipped (default on).
879 PH/13 After a successful wildlsearch lookup, discard the values of numeric
880 variables because (a) they are in the wrong storage pool and (b) even if
881 they were copied, it wouldn't work properly because of the caching.
883 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
884 checking when decoding. Apparently there are clients that generate
885 overlong encoded strings. Why am I not surprised?
887 PH/15 If the first argument of "${if match_address" was not empty, but did not
888 contain an "@" character, Exim crashed. Now it writes a panic log message
889 and treats the condition as false.
891 PH/16 In autoreply, treat an empty string for "once" the same as unset.
893 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
894 extension "envelope-auth". The code is finished and in agreement with
895 other implementations, but there is no documentation so far and in fact,
896 nobody wrote the draft yet. This extension is currently #undef'ed, thus
897 not changing the active code.
899 Print executed "if" and "elsif" statements when debugging is used. This
900 helps a great deal to understand what a filter does.
902 Document more things not specified clearly in RFC3028. I had all this
903 sorted out, when out of a sudden new issues came to my mind. Oops."
905 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
908 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
909 canonical form (as documented). However, after a host name lookup from
910 the IP address, check_host() was doing a simple string comparison with
911 addresses acquired from the DNS when checking that the found name did
912 have the original IP as one of its addresses. Since any found IPv6
913 addresses are likely to be in abbreviated form, the comparison could
914 fail. Luckily, there already exists a function for doing the comparison
915 by converting both addresses to binary, so now that is used instead of
918 PH/20 There was another similar case to PH/19, when a complete host name was
919 given in a host list; looking up its IP address could give an abbreviated
920 form, whereas the current host's name might or might not be abbreviated.
921 The same fix has been applied.
927 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
928 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
931 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
932 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
934 PH/03 Typo: missing ".o" in src/pcre/Makefile.
936 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
937 header line, restrict the check to what is listed in RFCs 2369 and 2929.
938 Also, for "Auto-Submitted", treat anything other than "no" as
939 non-personal, in accordance with RFC 3834. (Previously it treated
940 anything starting "auto-" as non-personal.)
942 TF/01 The control=submission/name=... option had a problem with syntax
943 errors if the name included a slash character. The /name= option
944 now slurps the rest of the string, so it can include any characters
945 but it must come last in the list of options (after /sender_retain
948 PH/05 Some modifications to the interface to the fake nameserver for the new
956 TK/01 Added the "success_on_redirect" address verification option. See
957 NewStuff for rationale and an example.
959 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
961 PH/02 Patch to exigrep to allow it to work on syslog lines.
963 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
964 fread() to skip over the body file's header line, because in Cygwin the
965 header line is locked and is inaccessible.
967 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
968 co-exist for some time) to make it clear that it is the Exim ID that is
969 referenced, not the Message-ID: header line.
971 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
972 string_format() function, because snprintf() does not exist on all
975 PH/06 The use of forbid_filter_existstest now also locks out the use of the
976 ${stat: expansion item.
978 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
979 protocol synchronization error", to keep the pedants happy.
981 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
982 well as for IRIX systems, when gcc is being used. See the host.c source
985 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
987 PH/10 Named domain lists were not working if used in a queue_smtp_domains
990 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
991 transport and to the smtp transport in LMTP mode.
993 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
995 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
996 run a filter in a subprocess. This could lead to confusion in subsequent
997 lookups in the parent process. There should also be a search_tidyup() at
998 the end of the subprocess.
1000 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1001 only if the host matched helo_try_verify_hosts, which caused the
1002 verification to occur when the EHLO/HELO command was issued. The ACL just
1003 tested the remembered result. Now, if a previous verification attempt has
1004 not happened, "verify = helo" does it there and then.
1006 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1008 TK/03 Fix log output including CR from clamd.
1010 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1011 exist provoked a memory error which could cause a segfault.
1013 PH/15 Installed PCRE 6.2
1015 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1017 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1018 of the problem. Specifically, suggested +O2 rather than +O1 for the
1021 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1023 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1024 record, the port was not correctly logged when the outgoing_port log
1025 selector was set (it logged the transort's default port).
1027 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1028 fallback_hosts, and "hosts" in the smtp transport.
1030 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1031 host errors such as "Connection refused".
1033 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1034 authentication with radiusclient 0.4.9:
1036 - Error returned from rc_read_config was caught wrongly
1037 - Username/password not passed on to radius server due to wrong length.
1039 The presumption is that some radiusclient API changes for 4.51/PH/17
1040 were not taken care of correctly. The code is still untested by me (my
1041 Linux distribution still has 0.3.2 of radiusclient), but it was
1042 contributed by a Radius user.
1044 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1045 expanding the "port" option of the smtp transport.
1047 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1048 while reading a MIME header. Thanks to Tom Hughes for a patch.
1050 PH/24 Include config.h inside local_scan.h so that configuration settings are
1053 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1054 that in an ACL for RCPT (for example), you can examine exactly what was
1057 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1058 commands, but it was not correctly comparing the address with the actual
1059 client host address. Thus, it would show the EHLO address in Received:
1060 header lines when this was not necessary.
1062 PH/27 Added the % operator to ${eval:}.
1064 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1065 it should be ignoring failures (because with -C, for example, it has lost
1066 privilege). It wasn't ignoring creation failures other than "already
1069 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1072 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1073 because an input line was too long, either on its own, or by virtue of
1074 too many continuations, the temporary file was not being removed, and the
1075 return code was incorrect.
1077 PH/31 Missing "BOOL" in function definition in filtertest.c.
1079 PH/32 Applied Sieve patches from the maintainer.
1081 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1083 PH/33 Added "verify = not_blind".
1085 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1086 Local/Makefile (with some defaults set). These are used in built scripts
1087 such as exicyclog, but they have never been used in the exim_install
1088 script (though there are many overriding facilities there). I have
1089 arranged that the exim_install script now takes note of these two
1092 PH/35 Installed configuration files for Dragonfly.
1094 PH/36 When a locally submitted message by a trusted user did not contain a
1095 From: header, and the sender address was obtained from -f or from an SMTP
1096 MAIL command, and the trusted user did not use -F to supply a sender
1097 name, $originator_name was incorrectly used when constructing a From:
1098 header. Furthermore, $originator_name was used for submission mode
1099 messages from external hosts without From: headers in a similar way,
1100 which is clearly wrong.
1102 PH/37 Added control=suppress_local_fixups.
1104 PH/38 When log_selector = +received_sender was set, and the addition of the
1105 sender made the log line's construction buffer exactly full, or one byte
1106 less than full, an overflow happened when the terminating "\n" was
1109 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1110 when the result of a list match is failure because a DNS lookup failed.
1112 PH/40 RM_COMMAND is now used in the building process.
1114 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1115 the "build-* directories that it finds.
1117 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1118 address in a domain literal was a prefix of an interface address.
1120 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1121 when verifying a sender address, unless rewrite_headers is false.
1123 PH/44 Wrote a long comment about why errors_to addresses are verified as
1124 recipients, not senders.
1126 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1127 the ratelimit ACL was added.
1129 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1131 PH/46 Added extra information about PostgreSQL errors to the error string.
1133 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1134 avoiding the need to install special zones in a real server. This is
1135 backwards compatible; if it can't find the fake resolver, it drops back.
1136 Thus, both old and new test suites can be run.
1138 TF/02 Added util/ratelimit.pl
1140 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1141 clock is set back in time.
1143 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1144 Candler <B.Candler@pobox.com>.
1146 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1147 OK for addresses that are the result of redirections.
1149 PH/48 A number of further additions for the benefit of the new test suite,
1150 including a fake gethostbyname() that interfaces to the fake DNS resolver
1153 TF/06 The fix for widen_domains has also been applied to qualify_single and
1154 search_parents which are the other dnslookup options that can cause
1157 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1160 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1162 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1163 service" if it encountered an MX record with an empty target hostname.
1164 The message is now "an MX or SRV record indicated no SMTP service".
1166 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1167 if the DNS of the sending site is misconfigured. This is quite a
1168 common situation. This change restores the behaviour of treating a
1169 helo verification defer as a failure.
1171 PH/51 If self=fail was set on a router, the bounce message did not include the
1172 actual error message.
1178 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1180 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1181 command itself ended in error, the underlying message about the transport
1182 filter timeout was being overwritten with the pipe command error. Now the
1183 underlying error message should be appended to the second error message.
1185 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1187 PH/02 Reduce the amount of output that "make" produces by default. Full output
1188 can still be requested.
1190 PH/03 The warning log line about a condition test deferring for a "warn" verb
1191 was being output only once per connection, rather than after each
1192 occurrence (because it was using the same function as for successful
1193 "warn" verbs). This seems wrong, so I have changed it.
1195 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1196 it should not have, which might have caused a crash in the right
1197 circumstances, but probably never did.
1199 PH/04 Installed a modified version of Tony Finch's patch to make submission
1200 mode fix the return path as well as the Sender: header line, and to
1201 add a /name= option so that you can make the user's friendly name appear
1204 TF/03 Added the control = fakedefer ACL modifier.
1206 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1207 Mark Lowes for thorough testing.
1209 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1211 TK/03 Merged latest SRS patch from Miles Wilton.
1213 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1214 with the definition in sysexits.h (which is #included earlier).
1215 Fortunately, Exim does not actually use EX_OK. The code used to try to
1216 preserve the sysexits.h value, by assumimg that macro definitions were
1217 scanned for macro replacements. I have been disabused of this notion,
1218 so now the code just undefines EX_OK before #including unistd.h.
1220 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1221 in the smtp transport. When a block could not be written in a single
1222 write() function, the timeout was being re-applied to each part-write.
1223 This seems wrong - if the receiver was accepting one byte at a time it
1224 would take for ever. The timeout is now adjusted when this happens. It
1225 doesn't have to be particularly precise.
1227 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1228 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1230 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1231 without a domain if the check to <postmaster@domain> fails.
1233 SC/01 Eximstats: added -xls and the ability to specify output files
1234 (patch written by Frank Heydlauf).
1236 SC/02 Eximstats: use FileHandles for outputing results.
1238 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1240 SC/04 Eximstats: fixed display of large numbers with -nvr option
1242 SC/05 Eximstats: fixed merging of reports with empty tables.
1244 SC/06 Eximstats: added the -include_original_destination flag
1246 SC/07 Eximstats: removed tabs and trailing whitespace.
1248 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1250 TK/06 MBOX spool code: Add real "From " MBOX separator line
1251 so the .eml file is really in mbox format (even though
1252 most programs do not really care). Patch from Alex Miller.
1254 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1255 The latter is generated from $received_to and is only set if the
1256 message has one envelope recipient. SA can use these headers,
1257 obviously out-of-the-box. Patch from Alex Miller.
1259 PH/08 The ${def test on a variable was returning false if the variable's
1260 value was "0", contrary to what the specification has always said!
1261 The result should be true unless the variable is empty.
1263 PH/09 The syntax error of a character other than { following "${if
1264 def:variable_name" (after optional whitespace) was not being diagnosed.
1265 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1266 accidental colon was present, for example, could give incorrect results.
1268 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1269 result is used (not including appendfile, where other changes are about
1272 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1273 This involved changing a lot of size variables from int to off_t. It
1274 should work with maildirs and everything.
1276 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1277 spamd dying while we are connected to it.
1279 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1282 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1283 function for BDB 4.3.
1285 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1287 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1288 the sources for PCRE have been re-arranged and more files are now
1291 PH/15 The code I had for printing potentially long long variables in PH/11
1292 above was not the best (it lost precision). The length of off_t variables
1293 is now inspected at build time, and an appropriate printing format (%ld
1294 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1295 to be "long long int" or "long int". This is needed for the internal
1296 formatting function string_vformat().
1298 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1299 the configuration file to be ":syslog", then the script "guesses" where
1300 the logs files are, rather than using the compiled in default. In our
1301 case the guess is not the same as the compiled default, so the script
1302 suddenly stopped working when I started to use syslog. The patch checks
1303 to see if log_file_path is "". If so, it attempts to read it from exim
1304 with no configuration file to get the compiled in version, before it
1305 falls back to the previous guessing code."
1307 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1308 implementing BATV in an Exim configuration. See NewStuff for the gory
1311 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1312 Makefile that are specific to HP-UX.
1314 PH/18 If the "use_postmaster" option was set for a recipient callout together
1315 with the "random" option, the postmaster address was used as the MAIL
1316 FROM address for the random test, but not for the subsequent recipient
1317 test. It is now used for both.
1319 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1320 patch removes a few documentation additions to RFC 3028, because the
1321 latest draft now contains them. It adds the new en;ascii-case comparator
1322 and a new error check for 8bit text in MIME parts. Comparator and
1323 require names are now matched exactly. I enabled the subaddress
1324 extension, but it is not well tested yet (read: it works for me)."
1326 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1327 rework some of the code of TK/09 above to avoid the hardwired use of
1328 "%lld" and "long long". Replaced the call to snprintf() with a call to
1331 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1332 records point to non-existent hosts", "retry timeout exceeded", and
1333 "retry time not reached for any host after a long failure period".
1335 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1336 experimental DomainKeys support:
1338 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1339 (2) On an error such as an illegally used "control", the wrong name for
1340 the control was given.
1342 These problems did NOT occur unless DomainKeys support was compiled.
1344 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1346 PH/24 Added ${if match_ip condition.
1348 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1349 databases so that it will be absolutely obvious if a crash occurs in the
1350 DB library. This is a regular occurrence (often caused by mis-matched
1353 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1354 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1355 picked up on a user's system that detects such things. There doesn't seem
1356 to be a gcc warning option for this - only an attribute that has to be
1357 put on the function's prototype. It seems that in Fedora Core 4 they have
1358 set this on a number of new functions. No doubt there will be more in due
1361 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1362 specify a transport. However, if an address that was verified by such a
1363 router was the subject of a callout, Exim crashed because it tried to
1364 read the rcpt_include_affixes from the non-existent transport. Now it
1365 just assumes that the setting of that option is false. This bug was
1366 introduced by 4.51/PH/31.
1368 PH/28 Changed -d+all to exclude +memory, because that information is very
1369 rarely of interest, but it makes the output a lot bigger. People tend to
1370 do -d+all out of habit.
1372 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1373 code in os-type was giving problems when libc.so lives in lib64, like on
1376 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1377 aren't the modern standard, and it seems that some systems' include files
1378 don't always have them. Exim was already checking for some of the newer
1379 ones like T_AAAA, and defining it itself. I've added checks for all the
1380 record types that Exim uses.
1382 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1383 not automatically generating a new one, as it is supposed to. This
1384 prevented TLS from working. If the file did exist, but contained invalid
1385 data, a new version was generated, as expected. It was only the case of a
1386 non-existent file that was broken.
1388 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1389 with a change in libdomainkeys > 0.64.
1391 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1392 from DNS. If the selector record carries the flag, it now has
1393 precedence over the domain-wide flag.
1395 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1397 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1398 the use of a port name that isn't defined in /etc/services) occurred, the
1399 message was deferred as in a normal delivery, and thus remained on the
1400 spool, instead of being failed because of the mua_wrapper setting. This
1401 is now fixed, and I tidied up some of the mua_wrapper messages at the
1404 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1405 the messages in a hash of arrays rather than using individual hashes.
1406 This is a bit cleaner and results in dramatic memory savings, albeit
1407 at a slight CPU cost.
1409 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1410 as requested by Marc Sherman.
1412 SC/10 Eximstats: added histograms for user specified patterns as requested
1415 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1417 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1418 fopen() in the content-scanning modules that did not already have it.
1424 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1425 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1427 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1429 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1431 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1432 to test Sieve filters that use "vacation".
1434 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1435 that changes the way the GnuTLS parameters are stored in the cache file.
1436 The new format can be generated externally. For backward compatibility,
1437 if the data in the cache doesn't make sense, Exim assumes it has read an
1438 old-format file, and it generates new data and writes a new file. This
1439 means that you can't go back to an older release without removing the
1442 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1443 work if there are any delivery delays because "one_time" forces the
1444 parent to be marked "delivered", so its unseen clone is never tried
1445 again. For this reason, Exim now forbids the simultaneous setting of
1448 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1449 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1450 entry if you want to know the details. The fix, however, neglected to
1451 consider the case when local delivery batching is involved. The test for
1452 "previously delivered" was not happening when checking to see if an
1453 address could be batched with a previous (undelivered) one; under
1454 certain circumstances this could lead to multiple deliveries to the same
1457 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1458 in its include files, and this causes problems building Exim.
1460 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1461 header_syntax) but Exim was just ignoring anything given after a slash.
1462 In particular, this caused confusion with an attempt to use "verify =
1463 reverse_host_lookup/defer_ok". An error is now given when options are
1464 supplied for verify items that do not have them. (Maybe reverse_host_
1465 lookup should have a defer_ok option, but that's a different point.)
1467 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1468 defined by RFC 821) to 2048, because there were problems with some AUTH
1469 commands, and RFC 1869 says the size should be increased for extended
1470 SMTP commands that take arguments.
1472 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1475 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1476 "unknown" error; now it says that the functionality isn't in the binary.
1478 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1479 an address' error message when a string expansion fails (syntax or
1480 whatever). Otherwise the password may appear in the log. Following change
1481 PH/42 below, there is no longer a chance of it appearing in a bounce
1484 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1486 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1487 of its addresses was ever tried. (Bugzilla bug #2.)
1489 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1490 the result incorrectly in the debug output. (It correctly added a newline
1491 to what was transported.)
1493 TF/01 Added $received_time.
1495 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1496 commented out examples of how to interface to a virus scanner and to
1497 SpamAssassin. Also added commented examples of av_scanner and
1498 spamd_address settings.
1500 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1501 and controls are allowed in which ACLs. There were a couple of minor
1502 errors. Some of the entries in the conditions table (which is a table of
1503 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1504 negation of where the condition IS allowed.
1506 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1508 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1509 header file does not have a version number, so I've had to invent a new
1510 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1511 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1512 radiusclient), but it was contributed by a Radius user.
1514 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1515 files or directories, for OpenSSL.
1517 PH/19 When an Exim process that is running as root has to create an Exim log
1518 file, it does so in a subprocess that runs as exim:exim so as to get the
1519 ownership right at creation (otherwise, other Exim processes might see
1520 the file with the wrong ownership). There was no test for failure of this
1521 fork() call, which would lead to the process getting stuck as it waited
1522 for a non-existent subprocess. Forks do occasionally fail when resources
1523 run out. I reviewed all the other calls to fork(); they all seem to check
1526 PH/20 When checking for unexpected SMTP input at connect time (before writing
1527 the banner), Exim was not dealing correctly with a non-positive return
1528 from the read() function. If the client had disconnected by this time,
1529 the result was a log entry for a synchronization error with an empty
1530 string after "input=" when read() returned zero. If read() returned -1
1531 (an event I could not check), uninitialized data bytes were printed.
1532 There were reports of junk text (parts of files, etc) appearing after
1535 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1537 PH/22 Added support for macro redefinition, and (re)definition in between
1538 driver and ACL definitions.
1540 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1541 forgetting to use the resulting value; it was using the unexpanded value.
1543 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1544 hadn't been configured. The fix is from Juergen Kreileder, who
1545 understands it better than I do:
1547 "Here's what I see happening with three configured cyrus_sasl
1548 authenticators configured (plain, login, cram-md5):
1550 On startup auth_cyrus_sasl_init() gets called for each of these.
1551 This means three calls to sasl_listmech() without a specified mech_list.
1552 => SASL tests which mechs of all available mechs actually work
1553 => three warnings about OTP not working
1554 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1556 With the patch, sasl_listmech() also gets called three times. But now
1557 SASL's mech_list option is set to the server_mech specified in the the
1558 authenticator. Or in other words, the answer from sasl_listmech()
1559 gets limited to just the mech you're testing for (which is different
1561 => the return list contains just 'plain' or 'login', 'cram-md5' or
1562 nothing depending on the value of ob->server_mech.
1564 I've just tested the patch: Authentication still works fine,
1565 unavailable mechs specified in the exim configuration are still
1566 caught, and the auth.log warnings about OTP are gone."
1568 PH/25 When debugging is enabled, the contents of the command line are added
1569 to the debugging output, even when log_selector=+arguments is not
1572 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1573 answer is "GNU", and only if the return is "GNU/something" is the answer
1576 PH/27 $acl_verify_message is now set immediately after the failure of a
1577 verification in an ACL, and so is available in subsequent modifiers. In
1578 particular, the message can be preserved by coding like this:
1580 warn !verify = sender
1581 set acl_m0 = $acl_verify_message
1583 Previously, $acl_verify_message was set only while expanding "message"
1584 and "log_message" when a very denied access.
1586 PH/28 Modified OS/os.c-Linux with
1588 -#ifndef OS_LOAD_AVERAGE
1589 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1591 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1592 nomenclature these days.)
1594 PH/29 Installed patch from the Sieve maintainer that adds the options
1595 sieve_useraddress and sieve_subaddress to the redirect router.
1597 PH/30 In these circumstances:
1598 . Two addresses routed to the same list of hosts;
1599 . First host does not offer TLS;
1600 . First host accepts first address;
1601 . First host gives temporary error to second address;
1602 . Second host offers TLS and a TLS session is established;
1603 . Second host accepts second address.
1604 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1605 and peerdn, if requested) that were in fact used only for the second
1608 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1609 attention to any local part prefix or suffix that was matched by the
1610 router that accepted the address. It now behaves in the same way as it
1611 does for delivery: the affixes are removed from the local part unless
1612 rcpt_include_affixes is set on the transport.
1614 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1615 timeout during the DATA phase of an incoming message.
1617 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1618 applied a patch sanctioned by the Sieve maintainer.
1620 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1621 the uid or gid is negative. A case of a negative gid caused this to be
1622 noticed. The fix allows for either to be negative.
1624 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1625 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1628 PH/36 The change PH/12 above was broken. Fixed it.
1630 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1631 the grounds that routing the same address twice would always produce the
1632 same answer. This might have been true once, but it is certainly no
1633 longer true now. Routing a child address may depend on the previous
1634 routing that produced that child. Some complicated redirection strategies
1635 went wrong when messages had multiple recipients, and made Exim's
1636 behaviour dependent on the order in which the addresses were given.
1638 I have moved the duplicate checking until after the routing is complete.
1639 Exim scans the addresses that are assigned to local and remote
1640 transports, and removes any duplicates. This means that more work will be
1641 done, as duplicates will always all be routed, but duplicates are
1642 presumably rare, so I don't expect this is of any significance.
1644 For deliveries to pipes, files, and autoreplies, the duplicate checking
1645 still happens during the routing process, since they are not going to be
1648 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1649 It corrects a timeout issue with spamd. This is Ian's comment: "The
1650 background is that sometimes spamd either never reads data from a
1651 connection it has accepted, or it never writes response data. The exiscan
1652 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1653 blindly assumes that writes won't block so it may never time out."
1655 PH/39 Allow G after quota size as well as K and M.
1657 PH/40 The value set for $authenticated_id in an authenticator may not contain
1658 binary zeroes or newlines because the value is written to log lines and
1659 to spool files. There was no check on this. Now the value is run through
1660 the string_printing() function so that such characters are converted to
1661 printable escape sequences.
1663 PH/41 $message_linecount is a new variable that contains the total number of
1664 lines in the message. Compare $body_linecount, which is the count for the
1667 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1668 bounce and delay warning messages, except in certain special cases, which
1671 (a) An SMTP error message from a remote host;
1672 (b) A message specified in a :fail: redirection;
1673 (c) A message specified in a "fail" command in a system filter;
1674 (d) A message specified in a FAIL return from the queryprogram router;
1675 (e) A message specified by the cannot_route_message router option.
1677 In these cases only, Exim does include the error details in bounce and
1678 warning messages. There are also a few cases where bland messages such
1679 as "unrouteable address" or "local delivery error" are given.
1681 PH/43 $value is now also set for the "else" part of a ${run expansion.
1683 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1684 being worked on, but at least Exim now implements the latest version to
1687 PH/45 In a pipe transport, although a timeout while waiting for the pipe
1688 process to complete was treated as a delivery failure, a timeout while
1689 writing the message to the pipe was logged, but erroneously treated as a
1690 successful delivery. Such timeouts include transport filter timeouts. For
1691 consistency with the overall process timeout, these timeouts are now
1692 treated as errors, giving rise to delivery failures by default. However,
1693 there is now a new Boolean option for the pipe transport called
1694 timeout_defer, which, if set TRUE, converts the failures into defers for
1695 both kinds of timeout. A transport filter timeout is now identified in
1698 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1699 systems where "make" and "gmake" are different, calling "gmake" at top
1700 level broke things. I've arranged for the value of $(MAKE) to be passed
1701 from the Makefile to this script so that it can call the same version of
1705 A note about Exim versions 4.44 and 4.50
1706 ----------------------------------------
1708 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1709 changes of various kinds. As a consequence, a big documentation update was
1710 needed. This delayed the release for rather longer than seemed good, especially
1711 in the light of a couple of (minor) security issues. Therefore, the changes
1712 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1713 release. So 4.44 and 4.50 are in effect two different branches that both start
1716 I have left the 4.50 change log unchanged; it contains all the changes since
1717 4.43. The change log for 4.44 is below; many of its items are identical to
1718 those for 4.50. This seems to be the most sensible way to preserve the
1719 historical information.
1725 1. Minor wording change to the doc/README.SIEVE file.
1727 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1728 computation of the current number of files was incorrect.
1730 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1731 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1734 4. Give more explanation in the error message when the command for a transport
1735 filter fails to execute.
1737 5. There are several places where Exim runs a non-Exim command in a
1738 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1739 was being done only for the command run by the queryprogram router. It is
1740 now done for all such subprocesses. The other cases are: ${run, transport
1741 filters, and the commands run by the lmtp and pipe transports.
1743 6. Added CONFIGURE_GROUP build-time option.
1745 7. Some older OS have a limit of 256 on the maximum number of file
1746 descriptors. Exim was using setrlimit() to set 1000 as a large value
1747 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1748 systems. I've change it so that if it can't get 1000, it tries for 256.
1750 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1751 was an oversight, and furthermore, ever since the addition of extra
1752 controls (e.g. 4.43/32), the checks on when to allow different forms of
1753 "control" were broken. There should now be diagnostics for all cases when a
1754 control that does not make sense is encountered.
1756 9. Added the /retain_sender option to "control=submission".
1758 10. $recipients is now available in the predata ACL (oversight).
1760 11. Tidy the search cache before the fork to do a delivery from a message
1761 received from the command line. Otherwise the child will trigger a lookup
1762 failure and thereby defer the delivery if it tries to use (for example) a
1763 cached ldap connection that the parent has called unbind on.
1765 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1766 of $address_data from the recipient verification was clobbered by the
1767 sender verification.
1769 13. The value of address_data from a sender verification is now available in
1770 $sender_address_data in subsequent conditions in the ACL statement.
1772 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1774 15. Added a new option "connect=<time>" to callout options, to set a different
1777 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1778 was its contents. (It was OK if the option was not defined at all.)
1780 17. A "Completed" log line is now written for messages that are removed from
1781 the spool by the -Mrm option.
1783 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1784 information about exactly what failed.
1786 19. Added -dd to debug only the daemon process.
1788 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1789 handles timeouts, both on the server side and network timeouts. Renamed the
1790 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1792 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1793 It is now set to "smtps".
1795 22. $host_address is now set to the target address during the checking of
1796 ignore_target_hosts.
1798 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1799 being passed; this would have caused $sender_host_name to have been used if
1800 matching the list had actually called for a host name (not very likely,
1801 since this list is usually IP addresses). A host name is now passed as
1804 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1805 code that specifies a non-restarting handler (typically sigaction() in
1806 modern systems) in an attempt to fix a rare and obscure crash bug.
1808 25. Narrowed the window for a race in the daemon that could cause it to ignore
1809 SIGCHLD signals. This is not a major problem, because they are used only to
1810 wake it up if nothing else does.
1812 26. A malformed maildirsize file could cause Exim to calculate negative values
1813 for the mailbox size or file count. Odd effects could occur as a result.
1814 The maildirsize information is now recalculated if the size or filecount
1817 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1818 support for a long time. Removed HAVE_SYS_VFS_H.
1820 28. Installed the latest version of exipick from John Jetmore.
1822 29. In an address list, if the pattern was not a regular expression, an empty
1823 subject address (from a bounce message) matched only if the pattern was an
1824 empty string. Non-empty patterns were not even tested. This was the wrong
1825 because it is perfectly reasonable to use an empty address as part of a
1826 database query. An empty address is now tested by patterns that are
1827 lookups. However, all the other forms of pattern expect the subject to
1828 contain a local part and a domain, and therefore, for them, an empty
1829 address still always fails if the pattern is not itself empty.
1831 30. Exim went into a mad DNS loop when attempting to do a callout where the
1832 host was specified on an smtp transport, and looking it up yielded more
1833 than one IP address.
1835 31. Re-factored the code for checking spool and log partition space into a
1836 function that finds that data and another that does the check. The former
1837 is then used to implement four new variables: $spool_space, $log_space,
1838 $spool_inodes, and $log_inodes.
1840 32. The RFC2047 encoding function was originally intended for short strings
1841 such as real names; it was not keeping to the 75-character limit for
1842 encoded words that the RFC imposes. It now respects the limit, and
1843 generates multiple encoded words if necessary. To be on the safe side, I
1844 have increased the buffer size for the ${rfc2047: expansion operator from
1847 33. It is now permitted to omit both strings after an "if" condition; if the
1848 condition is true, the result is "true". As before, when the second string
1849 is omitted, a false condition yields an empty string. This makes it less
1850 cumbersome to write custom ACL and router conditions.
1852 34. Failure to deliver a bounce message always caused it to be frozen, even if
1853 there was an errors_to setting on the router. The errors_to setting is now
1856 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1857 canonical form (fully expanded) before being placed in
1858 $sender_host_address.
1860 36. The table in the code that translates DNS record types into text (T_A to
1861 "A" for instance) was missing entries for NS and CNAME. It is just possible
1862 that this could have caused confusion if both these types were looked up
1863 for the same domain, because the text type is used as part of Exim's
1864 per-process caching. But the chance of anyone hitting this buglet seems
1867 37. The dnsdb lookup has been extended in a number of ways.
1869 (1) There is a new type, "zns", which walks up the domain tree until it
1870 finds some nameserver records. It should be used with care.
1872 (2) There is a new type, "mxh", which is like "mx" except that it returns
1873 just the host names, not the priorities.
1875 (3) It is now possible to give a list of domains (or IP addresses) to be
1876 looked up. The behaviour when one of the lookups defers can be
1877 controlled by a keyword.
1879 (4) It is now possible to specify the separator character for use when
1880 multiple records are returned.
1882 38. The dnslists ACL condition has been extended: it is now possible to supply
1883 a list of IP addresses and/or domains to be looked up in a particular DNS
1886 39. Added log_selector=+queue_time_overall.
1888 40. When running the queue in the test harness, wait just a tad after forking a
1889 delivery process, to get repeatability of debugging output.
1891 41. Include certificate and key file names in error message when GnuTLS fails
1892 to set them up, because the GnuTLS error message doesn't include the name
1893 of the failing file when there is a problem reading it.
1895 42. Allow both -bf and -bF in the same test run.
1897 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1899 44. The "Exiscan patch" is now merged into the mainline Exim source.
1901 45. Sometimes the final signoff response after QUIT could fail to get
1902 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1903 < 0 before doing a fflush(). This bug looks as though it goes back to the
1904 introduction of TLS in release 3.20, but "sometimes" must have been rare
1905 because the tests only now provoked it.
1907 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1908 (this can affect the format of dates).
1910 47. exim_tidydb, when checking for the continued existence of a message for
1911 which it has found a message-specific retry record, was not finding
1912 messages that were in split spool directories. Consequently, it was
1913 deleting retry records that should have stayed in existence.
1915 48. Steve fixed some bugs in eximstats.
1917 49. The SPA authentication driver was not abandoning authentication and moving
1918 on to the next authenticator when an expansion was forced to fail,
1919 contradicting the general specification for all authenticators. Instead it
1920 was generating a temporary error. It now behaves as specified.
1922 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1923 (the order specifies the preference for clients). The order is now AES256,
1924 AES128, 3DES, ARCFOUR128.
1926 51. Small patch to Sieve code - explicitly set From: when generating an
1929 52. Exim crashed if a remote delivery caused a very long error message to be
1930 recorded - for instance if somebody sent an entire SpamAssassin report back
1931 as a large number of 550 error lines. This bug was coincidentally fixed by
1932 increasing the size of one of Exim's internal buffers (big_buffer) that
1933 happened as part of the Exiscan merge. However, to be on the safe side, I
1934 have made the code more robust (and fixed the comments that describe what
1937 53. Now that there can be additional text after "Completed" in log lines (if
1938 the queue_time_overall log selector is set), a one-byte patch to exigrep
1939 was needed to allow it to recognize "Completed" as not the last thing in
1942 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1943 patch that reportedly fixes this has been added. I am not expert enough to
1944 create a test for it. This is what the patch creator wrote:
1946 "I found a little strange behaviour of ldap code when working with
1947 Windows 2003 AD Domain, where users was placed in more than one
1948 Organization Units. When I tried to give exim partial DN, the exit code
1949 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1950 But simultaneously result of request was absolutely normal ldap result,
1951 so I produce this patch..."
1953 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1954 REFERENCE, so I have modified the code to exclude the patch when that macro
1957 55. Some experimental protocols are using DNS PTR records for new purposes. The
1958 keys for these records are domain names, not reversed IP addresses. The
1959 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1960 leaves it alone. Component reversal etc. now happens only for IP addresses.
1963 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1965 57. Double the size of the debug message buffer (to 2048) so that more of very
1966 long debug lines gets shown.
1968 58. The exicyclog utility now does better if the number of log files to keep
1969 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1971 59. Two changes related to the smtp_active_hostname option:
1973 (1) $smtp_active_hostname is now available as a variable.
1974 (2) The default for smtp_banner uses $smtp_active_hostname instead
1975 of $primary_hostname.
1977 60. The host_aton() function is supposed to be passed a string that is known
1978 to be a valid IP address. However, in the case of IPv6 addresses, it was
1979 not checking this. This is a hostage to fortune. Exim now panics and dies
1980 if the condition is not met. A case was found where this could be provoked
1981 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1982 components; fortuitously, this particular loophole had already been fixed
1983 by change 4.50/55 above.
1985 If there are any other similar loopholes, the new check in host_aton()
1986 itself should stop them being exploited. The report I received stated that
1987 data on the command line could provoke the exploit when Exim was running as
1988 exim, but did not say which command line option was involved. All I could
1989 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1990 running as the user.
1993 61. There was a buffer overflow vulnerability in the SPA authentication code
1994 (which came originally from the Samba project). I have added a test to the
1995 spa_base64_to_bits() function which I hope fixes it.
1998 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1999 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2001 63. The daemon start-up calls getloadavg() while still root for those OS that
2002 need the first call to be done as root, but it missed one case: when
2003 deliver_queue_load_max is set with deliver_drop_privilege. This is
2004 necessary for the benefit of the queue runner, because there is no re-exec
2005 when deliver_drop_privilege is set.
2007 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2008 This has been fixed.
2010 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2011 was in use, was not putting the data itself into the right store pool;
2012 consequently, it could be overwritten for a subsequent message in the same
2013 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2016 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2018 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2019 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2020 the function were treating the return as a boolean value, which happened to
2021 work because 0=false and not-0=true, but is not correct code.
2023 68. The host_aton() function was not handling scoped IPv6 addresses (those
2024 with, for example, "%eth0" on the end) correctly.
2026 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2027 negated items (that is, ~something) in unsigned ints. Some compilers
2028 apparently mutter when there is no cast.
2030 70. If an address verification called from an ACL failed, and did not produce a
2031 user-specific message (i.e. there was only a "system" message), nothing was
2032 put in $acl_verify_message. In this situation, it now puts the system
2035 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2036 session; change 4.31/43 added the unwanted input to the log line - except
2037 that it did not do this in the start of session case. It now does.
2039 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2040 This is wrong and can cause the other end to generate a synchronization
2041 error if it is another Exim or anything else that does the synchronization
2042 check. A QUIT command is no longer sent after a timeout.
2044 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2045 during host lookups.
2047 74. The defer_ok option of callout verification was not working if it was used
2048 when verifying addresses in header lines, that is, for this case:
2050 verify = header_sender/callout=defer_ok
2052 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2053 those file descriptors could be used for SMTP connections. If anything
2054 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2055 could be sent to the SMTP client, causing chaos. The daemon now opens
2056 stdin, stdout, and stderr to /dev/null when it puts itself into the
2059 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2060 log by default. The user can override this with suitable Perl magic.
2062 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2063 the log message when discard triggers, was not working for the DATA ACL or
2064 for the non-SMTP ACL.
2066 78. Error message wording change in sieve.c.
2068 79. If smtp_accept_max_per_host was set, the number of connections could be
2069 restricted to fewer than expected, because the daemon was trying to set up
2070 a new connection before checking whether the processes handling previous
2071 connections had finished. The check for completed processes is now done
2072 earlier. On busy systems, this bug wouldn't be noticed because something
2073 else would have woken the daemon, and it would have reaped the completed
2076 80. If a message was submitted locally by a user whose login name contained one
2077 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2078 It caused a spool format error. I have fixed the spool reading code. A
2079 related problem was that the "from" clause in the Received: line became
2080 illegal because of the space(s). It is now covered by ${quote_local_part.
2082 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2085 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2087 83. Patch from Sieve maintainer to fix unterminated string problem in
2088 "vacation" handling.
2090 84. Some minor changes to the Linux configuration files to help with other
2091 OS variants using glibc.
2093 85. One more patch for Sieve to update vacation handling to latest spec.
2096 ----------------------------------------------------
2097 See the note above about the 4.44 and 4.50 releases.
2098 ----------------------------------------------------
2104 1. Change 4.43/35 introduced a bug that caused file counts to be
2105 incorrectly computed when quota_filecount was set in an appendfile
2108 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2109 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2112 3. Give more explanation in the error message when the command for a transport
2113 filter fails to execute.
2115 4. There are several places where Exim runs a non-Exim command in a
2116 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2117 was being done only for the command run by the queryprogram router. It is
2118 now done for all such subprocesses. The other cases are: ${run, transport
2119 filters, and the commands run by the lmtp and pipe transports.
2121 5. Some older OS have a limit of 256 on the maximum number of file
2122 descriptors. Exim was using setrlimit() to set 1000 as a large value
2123 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2124 systems. I've change it so that if it can't get 1000, it tries for 256.
2126 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2127 was an oversight, and furthermore, ever since the addition of extra
2128 controls (e.g. 4.43/32), the checks on when to allow different forms of
2129 "control" were broken. There should now be diagnostics for all cases when a
2130 control that does not make sense is encountered.
2132 7. $recipients is now available in the predata ACL (oversight).
2134 8. Tidy the search cache before the fork to do a delivery from a message
2135 received from the command line. Otherwise the child will trigger a lookup
2136 failure and thereby defer the delivery if it tries to use (for example) a
2137 cached ldap connection that the parent has called unbind on.
2139 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2140 of $address_data from the recipient verification was clobbered by the
2141 sender verification.
2143 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2144 was its contents. (It was OK if the option was not defined at all.)
2146 11. A "Completed" log line is now written for messages that are removed from
2147 the spool by the -Mrm option.
2149 12. $host_address is now set to the target address during the checking of
2150 ignore_target_hosts.
2152 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2153 being passed; this would have caused $sender_host_name to have been used if
2154 matching the list had actually called for a host name (not very likely,
2155 since this list is usually IP addresses). A host name is now passed as
2158 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2159 code that specifies a non-restarting handler (typically sigaction() in
2160 modern systems) in an attempt to fix a rare and obscure crash bug.
2162 15. Narrowed the window for a race in the daemon that could cause it to ignore
2163 SIGCHLD signals. This is not a major problem, because they are used only to
2164 wake it up if nothing else does.
2166 16. A malformed maildirsize file could cause Exim to calculate negative values
2167 for the mailbox size or file count. Odd effects could occur as a result.
2168 The maildirsize information is now recalculated if the size or filecount
2171 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2172 support for a long time. Removed HAVE_SYS_VFS_H.
2174 18. Updated exipick to current release from John Jetmore.
2176 19. Allow an empty sender to be matched against a lookup in an address list.
2177 Previously the only cases considered were a regular expression, or an
2180 20. Exim went into a mad DNS lookup loop when doing a callout where the
2181 host was specified on the transport, if the DNS lookup yielded more than
2184 21. The RFC2047 encoding function was originally intended for short strings
2185 such as real names; it was not keeping to the 75-character limit for
2186 encoded words that the RFC imposes. It now respects the limit, and
2187 generates multiple encoded words if necessary. To be on the safe side, I
2188 have increased the buffer size for the ${rfc2047: expansion operator from
2191 22. Failure to deliver a bounce message always caused it to be frozen, even if
2192 there was an errors_to setting on the router. The errors_to setting is now
2195 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2196 canonical form (fully expanded) before being placed in
2197 $sender_host_address.
2199 24. Updated eximstats to version 1.33
2201 25. Include certificate and key file names in error message when GnuTLS fails
2202 to set them up, because the GnuTLS error message doesn't include the name
2203 of the failing file when there is a problem reading it.
2205 26. Expand error message when OpenSSL has problems setting up cert/key files.
2208 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2209 (this can affect the format of dates).
2211 28. exim_tidydb, when checking for the continued existence of a message for
2212 which it has found a message-specific retry record, was not finding
2213 messages that were in split spool directories. Consequently, it was
2214 deleting retry records that should have stayed in existence.
2216 29. eximstats updated to version 1.35
2217 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2218 1.35 - bugfix such that pie charts by volume are generated correctly
2220 30. The SPA authentication driver was not abandoning authentication and moving
2221 on to the next authenticator when an expansion was forced to fail,
2222 contradicting the general specification for all authenticators. Instead it
2223 was generating a temporary error. It now behaves as specified.
2225 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2226 (the order specifies the preference for clients). The order is now AES256,
2227 AES128, 3DES, ARCFOUR128.
2229 31. Small patch to Sieve code - explicitly set From: when generating an
2232 32. Exim crashed if a remote delivery caused a very long error message to be
2233 recorded - for instance if somebody sent an entire SpamAssassin report back
2234 as a large number of 550 error lines. This bug was coincidentally fixed by
2235 increasing the size of one of Exim's internal buffers (big_buffer) that
2236 happened as part of the Exiscan merge. However, to be on the safe side, I
2237 have made the code more robust (and fixed the comments that describe what
2240 33. Some experimental protocols are using DNS PTR records for new purposes. The
2241 keys for these records are domain names, not reversed IP addresses. The
2242 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2243 leaves it alone. Component reversal etc. now happens only for IP addresses.
2246 34. The host_aton() function is supposed to be passed a string that is known
2247 to be a valid IP address. However, in the case of IPv6 addresses, it was
2248 not checking this. This is a hostage to fortune. Exim now panics and dies
2249 if the condition is not met. A case was found where this could be provoked
2250 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2251 components; fortuitously, this particular loophole had already been fixed
2252 by change 4.50/55 or 4.44/33 above.
2254 If there are any other similar loopholes, the new check in host_aton()
2255 itself should stop them being exploited. The report I received stated that
2256 data on the command line could provoke the exploit when Exim was running as
2257 exim, but did not say which command line option was involved. All I could
2258 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2259 running as the user.
2262 35. There was a buffer overflow vulnerability in the SPA authentication code
2263 (which came originally from the Samba project). I have added a test to the
2264 spa_base64_to_bits() function which I hope fixes it.
2267 36. The daemon start-up calls getloadavg() while still root for those OS that
2268 need the first call to be done as root, but it missed one case: when
2269 deliver_queue_load_max is set with deliver_drop_privilege. This is
2270 necessary for the benefit of the queue runner, because there is no re-exec
2271 when deliver_drop_privilege is set.
2273 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2274 was in use, was not putting the data itself into the right store pool;
2275 consequently, it could be overwritten for a subsequent message in the same
2276 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2279 38. Sometimes the final signoff response after QUIT could fail to get
2280 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2281 < 0 before doing a fflush(). This bug looks as though it goes back to the
2282 introduction of TLS in release 3.20, but "sometimes" must have been rare
2283 because the tests only now provoked it.
2289 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2290 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2291 Now it returns an integer. A number of calls were still expecting a T/F
2292 return. Fortuitously, in all cases, the tests worked in OK situations,
2293 which is the norm. However, things would have gone wrong on any write
2294 failures on the smtp file descriptor. This function is used when sending
2295 messages over SMTP and also when doing verify callouts.
2297 2. When Exim is called to do synchronous delivery of a locally submitted
2298 message (the -odf or -odi options), it no longer closes stderr before doing
2301 3. Implemented the mua_wrapper option.
2303 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2305 5. Implemented the functions header_remove(), header_testname(),
2306 header_add_at_position(), and receive_remove_recipient(), and exported them
2309 6. If an ACL "warn" statement specified the addition of headers, Exim already
2310 inserted X-ACL-Warn: at the start if there was no header name. However, it
2311 was not making this test for the second and subsequent header lines if
2312 there were newlines in the string. This meant that an invalid header could
2313 be inserted if Exim was badly configured.
2315 7. Allow an ACL "warn" statement to add header lines at the start or after all
2316 the Received: headers, as well as at the end.
2318 8. Added the rcpt_4xx retry error code.
2320 9. Added postmaster_mailfrom=xxx to callout verification option.
2322 10. Added mailfrom=xxxx to the callout verification option, for verify=
2325 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2326 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2328 12. Inserted some casts to stop certain compilers warning when using pointer
2329 differences as field lengths or precisions in printf-type calls (mostly
2330 affecting debugging statements).
2332 13. Added optional readline() support for -be (dynamically loaded).
2334 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2335 same clock tick as a message's arrival, so that its received time was the
2336 same as the "first fail" time on the retry record, and that message
2337 remained on the queue past the ultimate address timeout, every queue runner
2338 would try a delivery (because it was past the ultimate address timeout) but
2339 after another failure, the ultimate address timeout, which should have then
2340 bounced the address, did not kick in. This was a "< instead of <=" error;
2341 in most cases the first failure would have been in the next clock tick
2342 after the received time, and all would be well.
2344 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2345 being recognized when the domain list was tested by the match_domain
2346 condition in an expansion string.
2348 16. Added the ${str2b64: operator.
2350 17. Exim was always calling setrlimit() to set a large limit for the number of
2351 processes, without checking whether the existing limit was already
2352 adequate. (It did check for the limit on file descriptors.) Furthermore,
2353 errors from getrlimit() and setrlimit() were being ignored. Now they are
2354 logged to the main and panic logs, but Exim does carry on, to try to do its
2355 job under whatever limits there are.
2357 18. Imported PCRE 5.0.
2359 19. Trivial typo in log message " temporarily refused connection" (the leading
2362 20. If the log selector return_path_on_delivery was set and an address was
2363 redirected to /dev/null, the delivery process crashed because it assumed
2364 that a return path would always be set for a "successful" delivery. In this
2365 case, the whole delivery is bypassed as an optimization, and therefore no
2368 21. Internal re-arrangement: the function for sending a challenge and reading
2369 a response while authentication was assuming a zero-terminated challenge
2370 string. It's now changed to take a pointer and a length, to allow for
2371 binary data in such strings.
2373 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2375 23. Exim was not respecting finduser_retries when seeking the login of the
2376 uid under which it was called; it was always trying 10 times. (The default
2377 setting of finduser_retries is zero.) Also, it was sleeping after the final
2378 failure, which is pointless.
2380 24. Implemented tls_on_connect_ports.
2382 25. Implemented acl_smtp_predata.
2384 26. If the domain in control=submission is set empty, Exim assumes that the
2385 authenticated id is a complete email address when it generates From: or
2386 Sender: header lines.
2388 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2389 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2390 chown and chgrp in /bin and hostname in /usr/bin.
2392 28. Exim was keeping the "process log" file open after each use, just as it
2393 does for the main log. This opens the possibility of it remaining open for
2394 long periods when the USR1 signal hits a daemon. Occasional processlog
2395 errors were reported, that could have been caused by this. Anyway, it seems
2396 much more sensible not to leave this file open at all, so that is what now
2399 29. The long-running daemon process does not normally write to the log once it
2400 has entered its main loop, and it closes the log before doing so. This is
2401 so that log files can straightforwardly be renamed and moved. However,
2402 there are a couple of unusual error situations where the daemon does write
2403 log entries, and I had neglected to close the log afterwards.
2405 30. The text of an SMTP error response that was received during a remote
2406 delivery was being truncated at 512 bytes. This is too short for some of
2407 the long messages that one sometimes sees. I've increased the limit to
2410 31. It is now possible to make retry rules that apply only when a message has a
2411 specific sender, in particular, an empty sender.
2413 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2414 it possible to be selective about when SMTP synchronization is enforced.
2416 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2418 32. Implemented hosts_connection_nolog.
2420 33. Added an ACL for QUIT.
2422 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2425 35. Added mailbox_size and mailbox_filecount to appendfile.
2427 36. Added control = no_multiline_responses to ACLs.
2429 37. There was a bug in the logic of the code that waits for the clock to tick
2430 in the case where the clock went backwards by a substantial amount such
2431 that the microsecond fraction of "now" was more than the microsecond
2432 fraction of "then" (but the whole seconds number was less).
2434 38. Added support for the libradius Radius client library this is found on
2435 FreeBSD (previously only the radiusclient library was supported).
2441 1. When certain lookups returned multiple values in the form name=value, the
2442 quoting of the values was not always being done properly. Specifically:
2443 (a) If the value started with a double quote, but contained no whitespace,
2445 (b) If the value contained whitespace other than a space character (i.e.
2446 tabs or newlines or carriage returns) it was not quoted.
2447 This fix has been applied to the mysql and pgsql lookups by writing a
2448 separate quoting function and calling it from the lookup code. The fix
2449 should probably also be applied to nisplus, ibase and oracle lookups, but
2450 since I cannot test any of those, I have not disturbed their existing code.
2452 2. A hit in the callout cache for a specific address caused a log line with no
2453 reason for rejecting RCPT. Now it says "Previous (cached) callout
2454 verification failure".
2456 3. There was an off-by-one bug in the queryprogram router. An over-long
2457 return line was truncated at 256 instead of 255 characters, thereby
2458 overflowing its buffer with the terminating zero. As well as fixing this, I
2459 have increased the buffer size to 1024 (and made a note to document this).
2461 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2462 when Exim is waiting for an SMTP response from a remote server, Exim
2463 restarts its select() call on the socket, thereby resetting its timeout.
2464 This is not a problem when such interrupts are rare. Somebody set up a cron
2465 job to run exiwhat every 2 minutes, which is less than the normal select()
2466 timeout (5 or 10 minutes). This meant that the select() timeout never
2467 kicked in because it was always reset. I have fixed this by comparing the
2468 time when an interrupt arrives with the time at the start of the first call
2469 to select(). If more time than the timeout has elapsed, the interrupt is
2470 treated as a timeout.
2472 5. Some internal re-factoring in preparation for the addition of Sieve
2473 extensions (by MH). In particular, the "personal" test is moved to a
2474 separate function, and given an option for scanning Cc: and Bcc: (which is
2475 not set for Exim filters).
2477 6. When Exim created an email address using the login of the caller as the
2478 local part (e.g. when creating a From: or Sender: header line), it was not
2479 quoting the local part when it contained special characters such as @.
2481 7. Installed new OpenBSD configuration files.
2483 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2484 try to make them clearer.
2486 9. Callout options, other than the timeout value, were being ignored when
2487 verifying sender addresses in header lines. For example, when using
2489 verify = header_sender/callout=no_cache
2491 the cache was (incorrectly) being used.
2493 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2494 only the Cygwin environment.
2496 11. When return_path_on_delivery was set as a log selector, if different remote
2497 addresses in the same message used different return paths and parallel
2498 remote delivery occurred, the wrong values would sometimes be logged.
2499 (Whenever a remote delivery process finished, the return path value from
2500 the most recently started remote delivery process was logged.)
2502 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2503 lines when AUTH and/or TLS are in use. This is the "received protocol"
2504 field. Exim used to use "asmtp" for authenticated SMTP, without any
2505 indication (in the protocol name) for TLS use. Now it follows the RFC and
2506 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2507 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2508 names appear in log lines as well as in Received: header lines.
2510 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2511 extensions, and comparison tests, and to fix some bugs.
2513 14. Changes to the "personal" filter test:
2515 (1) The test was buggy in that it was just doing the equivalent of
2516 "contains" tests on header lines. For example, if a user's address was
2517 anne@some.where, the "personal" test would incorrectly be true for
2519 To: susanne@some.where
2521 This test is now done by extracting each address from the header in turn,
2522 and checking the entire address. Other tests that are part of "personal"
2523 are now done using regular expressions (for example, to check local parts
2524 of addresses in From: header lines).
2526 (2) The list of non-personal local parts in From: addresses has been
2527 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2528 taken from the Sieve specification recommendations.
2530 (3) If the message contains any header line starting with "List-" it is
2531 treated as non-personal.
2533 (4) The test for "circular" in the Subject: header line has been removed
2534 because it now seems ill-conceived.
2536 15. Minor typos in src/EDITME comments corrected.
2538 16. Installed latest exipick from John Jetmore.
2540 17. If headers_add on a router specified a text string that was too long for
2541 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2542 of string_sprintf() is now avoided.
2544 18. $message_body_size was not set (it was always zero) when running the DATA
2545 ACL and the local_scan() function.
2547 19. For the "mail" command in an Exim filter, no default was being set for
2548 the once_repeat time, causing a random time value to be used if "once" was
2549 specified. (If the value happened to be <= 0, no repeat happened.) The
2550 default is now 0s, meaning "never repeat". The "vacation" command was OK
2551 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2552 (I found it when inspecting the code).
2554 20. There is now an overall timeout for performing a callout verification. It
2555 defaults to 4 times the callout timeout, which applies to individual SMTP
2556 commands during the callout. The overall timeout applies when there is more
2557 than one host that can be tried. The timeout is checked before trying the
2558 next host. This prevents very long delays if there are a large number of
2559 hosts and all are timing out (e.g. when the network connections are timing
2560 out). The value of the overall timeout can be changed by specifying an
2561 additional sub-option for "callout", called "maxwait". For example:
2563 verify = sender/callout=5s,maxwait=20s
2565 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2566 to the end before writing, but this should make it even safer).
2568 22. Exim was forgetting that it had advertised PIPELINING for the second and
2569 subsequent messages on an SMTP connection. It was also not resetting its
2570 memory on STARTTLS and an internal HELO.
2572 23. When Exim logs an SMTP synchronization error within a session, it now
2573 records whether PIPELINING has been advertised or not.
2575 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2576 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2577 rather than long int.
2579 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2581 26. Added the never_mail option to autoreply.
2587 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2588 crash if the getsockname() call failed; this can happen if a connection is
2589 closed very soon after it is established. The problem was simply in the
2590 order in which certain operations were done, causing Exim to try to write
2591 to the SMTP stream before it had set up the file descriptor. The bug has
2592 been fixed by making things happen in the correct order.
2598 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2599 before the connection was closed, thus losing the rejection response.
2601 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2602 some early Solaris releases, but causes trouble in current releases where
2603 socklen_t is defined.
2605 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2608 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2611 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2612 that releasing the top of it at the end releases what was used for sub-
2613 expansions (unless the block got too big). However, discard this block if
2614 the first thing is a variable or header, so that we can use its block when
2615 it is dynamic (useful for very large $message_headers, for example).
2617 6. Lookups now cache *every* query, not just the most recent. A new, separate
2618 store pool is used for this. It can be recovered when all lookup caches are
2619 flushed. Lookups now release memory at the end of their result strings.
2620 This has involved some general refactoring of the lookup sources.
2622 7. Some code has been added to the store_xxx() functions to reduce the amount
2623 of flapping under certain conditions.
2625 8. log_incoming_interface used to affect only the <= reception log lines. Now
2626 it causes the local interface and port to be added to several more SMTP log
2627 lines, for example "SMTP connection from", and rejection lines.
2629 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2631 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2633 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2634 could be overwritten at the end of the current message (or the start of a
2635 new message if it was set in a HELO ACL). The value is now preserved for
2636 the duration of the SMTP connection.
2638 12. If a transport had a headers_rewrite setting, and a matching header line
2639 contained an unqualified address, that address was qualified, even if it
2640 did not match any rewriting rules. The underlying bug was that the values
2641 of the flags that permit the existence of unqualified sender and recipient
2642 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2643 non-local messages, and by -bnq for local messages) were not being
2644 preserved with the message after it was received.
2646 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2647 "next input=" as part of the text comprising the host identity instead of
2648 the correct text. The code was using the same buffer for two different
2649 strings. However, depending on which order the printing function evaluated
2650 its arguments, the bug did not always show up. Under Linux, for example, my
2651 test suite worked just fine.
2653 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2654 doesn't work with some older versions of Perl. It has been changed to "my",
2655 which in any case is probably the better facility to use.
2657 15. A really picky compiler found some instances of statements for creating
2658 error messages that either had too many or two few arguments for the format
2661 16. The size of the buffer for calls to the DNS resolver has been increased
2662 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2663 for addresses that have a lot of PTR records. This alleviates a problem; it
2664 does not fully solve it.
2666 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2667 buffer now truncates the list and logs the incident, which is the same
2668 action as happens when Exim is looking up a host name and its aliases.
2669 Previously in this situation something unpredictable would happen;
2670 sometimes it was "internal error: store_reset failed".
2672 18. If a server dropped the connection unexpectedly when an Exim client was
2673 using GnuTLS and trying to read a response, the client delivery process
2674 crashed while trying to generate an error log message.
2676 19. If a "warn" verb in an ACL added multiple headers to a message in a single
2677 string, for example:
2679 warn message = H1: something\nH2: something
2681 the text was added as a single header line from Exim's point of view
2682 though it ended up OK in the delivered message. However, searching for the
2683 second and subsequent header lines using $h_h2: did not work. This has been
2684 fixed. Similarly, if a system filter added multiple headers in this way,
2685 the routers could not see them.
2687 20. Expanded the error message when iplsearch is called with an invalid key to
2688 suggest using net-iplsearch in a host list.
2690 21. When running tests using -bh, any delays imposed by "delay" modifiers in
2691 ACLs are no longer actually imposed (and a message to that effect is
2694 22. If a "gecos" field in a passwd entry contained escaped characters, in
2695 particular, if it contained a \" sequence, Exim got it wrong when building
2696 a From: or a Sender: header from that name. A second bug also caused
2697 incorrect handling when an unquoted " was present following a character
2698 that needed quoting.
2700 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2701 was not being matched caselessly.
2703 24. Arranged for all hyphens in the exim.8 source to be escaped with
2706 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2707 properly. Recipient callout cache records were still being keyed to include
2708 the sender, even when use_sender was set false. This led to far more
2709 callouts that were necessary. The sender is no longer included in the key
2710 when use_sender is false.
2712 26. Added "control = submission" modifier to ACLs.
2714 27. Added the ${base62d: operator to decode base 62 numbers.
2716 28. dnsdb lookups can now access SRV records.
2718 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2719 the configuration file.
2721 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2722 (-v) mode. This makes the output for a verbose queue run more intelligible.
2724 31. Added a use_postmaster feature to recipient callouts.
2726 32. Added the $body_zerocount variable, containing the number of binary zero
2727 bytes in the message body.
2729 33. The time of last modification of the "new" subdirectory is now used as the
2730 "mailbox time last read" when there is a quota error for a maildir
2733 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2735 35. Added +ignore_unknown as a special item in host lists.
2737 36. Code for decoding IPv6 addresses in host lists is now included, even if
2738 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2739 address was recognized as an IP address, but was then not correctly decoded
2740 into binary, causing unexpected and incorrect effects when compared with
2747 1. Very minor rewording of debugging text in manualroute to say "list of
2748 hosts" instead of "hostlist".
2750 2. If verify=header_syntax was set, and a header line with an unqualified
2751 address (no domain) and a large number of spaces between the end of the
2752 name and the colon was received, the reception process suffered a buffer
2753 overflow, and (when I tested it) crashed. This was caused by some obsolete
2754 code that should have been removed. The fix is to remove it!
2756 3. When running in the test harness, delay a bit after writing a bounce
2757 message to get a bit more predictability in the log output.
2759 4. Added a call to search_tidyup() just before forking a reception process. In
2760 theory, someone could use a lookup in the expansion of smtp_accept_max_
2761 per_host which, without the tidyup, could leave open a database connection.
2763 5. Added the variables $recipient_data and $sender_data which get set from a
2764 lookup success in an ACL "recipients" or "senders" condition, or a router
2765 "senders" option, similar to $domain_data and $local_part_data.
2767 6. Moved the writing of debug_print from before to after the "senders" test
2770 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2771 problems for message scanning, either using a data ACL, or using
2772 local_scan() because the Received: header was not generated till after they
2773 were called (in order to set the time as the time of reception completion).
2774 I have revised the way this works. The header is now generated after the
2775 body is received, but before the ACL or local_scan() are called. After they
2776 are run, the timestamp in the header is updated.
2782 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2783 before starting a queue runner without re-exec. This happened only when
2784 deliver_drop_privilege was set or when the Exim user was set to root. The
2785 effect of the bug was that timeouts during subsequent deliveries caused
2786 crashes instead of being properly handled. The handler is now left at its
2787 default (and expected) setting.
2789 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2790 message, again when deliver_drop_privilege is set or Exim is run as root.
2791 The bug described in (1) was not present in this case, but the tidying up
2792 of the other signals was missing. I have made the two cases consistent.
2794 3. The ignore_target_hosts setting on a manualroute router was being ignored
2795 for hosts that were looked up using the /MX notation.
2797 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2800 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2801 operated on the sender address. After changing the $sender_address to <>
2802 for the sender address verify, Exim was re-instated it as the original
2803 (before rewriting) address, but remembering that it had rewritten it, so it
2804 wasn't rewriting it again. This bug also had the effect of breaking the
2805 sender address verification caching when the sender address was rewritten.
2807 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2808 This has been changed so that if the ip literal address matches
2809 ignore_target_hosts, the router declines.
2811 7. Added expansion conditions match_domain, match_address, and match_local_
2812 part (NOT match_host).
2814 8. The placeholder for the Received: header didn't have a length field set.
2816 9. Added code to Exim itself and to exim_lock to test for a specific race
2817 condition that could lead to file corruption when using MBX delivery. The
2818 issue is with the lockfile that is created in /tmp. If this file is removed
2819 after a process has opened it but before that process has acquired a lock,
2820 there is the potential for a second process to recreate the file and also
2821 acquire a lock. This could lead to two Exim processes writing to the file
2822 at the same time. The added code performs the same test as UW imapd; it
2823 checks after acquiring the lock that its file descriptor still refers to
2824 the same named file.
2826 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2827 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2830 11. Added the smtp_active_hostname option. If used, this will typically be made
2831 to depend on the incoming interface address. Because $interface_address is
2832 not set up until the daemon has forked a reception process, error responses
2833 that can happen earlier (such as "too many connections") no longer contain
2836 12. If an expansion in a condition on a "warn" statement fails because a lookup
2837 defers, the "warn" statement is abandoned, and the next ACL statement is
2838 processed. Previously this caused the whole ACL to be aborted.
2840 13. Added the iplsearch lookup type.
2842 14. Added ident_timeout as a log selector.
2844 15. Added tls_certificate_verified as a log selector.
2846 16. Added a global option tls_require_ciphers (compare the smtp transport
2847 option of the same name). This controls incoming TLS connections.
2849 17. I finally figured out how to make tls_require_ciphers do a similar thing
2850 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2851 before starting the TLS session.
2853 18. Tabs are now shown as \t in -bP output.
2855 19. If the log selector return_path_on_delivery was set, Exim crashed when
2856 bouncing a message because it had too many Received: header lines.
2858 20. If two routers both had headers_remove settings, and the first one included
2859 a superfluous trailing colon, the final name in the first list and the
2860 first name in the second list were incorrectly joined into one item (with a
2861 colon in the middle).
2867 1. Added -C and -D options to the exinext utility, mainly to make it easier
2868 to include in the automated testing, but these could be helpful when
2869 multiple configurations are in use.
2871 2. The exinext utility was not formatting the output nicely when there was
2872 an alternate port involved in the retry record key, nor when there was a
2873 message id as well (for retries that were specific to a specific message
2874 and a specific host). It was also confused by IPv6 addresses, because of
2875 the additional colons they contain. I have fixed the IPv4 problem, and
2876 patched it up to do a reasonable job for IPv6.
2878 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2879 delivery, the log line now contains "pipelined" if PIPELINING was used.
2881 4. An SMTP transport process used to panic and die if the bind() call to set
2882 an explicit outgoing interface failed. This has been changed; it is now
2883 treated in the same way as a connect() failure.
2885 5. A reference to $sender_host_name in the part of a conditional expansion
2886 that was being skipped was still causing a DNS lookup. This no longer
2889 6. The def: expansion condition was not recognizing references to header lines
2890 that used bh_ and bheader_.
2892 7. Added the _cache feature to named lists.
2894 8. The code for checking quota_filecount in the appendfile transport was
2895 allowing one more file than it should have been.
2897 9. For compatibility with Sendmail, the command line option
2905 and sets the incoming protocol and host name (for trusted callers). The
2906 host name and its colon can be omitted when only the protocol is to be set.
2907 Note the Exim already has two private options, -pd and -ps, that refer to
2908 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2909 "s", but I don't think that's a major issue.
2911 10. A number of refactoring changes to the code, none of which should affect
2914 (a) The number of logging options was getting close to filling up the
2915 32-bit word that was used as a bit map. I have split them into two classes:
2916 those that are passed in the argument to log_write(), and those that are
2917 only ever tested independently outside of that function. These are now in
2918 separate 32-bit words, so there is plenty of room for expansion again.
2919 There is no change in the user interface or the logging behaviour.
2921 (b) When building, for example, log lines, the code previously used a
2922 macro that called string_cat() twice, in order to add two strings. This is
2923 not really sufficiently general. Furthermore, there was one instance where
2924 it was actually wrong because one of the argument was used twice, and in
2925 one call a function was used. (As it happened, calling the function twice
2926 did not affect the overall behaviour.) The macro has been replaced by a
2927 function that can join an arbitrary number of extra strings onto a growing
2930 (c) The code for expansion conditions now uses a table and a binary chop
2931 instead of a serial search (which was left over from when there were very
2932 few conditions). Also, it now recognizes conditions like "pam" even when
2933 the relevant support is not compiled in: a suitably worded error message is
2934 given if an attempt is made to use such a condition.
2936 11. Added ${time_interval:xxxxx}.
2938 12. A bug was causing one of the ddress fields not to be passed back correctly
2939 from remote delivery subprocesses. The field in question was not being
2940 subsequently used, so this caused to problems in practice.
2942 13. Added new log selectors queue_time and deliver_time.
2944 14. Might have fixed a bug in maildirsizefile handling that threw up
2945 "unexpected character" debug warnings, and recalculated the data
2946 unnecessarily. In any case, I expanded the warning message to give more
2949 15. Added the message "Restricted characters in address" to the statements in
2950 the default ACL that block characters like @ and % in local parts.
2952 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2953 Three changes have been made:
2955 (a) There was a serious bug; a negative response to MAIL caused the whole
2956 recipient domain to be cached as invalid, thereby blocking all messages
2957 to all local parts at the same domain, from all senders. This bug has
2958 been fixed. The domain is no longer cached after a negative response to
2959 MAIL if the sender used is not empty.
2961 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2964 (c) A new callout option, "use_sender" has been added for people who want
2965 the modified behaviour.
2971 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2974 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2975 indeed breaks things for older releases.
2977 3. Added additional logging to the case where there is a problem reading data
2978 from a filter that is running in a subprocess using a pipe, in order to
2979 try to track down a specific problem.
2981 4. Testing facility fudge: when running in the test harness and attempting
2982 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2983 getting "No route to host". Convert this to a timeout.
2985 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2988 6. Some OS don't have socklen_t but use size_t instead. This affects the
2989 fifth argument of getsockopt() amongst other things. This is now
2990 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2991 can be set for individual OS. I have set it for SunOS5, OSF1, and
2992 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2993 some earlier ones do not.
2995 7. Change 4.30/15 was not doing the test caselessly.
2997 8. The standard form for an IPv6 address literal was being rejected by address
2998 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2999 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3000 this, as well as the form without the "IPv6" on the front (but only when
3001 address literals are enabled, of course).
3003 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3005 10. Exim crashed if a message with an empty sender address specified by -f
3006 encountered a router with an errors_to setting. This could be provoked only
3007 by a command such as
3011 where an empty string was supplied; "<>" did not hit this bug.
3013 11. Installed PCRE release 4.5.
3015 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3016 remained set. It is now erased.
3018 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3019 times from message ids (which are base 36 rather than the normal 62).
3021 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3022 were being counted as actual protocol errors, and logged if the log
3023 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3024 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3025 and DATA following a set of rejected RCPTs do not count as protocol errors.
3026 In other words, Exim assumes they were pipelined, though this may not
3027 actually be the case. Of course, in all cases the client gets an
3028 appropriate error code.
3030 15. If a lookup fails in an ACL condition, a message about the failure may
3031 be available; it is used if testing the ACL cannot continue, because most
3032 such messages specify what the cause of the deferral is. However, some
3033 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3034 that caused an old message to be retained and used if a later statement
3035 caused a defer, replacing the real cause of the deferral.
3037 16. If an IP address had so many PTR records that the DNS lookup buffer
3038 was not large enough to hold them, Exim could crash while trying to process
3039 the truncated data. It now detects and logs this case.
3041 17. Further to 4.21/58, another change has been made: if (and only if) the
3042 first line of a message (the first header line) ends with CRLF, a bare LF
3043 in a subsequent header line has a space inserted after it, so as not to
3044 terminate the header.
3046 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3047 unnecessarily, used atoi() instead of strtol(), and didn't check the
3048 termination when getting file sizes from file names by regex.
3050 19. Completely re-implemented the support for maildirsize files, in the light
3051 of a number of problems with the previous contributed implementation
3052 (4.30/29). In particular:
3054 . If the quota is zero, the maildirsize file is maintained, but no quota is
3057 . If the maildir directory does not exist, it is created before any attempt
3058 to write a maildirsize file.
3060 . The quota value in the file is just a cache; if the quota is changed in
3061 the transport, the new value overrides.
3063 . A regular expression is available for excluding directories from the
3066 20. The autoreply transport checks the characters in options that define the
3067 message's headers; it allows continued headers, but it was checking with
3068 isspace() after an embedded newline instead of explicitly looking for a
3071 21. If all the "regular" hosts to which an address was routed had passed their
3072 expiry times, and had not reached their retry times, the address was
3073 bounced, even if fallback hosts were defined. Now Exim should go on to try
3076 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3077 equivalent code in the SMTP transport. Some hosts send humungous responses
3078 to HELO/EHLO, more than 1024 it seems.
3080 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3081 gives compiler warnings in some environments. I've now done it "properly",
3084 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3085 (because of problems with the built-in one) was declared to return uschar *
3086 instead of char *, causing compiler failure.
3088 25. Fixed a file descriptor leak when processing alias/forward files.
3090 26. Fixed a minor format string issue in dbfn.c.
3092 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3094 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3095 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3098 29. When a sender address is verified, it is cached, to save repeating the test
3099 when there is more than one recipient in a message. However, when the
3100 verification involves a callout, it is possible for different callout
3101 options to be set for different recipients. It is too complicated to keep
3102 track of this in the cache, so now Exim always runs a verification when a
3103 callout is required, relying on the callout cache for the optimization.
3104 The overhead is duplication of the address routing, but this should not be
3107 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3108 to be verified with callout=postmaster, and the main callout worked but the
3109 postmaster check failed, the verification correctly failed. However, if a
3110 subsequent RCPT command asked for sender verification *without* the
3111 postmaster check, incorrect caching caused this verification also to fail,
3114 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3115 it was not caching the DNS options (qualify_single, search_parents) that
3116 were used when the lookup failed. A subsequent lookup with different
3117 options therefore always gave the same answer, though there were cases
3118 where it should not have. (Example: a "domains = !$mx_any" option on a
3119 dnslookup router: the "domains" option is always processed without any
3120 widening, but the router might have qualify_single set.) Now Exim uses the
3121 cached value only when the same options are set.
3123 32. Added John Jetmore's "exipick" utility to the distribution.
3125 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3126 than a timeout (e.g. a certificate is required, and is not provided), an
3127 Exim server now closes the connection immediately. Previously it waited for
3128 the client to close - but if the client is SSL, it seems that they each
3129 wait for each other, leading to a delay before one of them times out.
3131 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3132 maintained 0.8.x compatibility because I don't think many are using it, and
3133 it is clearly obsolete.
3135 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3138 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3139 client certificate was expired. A simple patch fixes this, though I don't
3140 understand the full logic of why the verify callback is called multiple
3143 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3144 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3145 which causes problems with some clients (such as the Certicom SSL Plus
3146 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3147 disables the coutermeasure allowing Eudora to connect."
3149 38. Exim was not checking that a write() to a log file succeeded. This could
3150 lead to Bad Things if a log got too big, in particular if it hit a file
3151 size limit. Exim now panics and dies if it cannot write to a log file, just
3152 as it does if it cannot open a log file.
3154 39. Modified OS/Makefile-Linux so that it now contains
3156 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3158 The two -D definitions ensure that Exim is compiled with large file
3159 support, which makes it possible to handle log files that are bigger than
3162 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3163 instance) a domain was checked against a named list that involved a lookup,
3164 causing $domain_data to be set, then another domain was checked against the
3165 same list, then the first domain was re-checked, the value of $domain_data
3166 after the final check could be wrong. In particular, if the second check
3167 failed, it could be set empty. This bug probably also applied to
3170 41. The strip_trailing_dot option was not being applied to the address given
3171 with the -f command-line option.
3173 42. The code for reading a message's header from the spool was incrementing
3174 $received_count, but never initializing it. This meant that the value was
3175 incorrect (doubled) while delivering a message in the same process in which
3176 it was received. In the most common configuration of Exim, this never
3177 happens - a fresh exec is done - but it can happen when
3178 deliver_drop_privilege is set.
3180 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3181 it now includes up to 150 characters of the unexpected data in the log
3184 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3185 and building data strings. The size of both of these buffers was 10 000
3186 bytes - far larger than anybody would *ever* want, thought I. Needless to
3187 say, somebody hit the limit. I have increased the maximum line length to
3188 20 000 and the maximum data length of concatenated lines to 100 000. I have
3189 also fixed two bugs, because there was no checking on these buffers. Tsk,
3190 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3191 buffer is too small.
3193 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3194 lsearch lookups. Now it does.
3196 46. When parsing a route_list item in a manualroute router, a fixed-length
3197 buffer was used for the list of hosts. I made this 1024 bytes long,
3198 thinking that nobody would ever have a list of hosts that long. Wrong.
3199 Somebody had a whole pile of complicated expansion conditions, and the
3200 string was silently truncated, leading to an expansion error. It turns out
3201 that it is easier to change to an unlimited length (owing to other changes
3202 that have happened since this code was originally written) than to build
3203 structure for giving a limitation error. The length of the item that
3204 expands into the list of hosts is now unlimited.
3206 47. The lsearch lookup could not handle data where the length of text line was
3207 more than 4095 characters. Such lines were truncated, leading to shortened
3208 data being returned. It should now handle lines of any length.
3210 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3211 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3214 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3215 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3216 no longer happens in comment lines. A list of replacements is now placed
3217 at the head of all of the source files, except those whose only change is
3218 to replace PERL_COMMAND in the very first #! line.
3220 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3221 messages on the queue, with a bottom-up merge sort, using code contributed
3222 by Michael Haardt. This should make operations like -bp somewhat faster on
3223 large queues. It won't affect queue runners, except when queue_run_in_order
3226 51. Installed eximstats 1.31 in the distribution.
3228 52. Added support for SRV lookups to the dnslookup router.
3230 53. If an ACL referred to $message_body or $message_body_end, the value was not
3231 reset for any messages that followed in the same SMTP session.
3233 54. The store-handling optimization for building very long strings was not
3234 differentiating between the different store pools. I don't think this
3235 actually made any difference in practice, but I've tidied it.
3237 55. While running the routers to verify a sender address, $sender_address
3238 was still set to the sender address. This is wrong, because when routing to
3239 send a bounce to the sender, it would be empty. Therefore, I have changed
3240 it so that, while verifying a sender address, $sender_address is set to <>.
3241 (There is no change to what happens when verifying a recipient address.)
3243 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3244 target A or AAAA records (if not already returned) without resetting the
3245 qualify_single or search_parents options of the DNS resolver. These are
3246 inappropriate in this case because the targets of MX and SRV records must
3247 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3248 target that, when qualified, matched something in the local domain. These
3249 two options are now turned off when doing these lookups.
3251 57. It seems that at least some releases of Reiserfs (which does not have the
3252 concept of a fixed number of inodes) returns zero and not -1 for the
3253 number of available inodes. This interacted badly with check_spool_inodes,
3254 which assumed that -1 was the "no such thing" setting. What I have done is
3255 to check that the total number of inodes is greater than zero before doing
3256 the test of how many are available.
3258 58. When a "warn" ACL statement has a log_message modifier, the message is
3259 remembered, and not repeated. This is to avoid a lot of repetition when a
3260 message has many recipients that cause the same warning to be written.
3261 Howewer, Exim was preserving the list of already written lines for an
3262 entire SMTP session, which doesn't seem right. The memory is now reset if a
3263 new message is started.
3265 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3266 debugging output unless log_rewrite was also set.
3268 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3269 of (int)(handle) when we know that handle contains (void *)(-1).
3271 61. The Exim daemon panic-logs an error return when it closes the incoming
3272 connection. However "connection reset by peer" seems to be common, and
3273 isn't really an error worthy of noting specially, so that particular error
3276 62. When Exim is trying to find all the local interfaces, it used to panic and
3277 die if the ioctl to get the interface flags failed. However, it seems that
3278 on at least one OS (Solaris 9) it is possible to have an interface that is
3279 included in the list of interfaces, but for which you get a failure error
3280 for this call. This happens when the interface is not "plumbed" into a
3281 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3282 failure of the "get flags" call assumes that the interface is down.
3284 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3285 makes life easier for people who are doing arithmetic on fields extracted
3286 from dates, where you often get leading zeros that should not be
3287 interpreted as octal.
3289 64. Added qualify_domain to the redirect router, to override the global
3292 65. If a pathologically long header line contained very many addresses (the
3293 report of this problem mentioned 10 000) and each of them was rewritten,
3294 Exim could use up a very large amount of memory. (It kept on making new
3295 copies of the header line as it rewrote, and never released the old ones.)
3296 At the expense of a bit more processing, the header rewriting function has
3297 been changed so that it no longer eats memory in this way.
3299 66. The generation of the Received: header has been moved from the time that a
3300 message starts to be received, to the time that it finishes. The timestamp
3301 in the Received: header should now be very close to that of the <= log
3302 line. There are two side-effects of this change:
3304 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3305 logged header lines no longer include the local Received: line, because
3306 it has not yet been created. The same applies to a copy of the message
3307 that is returned to a non-SMTP sender when a message is rejected.
3309 (b) When a filter file is tested using -bf, no additional Received: header
3310 is added to the test message. After some thought, I decided that this
3313 This change does not affect the value of $received_for. It is still set
3314 after address rewriting, but before local_scan() is called.
3316 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3318 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3319 gave an unhelpful panic error message, and a defer error. I have managed to
3320 change this behaviour so that it now rejects any supplied certificate,
3321 which seems right, as the list of acceptable certificates is empty.
3323 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3324 gave an unhelpful defer error. I have not managed to make this reject any
3325 supplied certificates, but the error message it gives is "no certificate
3326 supplied", which is not helpful.
3328 70. exigrep's output now also includes lines that are not associated with any
3329 message, but which match the given pattern. Implemented by a patch from
3330 Martin Sluka, which also tidied up the Perl a bit.
3332 71. Recipient callout verification, like sender verification, was using <> in
3333 the MAIL FROM command. This isn't really the right thing, since the actual
3334 sender may affect whether the remote host accepts the recipient or not. I
3335 have changed it to use the actual sender in the callout; this means that
3336 the cache record is now keyed on a recipient/sender pair, not just the
3337 recipient address. There doesn't seem to be a real danger of callout loops,
3338 since a callout by the remote host to check the sender would use <>.
3339 [SEE ABOVE: changed after hitting problems.]
3341 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3342 temporary errors. However, in the case of such a code being given after
3343 the end of a data transmission (i.e. after ".") Exim was failing to write
3344 a retry record for the message. (Yes, there was some broken host that was
3345 actually sending 8xx at this point.)
3347 73. An unknown lookup type in a host list could cause Exim to panic-die when
3348 the list was checked. (An example that provoked this was putting <; in the
3349 middle of a list instead of at the start.) If this happened during a DATA
3350 ACL check, a -D file could be left lying around. This kind of configuration
3351 error no longer causes Exim to die; instead it causes a defer errror. The
3352 incident is still logged to the main and panic logs.
3354 74. Buglet left over from Exim 3 conversion. The message "too many messages
3355 in one connection" was written to the rejectlog but not the mainlog, except
3356 when address rewriting (yes!) was being logged.
3358 75. Added write_rejectlog option.
3360 76. When a system filter was run not as root (that is, when system_filter_user
3361 was set), the values of the $n variables were not being returned to the
3362 main process; thus, they were not subsequently available in the $sn
3365 77. Added +return_path_on_delivery log selector.
3367 78. A connection timeout was being treated differently from recipients deferred
3368 when testing hosts_max_try with a message that was older than the host's
3369 retry timeout. (The host should not be counted, thus allowing all hosts to
3370 be tried at least once before bouncing.) This may have been the cause of an
3371 occasionally reported bug whereby a message would remain on the queue
3372 longer than the retry timeout, but would be bounced if a delivery was
3373 forced. I say "may" because I never totally pinned down the problem;
3374 setting up timeout/retry tests is difficult. See also the next item.
3376 79. The ultimate address timeout was not being applied to errors that involved
3377 a combination of host plus message (for example, a timeout on a MAIL
3378 command). When an address resolved to a number of possible hosts, and they
3379 were not all tried for each delivery (e.g. because of hosts_max_try), a
3380 message could remain on the queue longer than the retry timeout.
3382 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3385 81. Fixed an obscure SMTP outgoing bug which required at least the following
3386 conditions: (a) there was another message waiting for the same server;
3387 (b) the server returned 5xx to all RCPT commands in the first message so
3388 that the message was not completed; (c) the server dropped the connection
3389 or gave a negative response to the RSET that Exim sends to abort the
3390 transaction. The observed case was a dropped connection after DATA that had
3391 been sent in pipelining mode. That is, the server had advertised PIPELINING
3392 but was not implementing it correctly. The effect of the bug was incorrect
3393 behaviour, such as trying another host, and this could lead to a crash.
3399 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3400 and daemon.c were passed as pointers to ints; they should have been
3401 pointers to socklen_t variables (which are typically unsigned ints).
3403 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3406 3. Fixed a really odd bug that affected only the testing scheme; patching a
3407 certain fixed string in the binary changed the value of another string that
3408 happened to be identical to the end of the original first string.
3410 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3411 name", it returns that address as the IP address. On some operating
3412 systems (e.g. Solaris), it also passes back the IP address string as the
3413 "host name". However, on others (e.g. Linux), it passes back an empty
3414 string. Exim wasn't checking for this, and was changing the host name to an
3415 empty string, assuming it had been canonicized.
3417 5. Although rare, it is permitted to have more than one PTR record for a given
3418 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3419 all the names associated with an address, because they do in Solaris.
3420 However, it seems that they do not in Linux for data that comes from the
3421 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3422 I found this out when I moved to a new Linux workstation and tried to run
3423 the Exim test suite.
3425 To get round this problem I have changed the code so that it now does its
3426 own call to the DNS to look up PTR records when searching for a host name.
3427 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3428 addresses that are only in /etc/hosts are still found.
3430 This behaviour is, however, controlled by an option called host_lookup_
3431 order, which defaults to "bydns:byaddr". If people want to use the other
3432 order, or indeed, just use one or the other means of lookup, they can
3433 specify it in this variable.
3435 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3436 some operating systems, this comes back from gethostbyaddr() as an empty
3437 string, and this is what Exim used to test for. However, it seems that in
3438 other systems, "." is yielded. Exim now tests for this case too.
3440 7. The values of check_spool_space and check_log_space are now held internally
3441 as a number of kilobytes instead of an absolute number of bytes. If a
3442 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3443 kilobyte. This means that much larger values can be stored.
3445 8. Exim monitor: an attempt to get the action menu when not actually pointing
3446 at a message produces an empty menu entitled "No message selected". This
3447 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3448 no entries in it ("Shell widget menu has zero width and/or height"). So I
3449 have added a single, blank menu entry in this case.
3451 9. Added ${quote_local_part.
3453 10. MIME decoding is now applied to the contents of Subject: header lines when
3456 11. Now that a reference to $sender_host_address automatically causes a reverse
3457 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3458 host lookup before query-style lookups in lists that might use this
3459 variable. This has therefore been abolished, and the "net-" prefix is no
3460 longer necessary for query-style lookups.
3462 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3463 have been a typo for LFLAGS, so it has been changed.
3465 13. The install script calls Exim with "-C /dev/null" in order to find the
3466 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3467 to be output. Howeve, since Exim outputs its version number before the
3468 error, it didn't break the script. It just looked ugly. I fixed this by
3469 always allowing "-C /dev/null" if the caller is root.
3471 14. Ignore overlarge ACL variable number when reading spool file - insurance
3472 against a later release with more variables having written the file.
3474 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3475 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3476 this, as well as the form without the "IPv6" on the front.
3478 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3479 OS/Makefile-Darwin file.
3481 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3482 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3484 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3485 variable. However, some OS use ints for inodes. I've added cast to long int
3486 to get rid of the compiler warning.
3488 19. I had forgotten to lock out "/../" in configuration file names when
3489 ALT_CONFIG_PREFIX was set.
3491 20. Routers used for verification do not need to specify transports. However,
3492 if such a router generated a host list, and callout was configured, Exim
3493 crashed, because it could not find a port number from the (non-existent)
3494 transport. It now assumes port 25 in this circumstance.
3496 21. Added the -t option to exigrep.
3498 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3499 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3500 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3501 them is set, LOOKUP_LSEARCH is forced.
3503 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3505 24. Added sender and host information to the "rejected by local_scan()" log
3506 line; previously there was no indication of these.
3508 25. Added .include_if_exists.
3510 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3511 Linux. It turns out that not all file systems support this. Apparently some
3512 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3513 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3514 not supported on the file descriptor, is now ignored when Exim is trying to
3515 sync a directory. This applies only to Linux.
3517 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3519 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3520 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3523 29. Added support for maildirsize files from supplied patch (modified a bit).
3525 30. The use of :fail: followed by an empty string could lead Exim to respond to
3526 sender verification failures with (e.g.):
3528 550 Verification failed for <xxx>
3529 550 Sender verify failed
3531 where the first response line was missing the '-' that indicates it is not
3532 the final line of the response.
3534 31. The loop for finding the name of the user that called Exim had a hardwired
3535 limit of 10; it now uses the value of finduser_retries, which is used for
3536 all other user lookups.
3538 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3541 33. Exim was neglecting to zero errno before one call of strtol() when
3542 expanding a string and expecting an integer value. On some systems this
3543 resulted in spurious "integer overflow" errors. Also, it was casting the
3544 result into an int without checking.
3546 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3547 did not work. The code looks as if it has *never* worked, though it appears
3548 to have been documented since at least releast 1.62. I have made it work.
3550 35. The "timeout_DNS" error in retry rules, also documented since at least
3551 1.62, also never worked. As it isn't clear exactly what this means, and
3552 clearly it isn't a major issue, I have abolished the feature by treating it
3553 as "timeout", and writing a warning to the main and panic logs.
3555 36. The display of retry rules for -brt wasn't always showing the error code
3558 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3559 timeout_connect_A, timeout_connect_MX.
3561 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3562 to the empty sender.
3564 39. The daemon was not analysing the content of -oX till after it had closed
3565 stderr and disconnected from the controlling terminal. This meant that any
3566 syntax errors were only noted on the panic log, and the return code from
3567 the command was 0. By re-arranging the code a little, I've made the
3568 decoding happen first, so such errors now appear on stderr, and the return
3569 code is 1. However, the actual setting up of the sockets still happens in
3570 the disconnected process, so errors there are still only recorded on the
3573 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3574 connections (as happens on some IP stacks) was logged at start up time as
3575 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3576 it from "IPv6 and IPv4", which means that two separate sockets are being
3579 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3580 says whether AF_INET or AF_INET6 was passed as an argument.
3582 42. Exiwhat output was messed up when time zones were included in log
3585 43. Exiwhat now gives more information about the daemon's listening ports,
3586 and whether -tls-on-connect was used.
3588 44. The "port" option of the smtp transport is now expanded.
3590 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3591 silently ignored. Now an error message is written to the main and panic
3594 46. There's a new ACL modifier called "logwrite" which writes to a log file
3595 as soon as it is encountered.
3597 47. Added $local_user_uid and $local_user_gid at routing time.
3599 48. Exim crashed when trying to verify a sender address that was being
3602 49. Exim was recognizing only a space character after ".include". It now also
3603 recognizes a tab character.
3605 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3606 extracting the relevant information from the specification. The man page no
3607 longer contains scrambled data for the -d option, and I've added a section
3608 at the front about calling Exim under different names.
3610 51. Added "extra_headers" argument to the "mail" command in filter files.
3612 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3615 53. Installed eximstats 1.29.
3617 54. Added transport_filter_timeout as a generic transport option.
3619 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3620 Cc: header lines. This was required by RFC 822, but it not required by RFC
3623 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3624 incoming messages that did not have them. Now it does so only if the
3625 message originates locally, that is, if there is no associated remote host
3626 address. When Resent- header lines are present, this applies to the Resent-
3627 lines rather than the non-Resent- lines.
3629 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3630 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3632 58. Messages for configuration errors now include the name of the main
3633 configuration file - useful now that there may be more than one file in a
3634 list (.included file names were always shown).
3636 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3637 for those rare installations that do not start the daemon as root or run it
3638 setuid root. I've cut out the call to initgroups() if the daemon is not
3641 60. The Exim user and group can now be bound into the binary as text strings
3642 that are looked up at the start of Exim's processing.
3644 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3646 62. Added $mailstore_basename variable.
3648 63. Installed patch to sieve.c from Michael Haardt.
3650 64. When Exim failed to open the panic log after failing to open the main log,
3651 the original message it was trying to log was written to stderr and debug
3652 output, but if they were not available (the usual case in production), it
3653 was lost. Now it is written to syslog before the two lines that record the
3654 failures to open the logs.
3656 65. Users' Exim filters run in subprocesses under the user's uid. It is
3657 possible for a "deliver" command or an alias in a "personal" command to
3658 provoke an address rewrite. If logging of address rewriting is configured,
3659 this fails because the process is not running as root or exim. There may be
3660 a better way of dealing with this, but for the moment (because 4.30 needs
3661 to be released), I have disabled address rewrite logging when running a
3662 filter in a non-root, non-exim process.
3668 1. The buildconfig auxiliary program wasn't quoting the value set for
3669 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3670 not defined. This bug was masked in 4.22 by the effect that was fixed in
3673 2. Some messages that were rejected after a message id was allocated were
3674 shown as "incomplete" by exigrep. It no longer does this for messages that
3675 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3677 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3678 have allow_domain_literals set, the ID did not get logged in the <= line.
3679 Domain literals are now always recognized in Message-ID: header lines.
3681 4. The first argument for a ${extract expansion item is the key name or field
3682 number. Leading and trailing spaces in this item were not being ignored,
3683 causing some misleading effects.
3685 5. When deliver_drop_privilege was set, single queue runner processes started
3686 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3687 same command in the process it spins off) were not dropping privilege.
3689 6. When the daemon running as "exim" started a queue runner, it always
3690 re-executed Exim in the spun-off process. This is a waste of effort when
3691 deliver_drop_privilege is set. The new process now just calls the
3692 queue-runner function directly.
3698 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3701 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3702 ignored. Though the use of -oP was forcing the writing of a pid file, it
3703 was always written to the default place.
3705 3. If the message "no IP address found for host xxxx" is generated during
3706 incoming verification, it is now followed by identification of the incoming
3707 connection (so you can more easily find what provoked it).
3709 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3711 5. Added some features to "harden" Exim a bit more against certain attacks:
3713 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3714 be put in Local/Makefile. This is like the never_users runtime option,
3715 but it cannot be overridden. The default setting is "root".
3717 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3718 prefix string with which any file named in a -C command line option
3721 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3722 is retained for -C and -D only if the caller of Exim is root. Without
3723 it, the exim user may also use -C and -D and retain privilege.
3725 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3726 command line option is disabled.
3728 6. Macro names set by the -D option must start with an upper case letter, just
3729 like macro names defined in the configuration file.
3731 7. Added "dereference=" facility to LDAP.
3733 8. Two instances of the typo "uknown" in the source files are fixed.
3735 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3736 the Configure-Makefile script screwed up while processing it.
3738 10. Incorporated PCRE 4.4.
3740 11. The SMTP synchronization check was not operating right at the start of an
3741 SMTP session. For example, it could not catch a HELO sent before the client
3742 waited for the greeting. There is now a check for outstanding input at the
3743 point when the greeting is written. Because of the duplex, asynchronous
3744 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3745 way, but not yet received, when the check is performed.
3747 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3748 on TCP/IP sockets, because this apparently causes some broken clients to
3751 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3752 unchanged) from the Cygwin maintainer.
3754 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3755 was supported instead of testing for mbx. Effectively a typo.
3757 15. The spa authenticator server code was not checking that the input it
3758 received was valid base64.
3760 16. The debug output line for the "set" modifier in ACLs was not showing the
3761 name of the variable that was being set.
3763 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3765 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3766 lookup on its own. Something else was needed to trigger the lookup. For
3767 example, a match in host_lookup or the need for a host name in a host list.
3768 Now, if $sender_host_name is referenced and the host name has not yet been
3769 looked up, a lookup is performed. If the lookup fails, the variable remains
3770 empty, and $host_lookup_failed is set to "1".
3772 19. Added "eqi" as a case-independent comparison operator.
3774 20. The saslauthd authentication condition could segfault if neither service
3775 nor realm was specified.
3777 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3778 error message that was logged was misleading, and incoming SMTP
3779 connections were dropped. The message is now more accurate, and temporary
3780 errors are given to SMTP connections.
3782 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3783 (except RSET) with a 421 error, until QUIT is received. However, it was
3784 failing to send a response to QUIT.
3786 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3787 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3788 ACL is now run after the helo_try_verify_hosts code.
3790 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3791 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3792 case-independent, so other case variants are also recognized). Apparently
3793 some systems use these upper case variants.
3795 25. If more than two messages were waiting for the same host, and a transport
3796 filter was specified for the transport, Exim sent two messages over the
3797 same TCP/IP connection, and then failed with "socket operation on non-
3798 socket" when it tried to send the third.
3800 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3802 27. The extern definition of crypt16() in expand.c was not being excluded when
3803 the OS had its own crypt16() function.
3805 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3806 as a preferred synonym for return_size_limit, both as an option and as an
3809 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3811 30. Changed the default configuration ACL to relax the local part checking rule
3812 for addresses that are not in any local domains. For these addresses,
3813 slashes and pipe symbols are allowed within local parts, but the sequence
3814 /../ is explicitly forbidden.
3816 31. SPA server authentication was not clearing the challenge buffer before
3819 32. log_message in a "warn" ACL statement was writing to the reject log as
3820 well as to the main log, which contradicts the documentation and doesn't
3821 seem right (because no rejection is happening). So I have stopped it.
3823 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3824 However, I am unable to do any testing of this.
3826 34. Fixed an infelicity in the appendfile transport. When checking directories
3827 for a mailbox, to see if any needed to be created, it was accidentally
3828 using path names with one or more superfluous leading slashes; tracing
3829 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3831 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3832 contents are added to the log line that is written for every discarded
3833 recipient. (Previously a log_message setting was ignored.)
3835 36. The ${quote: operator now quotes the string if it is empty.
3837 37. The install script runs exim in order to find its version number. If for
3838 some reason other than non-existence or emptiness, which it checks, it
3839 could not run './exim', it was installing it with an empty version number,
3840 i.e. as "exim-". This error state is now caught, and the installation is
3843 38. An argument was missing from the function that creates an error message
3844 when Exim fails to connect to the socket for saslauthd authentication.
3845 This could cause Exim to crash, or give a corrupted message.
3847 39. Added isip, isip4, and isip6 to ${if conditions.
3849 40. The ACL variables $acl_xx are now saved with the message, and can be
3850 accessed later in routers, transports, and filters.
3852 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3853 strings in the file are not string-expanded.
3855 42. If a MAIL command specified a SIZE value that was too large to fit into an
3856 int variable, the check against message_size_limit failed. Such values are
3857 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3858 day this will have to be increased, but I don't think I want to be around
3859 when emails are that large.
3866 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3867 iconv() is not standard in FreeBSD.
3869 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3870 IPv6 enabled. The observed symptom was a segmentation fault on return from
3871 the function os_common_find_running_interfaces() in src/os.c.
3873 3. In the check_special_case() function in daemon.c I had used "errno" as an
3874 argument name, which causes warnings on some systems. This was basically a
3875 typo, since it was named "eno" in the comments!
3877 4. The code that waits for the clock to tick (at a resolution of some fraction
3878 of a second) so as to ensure message-id uniqueness was always waiting for
3879 at least one whole tick, when it could have waited for less. [This is
3880 almost certainly not relevant at current processor speeds, where it is
3881 unlikely to ever wait at all. But we try to future-proof.]
3883 5. The function that sleeps for a time interval that includes fractions of a
3884 second contained a race. It did not block SIGALRM between setting the
3885 timer, and suspending (a couple of lines later). If the interval was short
3886 and the sigsuspend() was delayed until after it had expired, the suspension
3887 never ended. On busy systems this could lead to processes getting stuck for
3890 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3891 process, before it forks any delivery processes. The open lookup caching
3892 mechanism meant that the open file or database connection was passed into
3893 the delivery process. The problem was that delivery processes always tidy
3894 up cached lookup data. This could cause a problem for the next delivery
3895 process started by the queue runner, because the external queue runner
3896 process does not know about the closure. So the next delivery process
3897 still has data in the lookup cache. In the case of a file lookup, there was
3898 no problem because closing a file descriptor in a subprocess doesn't affect
3899 the parent. However, if the lookup was caching a connection to a database,
3900 the connection was closed, and the second delivery process was likely to
3901 see errors such as "PGSQL: query failed: server closed the connection
3902 unexpectedly". The problem has been fixed by closing all cached lookups
3903 in a queue runner before running a delivery process.
3905 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3906 seem to have the "const" qualifier which it has on other OS. I've
3909 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3910 *of the same type* (client or server) with the same public name that an
3911 error should be diagnosed.
3913 9. When Exim looked up a host name for an IP address, but failed to find the
3914 original IP address when looking up the host name (a safety check), it
3915 output the message "<ip address> does not match any IP for NULL", which was
3916 confusing, to say the least. The bug was that the host name should have
3917 appeared instead of "NULL".
3919 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3920 user that is built into the binary, and the -C or -D options is used, root
3921 privilege is dropped before the configuration file is read. In addition,
3922 logging is switched to stderr instead of the normal log files. If the
3923 configuration then re-defines the Exim user, the unprivileged environment
3924 is probably not what is expected, so Exim logs a panic warning message (but
3927 However, if deliver_drop_privilege is set, the unprivileged state may well
3928 be exactly what is intended, so the warning has been cut out in that case,
3929 and Exim is allowed to try to write to its normal log files.
3935 1. smtp_return_error_details was not giving details for temporary sender
3936 or receiver verification errors.
3938 2. Diagnose a configuration error if two authenticators have the same public
3941 3. Exim used not to create the message log file for a message until the first
3942 delivery attempt. This could be confusing when incoming messages were held
3943 for policy or load reasons. The message log file is now created at the time
3944 the message is received, and an initial "Received" line is written to it.
3946 4. The automatically generated man page for command line options had a minor
3947 bug that caused no ill effects; however, a more serious problem was that
3948 the procedure for building the man page automatically didn't always
3949 operate. Consequently, release 4.20 contains an out-of-date version. This
3950 shouldn't happen again.
3952 5. When building Exim with embedded Perl support, the script that builds the
3953 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3954 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3956 6. The freeze_tell option was not being used for messages that were frozen on
3957 arrival, either by an ACL or by local_scan().
3959 7. Added the smtp_incomplete_transaction log selector.
3961 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3962 was accepting AUTH without a new EHLO.
3964 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3965 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3968 10. Logging of TCP/IP connections (when configured) now happens in the main
3969 daemon process instead of the child process, so that the TCP/IP connection
3970 count is more accurate (but it can never be perfect).
3972 11. The use of "drop" in a nested ACL was not being handled correctly in the
3973 outer ACL. Now, if condition failure induced by the nested "drop" causes
3974 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3975 or "require"), the connection is dropped.
3977 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3978 that yield "discard" can now be used with an "accept" or a "discard" verb,
3979 but an error is generated for any others (because I can't see a useful way
3980 to define what should happen).
3982 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3983 are now processed in the same way as lines in the Exim configuration file.
3984 In particular, continuation lines are supported.
3986 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3988 15. Added -ti meaning -t -i.
3990 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3991 domains, and warn by logging if others are found.
3993 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3994 returned from GIFCONF when Exim is trying to find the list of interfaces on
3995 a host. The code in os.c has been modified to copy each ifreq to an aligned
3996 structure in all cases.
3998 Also, in some cases, the returned ifreq's were being copied to a 'struct
3999 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4000 means the last couple of bytes of an IPv6 address could be chopped if the
4001 ifreq contained only a normal sockaddr (14 bytes storage).
4003 18. Named domain lists were not supported in the hosts_treat_as_local option.
4004 An entry such as +xxxx was not recognized, and was treated as a literal
4007 19. Ensure that header lines added by a DATA ACL are included in the reject log
4008 if the ACL subsequently rejects the message.
4010 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4011 MD5 (which is deprecated).
4013 21. When testing a filter file using -bf, Exim was writing a message when it
4014 took the sender from a "From " line in the message, but it was not doing so
4015 when it took $return_path from a Return-Path: header line. It now does.
4017 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4018 with a valid header line field name (a series of printing characters
4019 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4021 23. Changed "disc" in the source to "disk" to conform to the documentation and
4022 the book and for uniformity.
4024 24. Ignore Sendmail's -Ooption=value command line item.
4026 25. When execve() failed while trying to run a command in a pipe transport,
4027 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4028 could be confused with a return value of 69 from the command itself. This
4029 has been changed to 127, the value the shell returns if it is asked to run
4030 a non-existent command. The wording for the related log line suggests a
4031 non-existent command as the problem.
4033 26. If received_header_text expands to an empty string, do not add a Received:
4034 header line to the message. (Well, it adds a token one on the spool, but
4035 marks it "old" so that it doesn't get used or transmitted.)
4037 27. Installed eximstats 1.28 (addition of -nt option).
4039 28. There was no check for failure on the call to getsockname() in the daemon
4040 code. This can fail if there is a shortage of resources on the system, with
4041 ENOMEM, for example. A temporary error is now given on failure.
4043 29. Contrary to the C standard, it seems that in some environments, the
4044 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4045 program. Exim now does this explicitly; it affects the formatting of
4046 timestamps using strftime().
4048 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4049 complaints. I've now initialized all the variables, to avoid this.
4051 32. Header lines added by a system filter were not being "seen" during
4052 transport-time rewrites.
4054 33. The info_callback() function passed to OpenSSL is set up with type void
4055 (*)(SSL *, int, int), as described somewhere. However, when calling the
4056 function (actually a macro) that sets it up, the type void(*)() is
4057 expected. I've put in a cast to prevent warnings from picky compilers.
4059 34. If a DNS black list lookup found a CNAME record, but there were no A
4060 records associated with the domain it pointed at, Exim crashed.
4062 35. If a DNS black list lookup returned more than one A record, Exim ignored
4063 all but the first. It now scans all returned addresses if a particular IP
4064 value is being sought. In this situation, the contents of the
4065 $dnslist_value variable are a list of all the addresses, separated by a
4068 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4069 to accept a host name and all its aliases if the forward lookup for any of
4070 them yielded the IP address of the incoming connection. Now it accepts only
4071 those names whose forward lookup yields the correct IP address. Any other
4072 names are discarded. This closes a loophole whereby a rogue DNS
4073 administrator could create reverse DNS records to break through a
4074 wildcarded host restriction in an ACL.
4076 37. If a user filter or a system filter that ran in a subprocess used any of
4077 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4078 the wrong values were passed to the pipe command ($thisaddress had the
4079 value of $0, $0 had the value of $1, etc). This bug was introduced by
4080 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4082 38. Improved the line breaking for long SMTP error messages from ACLs.
4083 Previously, if there was no break point between 40 and 75 characters, Exim
4084 left the rest of the message alone. Two changes have been made: (a) I've
4085 reduced the minimum length to 35 characters; (b) if it can't find a break
4086 point between 35 and 75 characters, it looks ahead and uses the first one
4087 that it finds. This may give the occasional overlong line, but at least the
4088 remaining text gets split now.
4090 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4091 file descriptors might be low, and that setting 1000 would always raise it.
4092 It turns out that in some environments, the limit is already over 1000 and
4093 that lowering it causes trouble. So now Exim takes care not to decrease it.
4095 40. When delivering a message, the value of $return_path is set to $sender_
4096 address at the start of routing (routers may change the value). By an
4097 oversight, this default was not being set up when an address was tested by
4098 -bt or -bv, which affected the outcome if any router or filter referred to
4101 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4102 log only when "message" or "log_message" are set. However, if one of the
4103 conditions was an address verification, or a call to a nested ACL, the
4104 messages generated by the underlying test were being passed through. This
4105 no longer happens. The underlying message is available in $acl_verify_
4106 message for both "message" and "log_message" expansions, so it can be
4107 passed through if needed.
4109 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4110 new expansion $bh_ to give the encoded byte string without charset
4111 translation. Translation happens only if iconv() is available; HAVE_ICONV
4112 indicates this at build time. HEADERS_CHARSET gives the charset to
4113 translate to; headers_charset can change it in the configuration, and
4114 "headers charset" can change it in an individual filter file.
4116 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4117 that creates RFC 2047 encoded "words" labels them as that charset instead
4118 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4119 expansion operator; (ii) when Exim creates a From: line for a local
4120 message; (iii) when a header line is rewritten to include a "phrase" part.
4122 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4123 buggy, causing it to skip the first lines of messages whose message ID
4124 ended in 'D'. This would not have bitten before Exim release 4.14, because
4125 message IDs were unlikely to end in 'D' before then. The effect was to have
4126 incorrect size information for certain domains.
4128 45. #include "config.h" was missing at the start of the crypt16.c module. This
4129 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4132 46. If there was a timeout during a "random" callout check, Exim treated it as
4133 a failure of the random address, and carried on sending RSET and the real
4134 address. If the delay was just some slowness somewhere, the response to the
4135 original RCPT would be taken as a response to RSET and so on, causing
4136 mayhem of various kinds.
4138 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4139 when I implemented it. It didn't allow for the fact that some option values
4140 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4141 the right test for positive values.
4143 48. Domain names in DNS records are case-independent. Exim always looks them up
4144 in lower case. Some resolvers return domain names in exactly the case they
4145 appear in the zone file, that is, they may contain uppercase letters. Not
4146 all resolvers do this - some return always lower case. Exim was treating a
4147 change of case by a resolver as a change of domain, similar to a widening
4148 of a domain abbreviation. This triggered its re-routing code and so it was
4149 trying to route what was effectively the same domain again. This normally
4150 caused routing to fail (because the router wouldn't handle the domain
4151 twice). Now Exim checks for this case specially, and just changes the
4152 casing of the domain that it ultimately uses when it transmits the message
4155 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4158 50. If a filter generated a file delivery with a non-absolute name (possible if
4159 no home directory exists for the router), the forbid_file option was not
4162 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4163 the existing equality matching.
4165 52. Exim was using ints instead of ino_t variables in some places where it was
4166 dealing with inode numbers.
4168 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4169 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4170 called TMPDIR, and if it finds it is different, it changes its value.
4172 54. The smtp_printf() function is now made available to local_scan() so
4173 additional output lines can be written before returning. There is also an
4174 smtp_fflush() function to enable the detection of a dropped connection.
4175 The variables smtp_input and smtp_batched_input are exported to
4178 55. Changed the default runtime configuration: the message "Unknown user"
4179 has been removed from the ACL, and instead placed on the localuser router,
4180 using the cannot_route_message feature. This means that any verification
4181 failures that generate their own messages won't get overridden. Similarly,
4182 the "Unrouteable address" message that was in the ACL for unverifiable
4183 relay addresses has also been removed.
4185 56. Added hosts_avoid_esmtp to the smtp transport.
4187 57. The exicyclog script was not checking for the esoteric option
4188 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4189 will work only if exicyclog is run under the appropriate euid.
4191 58. Following a discussion on the list, the rules by which Exim recognises line
4192 endings on incoming messages have been changed. The -dropcr and drop_cr
4193 options are now no-ops, retained only for backwards compatibility. The
4194 following line terminators are recognized: LF CRLF CR. However, special
4195 processing applies to CR:
4197 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4198 nor a local message in the state where . is a terminator.
4200 (ii) If a bare CR is encountered in a header line, an extra space is added
4201 after the line terminator so as not to end the header. The reasoning
4202 behind this is that bare CRs in header lines are most likely either
4203 to be mistakes, or people trying to play silly games.
4205 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4206 was being incorrectly given as 18 bytes larger than it should have been.
4207 This is a VOB (very old bug).
4209 60. This may never have affected anything current, but just in case it has:
4210 When the local host is found other than at the start of a list of hosts,
4211 the local host, those with the same MX, and any that follow, are discarded.
4212 When the list in question was part of a longer list of hosts, the following
4213 hosts (not currently being processed) were also being discarded. This no
4214 longer happens. I'm not sure if this situation could ever has previously
4217 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4220 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4221 header. This is something that is recommended in a new Internet Draft, and
4222 is something that is documented as being done by Sendmail. There are two
4223 possible values. For messages generated by the autoreply transport, Exim
4226 Auto-Submitted: auto-replied
4228 whereas for all other generated messages (e.g. bounces) it adds
4230 Auto-Submitted: auto-generated
4232 63. The "personal" condition in filters now includes a test for the
4233 Auto-Submitted: header. If it contains the string "auto-" the message it
4234 not considered personal.
4236 64. Added rcpt_include_affixes as a generic transport option.
4238 65. Added queue_only_override (default true).
4240 66. Added the syslog_duplication option.
4242 67. If what should have been the first header line of a message consisted of
4243 a space followed by a colon, Exim was mis-interpreting it as a header line.
4244 It isn't of course - it is syntactically invalid and should therefore be
4245 treated as the start of the message body. The misbehaviour could have
4246 caused a number of strange effects, including loss of data in subsequent
4247 header lines, and spool format errors.
4249 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4250 client host had authenticated. This control can now be exercised by an ACL
4251 for more flexibility.
4253 69. By default, callouts do not happen when testing with -bh. There is now a
4254 variant, -bhc, which does actually run the callout code, including
4255 consulting and updating the callout cache.
4257 70. Added support for saslauthd authentication, courtesy of Alexander
4260 71. If statvfs() failed on the spool or log directories while checking their
4261 size for availability, Exim confusingly gave the error "space shortage".
4262 Furthermore, in debugging mode it crashed with a floating point exception.
4263 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4264 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4265 serious problem, Exim now writes to the main and panic logs when this
4266 happens, with details of the failure. It then refuses to accept the
4267 incoming message, giving the message "spool directory problem" or "log
4268 directory problem" with a 421 code for SMTP messages.
4270 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4271 0, 1, and 2 exist, because some OS complain for execs without them (see
4272 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4273 descriptors. However, the code omitted to check that the open succeeded,
4274 causing mysterious errors if for some reason the permissions on /dev/null
4275 got screwed. Now Exim writes a message to the main and panic logs, and
4276 bombs out if it can't open /dev/null.
4278 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4279 interact so that it is all more flexible. It is supposed to remain
4280 backwards compatible. Also added extra_local_interfaces.
4282 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4283 to bomb out with an assertion failure - to the client this appears as a
4284 connection drop. This problem occurs in the part of the code that was taken
4285 from the Samba project. Fortunately, the assertion is in a very simple
4286 function, so I have fixed this by reproducing the function inline in the
4287 one place where it is called, and arranging for authentication to fail
4288 instead of killing the process with assert().
4290 75. The SPA client code was not working when the server requested OEM rather
4291 than Unicode encoding.
4293 76. Added code to make require_files with a specific uid setting more usable in
4294 the case where statting the file as root fails - usually a non-root-mounted
4295 NFS file system. When this happens and the failure is EACCES, Exim now
4296 forks a subprocess and does the per-uid checking as the relevant uid.
4298 77. Added process_log_path.
4300 78. If log_file_path was not explicitly set, a setting of check_log_space or
4301 check_log_inodes was ignored.
4303 79. If a space check for the spool or log partitions fails, the incident is now
4304 logged. Of course, in the latter case the data may get lost...
4306 80. Added the %p formatting code to string_format() so that it can be used to
4307 print addresses in debug_print(). Adjusted all the address printing in the
4308 debugging in store.c to use %p rather than %d.
4310 81. There was a concern that a line of code in smtp_in.c could overflow a
4311 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4312 initially expressed, the concern was not well-founded, because trailing
4313 spaces are removed early. However, if the trailing spaces were followed by
4314 a NULL, they did not get removed, so the overflow was possible. Two fixes
4317 (a) I re-wrote the offending code in a cleaner fashion.
4318 (b) If an incoming SMTP command contains a NULL character, it is rejected
4321 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4322 runs initgroups(), so that if the Exim user is in any additional groups,
4323 they will be used during message reception.
4329 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob