1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.413 2006/10/23 13:24:21 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.
171 SC/01 Use a glob alias rather than an array ref in eximstats generated
172 parser. This improves both readability and performance.
174 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
175 Don't display local sender or destination tables in eximstats unless
176 there is data to show.
177 Added average volumes into the eximstats top table text output.
179 SC/03 Collect data on the number of addresses (recipients) as well
180 as the number of messages in eximstats.
182 TF/01 Correct an error in the documentation for the redirect router. Exim
183 does (usually) call initgroups() when daemonizing.
185 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
186 with consistent privilege compared to when running as a daemon.
188 TF/03 Note in the spec that $authenticated_id is not set for local
189 submissions from trusted users.
191 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
192 Thanks to Dean Brooks <dean@iglou.com> for the patch.
194 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
195 by adding some example configuration directives to the default
196 configuration file. A little bit of work is required to uncomment the
197 directives and define how usernames and passwords are checked, but
198 there is now a framework to start from.
200 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
201 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
202 without this. I don't know how relevant this is to other LDAP libraries.
204 PH/02 Add the verb name to the "unknown ACL verb" error.
206 PH/03 Magnus Holmgren's patch for filter_prepend_home.
208 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
210 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
211 directory not expanded when it should be if an expanded home directory
212 was set for the address (which is overridden by the transport).
214 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
217 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
218 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
219 because it is too late at that time, and has no effect.
221 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
222 security issue with \' (bugzilla #107). I could not use the
223 PQescapeStringConn() function, because it needs a PGconn value as one of
226 PH/08 When testing addresses using -bt, indicate those final addresses that
227 are duplicates that would not cause an additional delivery. At least one
228 person was confused, thinking that -bt output corresponded to deliveries.
229 (Suppressing duplicates isn't a good idea as you lose the information
230 about possibly different redirections that led to the duplicates.)
232 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
233 systems where poll() doesn't work, in particular OS X.
235 PH/10 Added more information to debugging output for retry time not reached.
237 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
238 operations in malware.c.
240 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
243 PH/13 If write_rejectlog was set false when logging was sent to syslog with
244 syslog_duplication set false, log lines that would normally be written
245 both the the main log and to the reject log were not written to syslog at
248 PH/14 In the default configuration, change the use of "message" in ACL warn
249 statements to "add_header".
251 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
252 not followed by a command (e.g. "seen endif").
254 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
255 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
258 PH/17 Added extra conditions to the default value of delay_warning_condition
262 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
263 { match{$h_precedence:}{(?i)bulk|list|junk} } \
264 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
267 The Auto-Submitted: and various List- headers are standardised, whereas I
268 don't think Precedence: ever was.
270 PH/18 Refactored debugging code in route_finduser() to show more information,
271 in particular, the error code if getpwnam() issues one.
273 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
274 This is apparently needed in addition to the PH/07 change above to avoid
275 any possible encoding problems.
277 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
278 but not after initializing Perl.
280 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
281 output them only if debugging. By default they are written stderr,
282 apparently, which is not desirable.
284 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
287 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
290 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
292 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
293 authenticated or an ident call has been made. Suppress the default
294 values for $authenticated_id and $authenticated_sender (but permit -oMai
295 and -oMas) when testing with -bh.
297 PH/24 Re-jigged the order of the tests in the default configuration so that the
298 tests for valid domains and recipients precede the DNS black list and CSA
299 tests, on the grounds that those ones are more expensive.
301 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
302 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
303 This bug exists in every version of Exim that I still have, right back to
306 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
307 However, an attempt to turn on case-sensitivity in a regex key by
308 including (?-i) didn't work because the subject string was already
309 lowercased, and the effects were non-intuitive. It turns out that a
310 one-line patch can be used to allow (?-i) to work as expected.
316 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
317 other effects) broke the use of negated acl sub-conditions.
319 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
322 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
323 "Deny" causes Exim to reject the incoming connection with a 554 error.
324 Unfortunately, if there is a major crisis, such as a disk failure,
325 tcp-wrappers gives "deny", whereas what one would like would be some
326 kind of temporary error. A kludge has been added to help with this.
327 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
328 554 error is used if errno is still zero or contains ENOENT (which occurs
329 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
332 PH/03 Add -lutil to the default FreeBSD LIBS setting.
334 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
335 errors. Otherwise a message that provokes a temporary error (when other
336 messages do not) can cause a whole host to time out.
338 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
339 addresses were routed directly to files or pipes from a redirect router.
340 File deliveries just didn't batch; pipe deliveries might have suffered
343 PH/06 A failure to get a lock for a hints database would erroneously always say
344 "Failed to get write lock", even when it was really a read lock.
346 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
347 of 0600. This has been changed to use the value of the lockfile_mode
348 option (which defaults to 0600).
350 PH/08 Applied small patch from the Sieve maintainer.
352 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
353 folder from quota calculations, a direct delivery into this folder messed
354 up the contents of the maildirsize file. This was because the regex was
355 used only to exclude .Trash (or whatever) when the size of the mailbox
356 was calculated. There was no check that a delivery was happening into an
357 excluded directory. This bug has been fixed by ignoring all quota
358 processing for deliveries into excluded directories.
360 PH/10 Added the maildirfolder_create_regex option to appendfile.
366 PH/01 The code for finding all the local interface addresses on a FreeBSD
367 system running IPv6 was broken. This may well have applied to all BSD
368 systems, as well as to others that have similar system calls. The broken
369 code found IPv4 interfaces correctly, but gave incorrect values for the
370 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
371 that it would not match correctly against @[] and not recognize the IPv6
374 PH/02 The ipliteral router was not recognizing addresses of the form user@
375 [ipv6:....] because it didn't know about the "ipv6:" prefix.
377 PH/03 Added disable_ipv6.
379 PH/04 Changed $reply_address to use the raw form of the headers instead of the
380 decoded form, because it is most often used to construct To: headers
381 lines in autoreplies, and the decoded form may well be syntactically
382 invalid. However, $reply_address has leading white space removed, and all
383 newlines turned into spaces so that the autoreply transport does not
386 PH/05 If group was specified without a user on a router, and no group or user
387 was specified on a transport, the group from the router was ignored.
389 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
390 for visible compile-time settings that can be used to change these
391 numbers, for those that want even more. Backwards compatibility with old
392 spool files has been maintained. However, going back to a previous Exim
393 release will lost any variables that are in spool files.
395 PH/07 Two small changes when running in the test harness: increase delay when
396 passing a TCP/IP connection to a new process, in case the original
397 process has to generate a bounce, and remove special handling of
398 127.0.0.2 (sic), which is no longer necessary.
400 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
401 be the same on different OS.
403 PH/09 Moved a debug statement in filter processing to avoid a race problem when
406 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
407 whether --show-vars was specified or not
409 JJ/02 exipick: Added support for new ACL variable spool format introduced
412 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
413 syntactically invalid From: or Reply-to: line, and a filter used this to
414 generate an autoreply, and therefore failed to obtain an address for the
415 autoreply, Exim could try to deliver to a non-existent relative file
416 name, causing unrelated and misleading errors. What now happens is that
417 it logs this as a hard delivery error, but does not attempt to create a
420 PH/11 The exinext utility has a -C option for testing purposes, but although
421 the given file was scanned by exinext itself; it wasn't being passed on
422 when Exim was called.
424 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
425 an end-of-file indication when reading a command response.
427 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
428 compiled. In many other places in Exim, IPv6 addresses are always
429 recognized, so I have changed this. It also means that IPv4 domain
430 literals of the form [IPV4:n.n.n.n] are now always recognized.
432 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
433 used if the router is not running as root, for example, when verifying at
434 ACL time, or when using -bh. The debugging output from this situation was
435 non-existent - all you got was a failure to exec. I have made two
438 (a) Failures to set uid/gid, the current directory, or a process leader
439 in a subprocess such as that created by queryprogram now generate
440 suitable debugging ouput when -d is set.
442 (b) The queryprogram router detects when it is not running as root,
443 outputs suitable debugging information if -d is set, and then runs
444 the subprocess without attempting to change uid/gid.
446 PH/15 Minor change to Makefile for building test_host (undocumented testing
449 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
450 additional section of a DNS packet that returns MX or SRV records.
451 Instead, it always explicitly searches for A/AAAA records. This avoids
452 major problems that occur when a DNS server includes only records of one
453 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
454 fixed another bug: if SRV records were looked up and the corresponding
455 address records were *not* found in the additional section, the port
456 values from the SRV records were lost.
458 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
459 using the correct key (the original address) when searching the retry
460 rules in order to find which one to use for generating the retry hint.
462 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
463 adding the default one. Similarly, if it contains a Reply-To: header, the
464 errors_reply_to option, if set, is not used.
466 PH/19 When calculating a retry time, Exim used to measure the "time since
467 failure" by looking at the "first failed" field in the retry record. Now
468 it does not use this if it is later than than the arrival time of the
469 message. Instead it uses the arrival time. This makes for better
470 behaviour in cases where some deliveries succeed, thus re-setting the
471 "first failed" field. An example is a quota failure for a huge message
472 when small messages continue to be delivered. Without this change, the
473 "time since failure" will always be short, possible causing more frequent
474 delivery attempts for the huge message than are intended.
475 [Note: This change was subsequently modified - see PH/04 for 4.62.]
477 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
478 $1, $2, $3) because the numerical variables can be reset during some
479 expansion items (e.g. "match"), thereby losing the authentication data.
481 PH/21 Make -bV show the size of off_t variables so that the test suite can
482 decide whether to run tests for quotas > 2G.
484 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
485 mailbox_size, and mailbox_filecount in the appendfile transport. If a
486 filecount value is greater than 2G or if a quota value is greater than 2G
487 on a system where the size of off_t is not greater than 4, a panic error
490 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
491 never match. The debug and -bh output now contains an explicit error
492 message indicating a malformed IPv4 address or mask.
494 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
495 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
498 PH/25 Do not write to syslog when running in the test harness. The only
499 occasion when this arises is a failure to open the main or panic logs
500 (for which there is an explicit test).
502 PH/26 Added the /no_tell option to "control=freeze".
504 PH/27 If a host name lookup failed very early in a connection, for example, if
505 the IP address matched host_lookup and the reverse lookup yielded a name
506 that did not have a forward lookup, an error message of the form "no IP
507 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
508 could be logged. Now it outputs the IP address instead of "NULL".
510 PH/28 An enabling patch from MH: add new function child_open_exim2() which
511 allows the sender and the authenticated sender to be set when
512 submitting a message from within Exim. Since child_open_exim() is
513 documented for local_scan(), the new function should be too.
515 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
516 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
517 results in an empty string is now treated as unset.
519 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
521 PH/31 Added sender_verify_fail logging option.
523 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
524 needed by RFC 822 but not by RFC 2822 was commented out. I have now
525 tidied the source and removed it altogether.
527 PH/33 When a queue run was abandoned because the load average was too high, a
528 log line was always written; now it is written only if the queue_run log
529 selector is set. In addition, the log line for abandonment now contains
530 information about the queue run such as the pid. This is always present
531 in "start" and "stop" lines but was omitted from the "abandon" line.
533 PH/34 Omit spaces between a header name and the colon in the error message that
534 is given when verify = headers_syntax fails (if there are lots of them,
535 the message gets confusing).
537 PH/35 Change the default for dns_check_names_pattern to allow slashes within
538 names, as there are now some PTR records that contain slashes. This check
539 is only to protect against broken name servers that fall over on strange
540 characters, so the fact that it applies to all lookups doesn't matter.
542 PH/36 Now that the new test suite is complete, we can remove some of the
543 special code in Exim that was needed for the old test suite. For example,
544 sorting DNS records because real resolvers return them in an arbitrary
545 order. The new test suite's fake resolver always returns records in the
548 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
549 bounces) except when queue_only is set, to avoid logging races between
550 the different processes.
552 PH/38 Panic-die if .include specifies a non-absolute path.
554 PH/39 A tweak to the "H" retry rule from its user.
556 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
557 a label. They prevented compilation on older perls.
559 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
560 a warning to be raised on newish perls.
562 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
563 on queue. Changes to match documented behaviour of showing count of
564 messages matching specified criteria.
566 PH/40 Changed the default ident timeout from 30s to 5s.
568 PH/41 Added support for the use of login_cap features, on those BSD systems
569 that have them, for controlling the resources used by pipe deliveries.
571 PH/42 The content-scanning code uses fopen() to create files in which to put
572 message data. Previously it was not paying any attention to the mode of
573 the files. Exim runs with umask(0) because the rest of the code creates
574 files with open(), and sets the required mode explicitly. Thus, these
575 files were ending up world-writeable. This was not a big issue, because,
576 being within the spool directory, they were not world-accessible. I have
577 created a function called modefopen, which takes an additional mode
578 argument. It sets umask(777), creates the file, chmods it to the required
579 mode, then resets the umask. All the relevant calls to fopen() in the
580 content scanning code have been changed to use this function.
582 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
583 to 24 hours. This avoids potential overflow problems when processing G
584 and H retry rules. I suspect nobody ever tinkers with this value.
586 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
588 PH/45 When the plaintext authenticator is running as a client, the server's
589 challenges are checked to ensure they are valid base64 strings. By
590 default, the authentication attempt is cancelled if an invalid string is
591 received. Setting client_ignore_invalid_base64 true ignores these errors.
592 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
593 they are received. Thus, the responses can be made to depend on the
594 challenges. If an invalid string is ignored, an empty string is placed in
597 PH/46 Messages that are created by the autoreply transport now contains a
598 References: header, in accordance with RFCs 2822 and 3834.
600 PH/47 Added authenticated_sender_force to the smtp transport.
602 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
604 PH/49 Installed latest patch from the Sieve maintainer.
606 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
607 was also set, the appendfile transport was unnecessarily scanning a
608 directory of message files (e.g. for maildir delivery) to find the count
609 of files (along with the size), even though it did not need this
610 information. It now does the scan only if it needs to find either the
611 size of the count of files.
613 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
615 PH/52 Two bugs concerned with error handling when the smtp transport is
618 (i) Exim was not creating retry information for temporary errors given
619 for individual recipients after the DATA command when the smtp transport
620 was used in LMTP mode. This meant that they could be retried too
621 frequently, and not timed out correctly.
623 (ii) Exim was setting the flag that allows error details to be returned
624 for LMTP errors on RCPT commands, but not for LMTP errors for individual
625 recipients that were returned after the DATA command.
627 PH/53 This is related to PH/52, but is more general: for any failing address,
628 when detailed error information was permitted to be returned to the
629 sender, but the error was temporary, then after the final timeout, only
630 "retry timeout exceeded" was returned. Now it returns the full error as
631 well as "retry timeout exceeded".
633 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
634 do this, and (what is worse) MTAs that accept it.
636 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
637 will now be deprecated.
639 PH/56 New os.c-cygwin from the Cygwin maintainer.
641 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
642 formats (previously only available in exim formats via -bpr, -bpru,
643 and -bpra. Now also available in native and exiqgrep formats)
645 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
646 with very large, slow to parse queues
648 JJ/08 exipick: added ! as generic prefix to negate any criteria format
650 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
652 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
653 responses to authentication challenges, though it was showing the
654 challenges; (ii) I've removed the CR characters from the debug output for
657 PH/58 Allow for the insertion of a newline as well as a space when a string
658 is turned into more than one encoded-word during RFC 2047 encoding. The
659 Sieve code now uses this.
661 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
662 data_4xx, lost_connection, tls_required.
664 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
665 message was being sent as an SMTP response.
667 PH/61 Add -l and -k options to exicyclog.
669 PH/62 When verifying, if an address was redirected to one new address, so that
670 verification continued, and the new address failed or deferred after
671 having set something in $address_data, the value of $address_data was not
672 passed back to the ACL. This was different to the case when no
673 redirection occurred. The value is now passed back in both cases.
675 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
676 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
677 use login_cap.h, so on its own it isn't the distinguishing feature. The
678 new name refers directly to the setclassresources() function.
680 PH/65 Added configuration files for NetBSD3.
682 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
684 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
685 is preferred over IPv4.
687 PH/68 The bounce_return_message and bounce_return_body options were not being
688 honoured for bounces generated during the reception of non-SMTP messages.
689 In particular, this applied to messages rejected by the ACL. This bug has
690 been fixed. However, if bounce_return_message is true and bounce_return_
691 body is false, the headers that are returned for a non-SMTP message
692 include only those that have been read before the error was detected.
693 (In the case of an ACL rejection, they have all been read.)
695 PH/69 The HTML version of the specification is now built in a directory called
696 spec_html instead of spec.html, because the latter looks like a path with
697 a MIME-type, and this confuses some software.
699 PH/70 Catch two compiler warnings in sieve.c.
701 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
702 function verify_get_ident() calls ip_connect() to connect a socket, but
703 if the "connect()" function timed out, ip_connect() used to close the
704 socket. However, verify_get_ident() also closes the socket later, and in
705 between Exim writes to the log, which may get opened at this point. When
706 the socket was closed in ip_connect(), the log could get the same file
707 descriptor number as the socket. This naturally causes chaos. The fix is
708 not to close the socket in ip_connect(); the socket should be closed by
709 the function that creates it. There was only one place in the code where
710 this was missing, in the iplookup router, which I don't think anybody now
711 uses, but I've fixed it anyway.
713 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
714 well as to direct DNS lookups. Otherwise the handling of names in host
715 lists is inconsistent and therefore confusing.
721 PH/01 Two changes to the default runtime configuration:
723 (1) Move the checks for relay_from_hosts and authenticated clients from
724 after to before the (commented out) DNS black list checks.
726 (2) Add control=submission to the relay_from_hosts and authenticated
727 clients checks, on the grounds that messages accepted by these
728 statements are most likely to be submissions.
730 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
732 (1) Generate an error if the third argument for the ${prvs expansion is
735 (2) Treat a missing third argument of ${prvscheck as if it were an empty
738 (3) Reset the variables that are obtained from the first argument of
739 ${prvscheck and used in the second argument before leaving the code,
740 because their memory is reclaimed, so using them afterwards may do
743 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
744 one (it's much easier than Tom thought :-).
746 (5) Because of (4), we can now allow for the use of $prvscheck_result
747 inside the third argument.
749 PH/03 For some reason, the default setting of PATH when running a command from
750 a pipe transport was just "/usr/bin". I have changed it to
753 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
754 anything to be listed in the output from -bV.
756 PH/05 When a filter generated an autoreply, the entire To: header line was
757 quoted in the delivery log line, like this:
759 => >A.N.Other <ano@some.domain> <original@ddress> ...
761 This has been changed so that it extracts the operative address. There
762 may be more than one such address. If so, they are comma-separated, like
765 => >ano@some.domain,ona@other.domain <original@ddress> ...
767 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
768 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
769 not being looked up in the rDNS to get a host name, Exim was showing the
770 IP address twice in Received: lines, even though the IP addresses were
771 identical. For example:
773 Received: from [1.2.3.4] (helo=[1.2.3.4])
775 However, if the real host name was known, it was omitting the HELO data
776 if it matched the actual IP address. This has been tidied up so that it
777 doesn't show the same IP address twice.
779 PH/07 When both +timestamp and +memory debugging was on, the value given by
780 $tod_xxx expansions could be wrong, because the tod_stamp() function was
781 called by the debug printing, thereby overwriting the timestamp buffer.
782 Debugging no longer uses the tod_stamp() function when +timestamp is set.
784 PH/08 When the original message was included in an autoreply transport, it
785 always said "this is a copy of the message, including all the headers",
786 even if body_only or headers_only was set. It now gives an appropriate
789 PH/09 Applied a patch from the Sieve maintainer which:
791 o fixes some comments
792 o adds the (disabled) notify extension core
793 o adds some debug output for the result of if/elsif tests
794 o points to the current vacation draft in the documentation
795 and documents the missing references header update
799 o fixes a bug in processing the envelope test (when testing
800 multiple envelope elements, the last element determinted the
803 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
804 Electronic Mail") by including:
806 Auto-submitted: auto-generated
808 in the messages that it generates (bounce messages and others, such as
809 warnings). In the case of bounce messages for non-SMTP mesages, there was
810 also a typo: it was using "Auto_submitted" (underscore instead of
811 hyphen). Since every message generated by Exim is necessarily in response
812 to another message, thes have all been changed to:
814 Auto-Submitted: auto-replied
816 in accordance with these statements in the RFC:
818 The auto-replied keyword:
820 - SHOULD be used on messages sent in direct response to another
821 message by an automatic process,
823 - MUST NOT be used on manually-generated messages,
825 - MAY be used on Delivery Status Notifications (DSNs) and Message
826 Disposition Notifications (MDNs),
828 - MUST NOT be used on messages generated by automatic or periodic
829 processes, except for messages which are automatic responses to
832 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
833 to the default Received: header definition.
835 PH/12 Added log selector acl_warn_skipped (default on).
837 PH/13 After a successful wildlsearch lookup, discard the values of numeric
838 variables because (a) they are in the wrong storage pool and (b) even if
839 they were copied, it wouldn't work properly because of the caching.
841 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
842 checking when decoding. Apparently there are clients that generate
843 overlong encoded strings. Why am I not surprised?
845 PH/15 If the first argument of "${if match_address" was not empty, but did not
846 contain an "@" character, Exim crashed. Now it writes a panic log message
847 and treats the condition as false.
849 PH/16 In autoreply, treat an empty string for "once" the same as unset.
851 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
852 extension "envelope-auth". The code is finished and in agreement with
853 other implementations, but there is no documentation so far and in fact,
854 nobody wrote the draft yet. This extension is currently #undef'ed, thus
855 not changing the active code.
857 Print executed "if" and "elsif" statements when debugging is used. This
858 helps a great deal to understand what a filter does.
860 Document more things not specified clearly in RFC3028. I had all this
861 sorted out, when out of a sudden new issues came to my mind. Oops."
863 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
866 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
867 canonical form (as documented). However, after a host name lookup from
868 the IP address, check_host() was doing a simple string comparison with
869 addresses acquired from the DNS when checking that the found name did
870 have the original IP as one of its addresses. Since any found IPv6
871 addresses are likely to be in abbreviated form, the comparison could
872 fail. Luckily, there already exists a function for doing the comparison
873 by converting both addresses to binary, so now that is used instead of
876 PH/20 There was another similar case to PH/19, when a complete host name was
877 given in a host list; looking up its IP address could give an abbreviated
878 form, whereas the current host's name might or might not be abbreviated.
879 The same fix has been applied.
885 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
886 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
889 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
890 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
892 PH/03 Typo: missing ".o" in src/pcre/Makefile.
894 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
895 header line, restrict the check to what is listed in RFCs 2369 and 2929.
896 Also, for "Auto-Submitted", treat anything other than "no" as
897 non-personal, in accordance with RFC 3834. (Previously it treated
898 anything starting "auto-" as non-personal.)
900 TF/01 The control=submission/name=... option had a problem with syntax
901 errors if the name included a slash character. The /name= option
902 now slurps the rest of the string, so it can include any characters
903 but it must come last in the list of options (after /sender_retain
906 PH/05 Some modifications to the interface to the fake nameserver for the new
914 TK/01 Added the "success_on_redirect" address verification option. See
915 NewStuff for rationale and an example.
917 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
919 PH/02 Patch to exigrep to allow it to work on syslog lines.
921 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
922 fread() to skip over the body file's header line, because in Cygwin the
923 header line is locked and is inaccessible.
925 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
926 co-exist for some time) to make it clear that it is the Exim ID that is
927 referenced, not the Message-ID: header line.
929 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
930 string_format() function, because snprintf() does not exist on all
933 PH/06 The use of forbid_filter_existstest now also locks out the use of the
934 ${stat: expansion item.
936 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
937 protocol synchronization error", to keep the pedants happy.
939 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
940 well as for IRIX systems, when gcc is being used. See the host.c source
943 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
945 PH/10 Named domain lists were not working if used in a queue_smtp_domains
948 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
949 transport and to the smtp transport in LMTP mode.
951 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
953 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
954 run a filter in a subprocess. This could lead to confusion in subsequent
955 lookups in the parent process. There should also be a search_tidyup() at
956 the end of the subprocess.
958 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
959 only if the host matched helo_try_verify_hosts, which caused the
960 verification to occur when the EHLO/HELO command was issued. The ACL just
961 tested the remembered result. Now, if a previous verification attempt has
962 not happened, "verify = helo" does it there and then.
964 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
966 TK/03 Fix log output including CR from clamd.
968 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
969 exist provoked a memory error which could cause a segfault.
971 PH/15 Installed PCRE 6.2
973 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
975 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
976 of the problem. Specifically, suggested +O2 rather than +O1 for the
979 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
981 PH/20 If a delivery was routed to a non-standard port by means of an SRV
982 record, the port was not correctly logged when the outgoing_port log
983 selector was set (it logged the transort's default port).
985 PH/21 Added support for host-specific ports to manualroute, queryprogram,
986 fallback_hosts, and "hosts" in the smtp transport.
988 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
989 host errors such as "Connection refused".
991 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
992 authentication with radiusclient 0.4.9:
994 - Error returned from rc_read_config was caught wrongly
995 - Username/password not passed on to radius server due to wrong length.
997 The presumption is that some radiusclient API changes for 4.51/PH/17
998 were not taken care of correctly. The code is still untested by me (my
999 Linux distribution still has 0.3.2 of radiusclient), but it was
1000 contributed by a Radius user.
1002 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1003 expanding the "port" option of the smtp transport.
1005 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1006 while reading a MIME header. Thanks to Tom Hughes for a patch.
1008 PH/24 Include config.h inside local_scan.h so that configuration settings are
1011 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1012 that in an ACL for RCPT (for example), you can examine exactly what was
1015 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1016 commands, but it was not correctly comparing the address with the actual
1017 client host address. Thus, it would show the EHLO address in Received:
1018 header lines when this was not necessary.
1020 PH/27 Added the % operator to ${eval:}.
1022 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1023 it should be ignoring failures (because with -C, for example, it has lost
1024 privilege). It wasn't ignoring creation failures other than "already
1027 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1030 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1031 because an input line was too long, either on its own, or by virtue of
1032 too many continuations, the temporary file was not being removed, and the
1033 return code was incorrect.
1035 PH/31 Missing "BOOL" in function definition in filtertest.c.
1037 PH/32 Applied Sieve patches from the maintainer.
1039 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1041 PH/33 Added "verify = not_blind".
1043 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1044 Local/Makefile (with some defaults set). These are used in built scripts
1045 such as exicyclog, but they have never been used in the exim_install
1046 script (though there are many overriding facilities there). I have
1047 arranged that the exim_install script now takes note of these two
1050 PH/35 Installed configuration files for Dragonfly.
1052 PH/36 When a locally submitted message by a trusted user did not contain a
1053 From: header, and the sender address was obtained from -f or from an SMTP
1054 MAIL command, and the trusted user did not use -F to supply a sender
1055 name, $originator_name was incorrectly used when constructing a From:
1056 header. Furthermore, $originator_name was used for submission mode
1057 messages from external hosts without From: headers in a similar way,
1058 which is clearly wrong.
1060 PH/37 Added control=suppress_local_fixups.
1062 PH/38 When log_selector = +received_sender was set, and the addition of the
1063 sender made the log line's construction buffer exactly full, or one byte
1064 less than full, an overflow happened when the terminating "\n" was
1067 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1068 when the result of a list match is failure because a DNS lookup failed.
1070 PH/40 RM_COMMAND is now used in the building process.
1072 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1073 the "build-* directories that it finds.
1075 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1076 address in a domain literal was a prefix of an interface address.
1078 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1079 when verifying a sender address, unless rewrite_headers is false.
1081 PH/44 Wrote a long comment about why errors_to addresses are verified as
1082 recipients, not senders.
1084 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1085 the ratelimit ACL was added.
1087 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1089 PH/46 Added extra information about PostgreSQL errors to the error string.
1091 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1092 avoiding the need to install special zones in a real server. This is
1093 backwards compatible; if it can't find the fake resolver, it drops back.
1094 Thus, both old and new test suites can be run.
1096 TF/02 Added util/ratelimit.pl
1098 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1099 clock is set back in time.
1101 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1102 Candler <B.Candler@pobox.com>.
1104 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1105 OK for addresses that are the result of redirections.
1107 PH/48 A number of further additions for the benefit of the new test suite,
1108 including a fake gethostbyname() that interfaces to the fake DNS resolver
1111 TF/06 The fix for widen_domains has also been applied to qualify_single and
1112 search_parents which are the other dnslookup options that can cause
1115 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1118 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1120 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1121 service" if it encountered an MX record with an empty target hostname.
1122 The message is now "an MX or SRV record indicated no SMTP service".
1124 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1125 if the DNS of the sending site is misconfigured. This is quite a
1126 common situation. This change restores the behaviour of treating a
1127 helo verification defer as a failure.
1129 PH/51 If self=fail was set on a router, the bounce message did not include the
1130 actual error message.
1136 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1138 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1139 command itself ended in error, the underlying message about the transport
1140 filter timeout was being overwritten with the pipe command error. Now the
1141 underlying error message should be appended to the second error message.
1143 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1145 PH/02 Reduce the amount of output that "make" produces by default. Full output
1146 can still be requested.
1148 PH/03 The warning log line about a condition test deferring for a "warn" verb
1149 was being output only once per connection, rather than after each
1150 occurrence (because it was using the same function as for successful
1151 "warn" verbs). This seems wrong, so I have changed it.
1153 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1154 it should not have, which might have caused a crash in the right
1155 circumstances, but probably never did.
1157 PH/04 Installed a modified version of Tony Finch's patch to make submission
1158 mode fix the return path as well as the Sender: header line, and to
1159 add a /name= option so that you can make the user's friendly name appear
1162 TF/03 Added the control = fakedefer ACL modifier.
1164 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1165 Mark Lowes for thorough testing.
1167 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1169 TK/03 Merged latest SRS patch from Miles Wilton.
1171 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1172 with the definition in sysexits.h (which is #included earlier).
1173 Fortunately, Exim does not actually use EX_OK. The code used to try to
1174 preserve the sysexits.h value, by assumimg that macro definitions were
1175 scanned for macro replacements. I have been disabused of this notion,
1176 so now the code just undefines EX_OK before #including unistd.h.
1178 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1179 in the smtp transport. When a block could not be written in a single
1180 write() function, the timeout was being re-applied to each part-write.
1181 This seems wrong - if the receiver was accepting one byte at a time it
1182 would take for ever. The timeout is now adjusted when this happens. It
1183 doesn't have to be particularly precise.
1185 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1186 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1188 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1189 without a domain if the check to <postmaster@domain> fails.
1191 SC/01 Eximstats: added -xls and the ability to specify output files
1192 (patch written by Frank Heydlauf).
1194 SC/02 Eximstats: use FileHandles for outputing results.
1196 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1198 SC/04 Eximstats: fixed display of large numbers with -nvr option
1200 SC/05 Eximstats: fixed merging of reports with empty tables.
1202 SC/06 Eximstats: added the -include_original_destination flag
1204 SC/07 Eximstats: removed tabs and trailing whitespace.
1206 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1208 TK/06 MBOX spool code: Add real "From " MBOX separator line
1209 so the .eml file is really in mbox format (even though
1210 most programs do not really care). Patch from Alex Miller.
1212 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1213 The latter is generated from $received_to and is only set if the
1214 message has one envelope recipient. SA can use these headers,
1215 obviously out-of-the-box. Patch from Alex Miller.
1217 PH/08 The ${def test on a variable was returning false if the variable's
1218 value was "0", contrary to what the specification has always said!
1219 The result should be true unless the variable is empty.
1221 PH/09 The syntax error of a character other than { following "${if
1222 def:variable_name" (after optional whitespace) was not being diagnosed.
1223 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1224 accidental colon was present, for example, could give incorrect results.
1226 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1227 result is used (not including appendfile, where other changes are about
1230 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1231 This involved changing a lot of size variables from int to off_t. It
1232 should work with maildirs and everything.
1234 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1235 spamd dying while we are connected to it.
1237 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1240 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1241 function for BDB 4.3.
1243 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1245 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1246 the sources for PCRE have been re-arranged and more files are now
1249 PH/15 The code I had for printing potentially long long variables in PH/11
1250 above was not the best (it lost precision). The length of off_t variables
1251 is now inspected at build time, and an appropriate printing format (%ld
1252 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1253 to be "long long int" or "long int". This is needed for the internal
1254 formatting function string_vformat().
1256 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1257 the configuration file to be ":syslog", then the script "guesses" where
1258 the logs files are, rather than using the compiled in default. In our
1259 case the guess is not the same as the compiled default, so the script
1260 suddenly stopped working when I started to use syslog. The patch checks
1261 to see if log_file_path is "". If so, it attempts to read it from exim
1262 with no configuration file to get the compiled in version, before it
1263 falls back to the previous guessing code."
1265 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1266 implementing BATV in an Exim configuration. See NewStuff for the gory
1269 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1270 Makefile that are specific to HP-UX.
1272 PH/18 If the "use_postmaster" option was set for a recipient callout together
1273 with the "random" option, the postmaster address was used as the MAIL
1274 FROM address for the random test, but not for the subsequent recipient
1275 test. It is now used for both.
1277 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1278 patch removes a few documentation additions to RFC 3028, because the
1279 latest draft now contains them. It adds the new en;ascii-case comparator
1280 and a new error check for 8bit text in MIME parts. Comparator and
1281 require names are now matched exactly. I enabled the subaddress
1282 extension, but it is not well tested yet (read: it works for me)."
1284 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1285 rework some of the code of TK/09 above to avoid the hardwired use of
1286 "%lld" and "long long". Replaced the call to snprintf() with a call to
1289 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1290 records point to non-existent hosts", "retry timeout exceeded", and
1291 "retry time not reached for any host after a long failure period".
1293 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1294 experimental DomainKeys support:
1296 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1297 (2) On an error such as an illegally used "control", the wrong name for
1298 the control was given.
1300 These problems did NOT occur unless DomainKeys support was compiled.
1302 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1304 PH/24 Added ${if match_ip condition.
1306 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1307 databases so that it will be absolutely obvious if a crash occurs in the
1308 DB library. This is a regular occurrence (often caused by mis-matched
1311 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1312 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1313 picked up on a user's system that detects such things. There doesn't seem
1314 to be a gcc warning option for this - only an attribute that has to be
1315 put on the function's prototype. It seems that in Fedora Core 4 they have
1316 set this on a number of new functions. No doubt there will be more in due
1319 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1320 specify a transport. However, if an address that was verified by such a
1321 router was the subject of a callout, Exim crashed because it tried to
1322 read the rcpt_include_affixes from the non-existent transport. Now it
1323 just assumes that the setting of that option is false. This bug was
1324 introduced by 4.51/PH/31.
1326 PH/28 Changed -d+all to exclude +memory, because that information is very
1327 rarely of interest, but it makes the output a lot bigger. People tend to
1328 do -d+all out of habit.
1330 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1331 code in os-type was giving problems when libc.so lives in lib64, like on
1334 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1335 aren't the modern standard, and it seems that some systems' include files
1336 don't always have them. Exim was already checking for some of the newer
1337 ones like T_AAAA, and defining it itself. I've added checks for all the
1338 record types that Exim uses.
1340 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1341 not automatically generating a new one, as it is supposed to. This
1342 prevented TLS from working. If the file did exist, but contained invalid
1343 data, a new version was generated, as expected. It was only the case of a
1344 non-existent file that was broken.
1346 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1347 with a change in libdomainkeys > 0.64.
1349 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1350 from DNS. If the selector record carries the flag, it now has
1351 precedence over the domain-wide flag.
1353 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1355 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1356 the use of a port name that isn't defined in /etc/services) occurred, the
1357 message was deferred as in a normal delivery, and thus remained on the
1358 spool, instead of being failed because of the mua_wrapper setting. This
1359 is now fixed, and I tidied up some of the mua_wrapper messages at the
1362 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1363 the messages in a hash of arrays rather than using individual hashes.
1364 This is a bit cleaner and results in dramatic memory savings, albeit
1365 at a slight CPU cost.
1367 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1368 as requested by Marc Sherman.
1370 SC/10 Eximstats: added histograms for user specified patterns as requested
1373 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1375 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1376 fopen() in the content-scanning modules that did not already have it.
1382 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1383 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1385 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1387 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1389 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1390 to test Sieve filters that use "vacation".
1392 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1393 that changes the way the GnuTLS parameters are stored in the cache file.
1394 The new format can be generated externally. For backward compatibility,
1395 if the data in the cache doesn't make sense, Exim assumes it has read an
1396 old-format file, and it generates new data and writes a new file. This
1397 means that you can't go back to an older release without removing the
1400 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1401 work if there are any delivery delays because "one_time" forces the
1402 parent to be marked "delivered", so its unseen clone is never tried
1403 again. For this reason, Exim now forbids the simultaneous setting of
1406 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1407 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1408 entry if you want to know the details. The fix, however, neglected to
1409 consider the case when local delivery batching is involved. The test for
1410 "previously delivered" was not happening when checking to see if an
1411 address could be batched with a previous (undelivered) one; under
1412 certain circumstances this could lead to multiple deliveries to the same
1415 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1416 in its include files, and this causes problems building Exim.
1418 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1419 header_syntax) but Exim was just ignoring anything given after a slash.
1420 In particular, this caused confusion with an attempt to use "verify =
1421 reverse_host_lookup/defer_ok". An error is now given when options are
1422 supplied for verify items that do not have them. (Maybe reverse_host_
1423 lookup should have a defer_ok option, but that's a different point.)
1425 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1426 defined by RFC 821) to 2048, because there were problems with some AUTH
1427 commands, and RFC 1869 says the size should be increased for extended
1428 SMTP commands that take arguments.
1430 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1433 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1434 "unknown" error; now it says that the functionality isn't in the binary.
1436 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1437 an address' error message when a string expansion fails (syntax or
1438 whatever). Otherwise the password may appear in the log. Following change
1439 PH/42 below, there is no longer a chance of it appearing in a bounce
1442 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1444 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1445 of its addresses was ever tried. (Bugzilla bug #2.)
1447 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1448 the result incorrectly in the debug output. (It correctly added a newline
1449 to what was transported.)
1451 TF/01 Added $received_time.
1453 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1454 commented out examples of how to interface to a virus scanner and to
1455 SpamAssassin. Also added commented examples of av_scanner and
1456 spamd_address settings.
1458 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1459 and controls are allowed in which ACLs. There were a couple of minor
1460 errors. Some of the entries in the conditions table (which is a table of
1461 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1462 negation of where the condition IS allowed.
1464 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1466 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1467 header file does not have a version number, so I've had to invent a new
1468 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1469 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1470 radiusclient), but it was contributed by a Radius user.
1472 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1473 files or directories, for OpenSSL.
1475 PH/19 When an Exim process that is running as root has to create an Exim log
1476 file, it does so in a subprocess that runs as exim:exim so as to get the
1477 ownership right at creation (otherwise, other Exim processes might see
1478 the file with the wrong ownership). There was no test for failure of this
1479 fork() call, which would lead to the process getting stuck as it waited
1480 for a non-existent subprocess. Forks do occasionally fail when resources
1481 run out. I reviewed all the other calls to fork(); they all seem to check
1484 PH/20 When checking for unexpected SMTP input at connect time (before writing
1485 the banner), Exim was not dealing correctly with a non-positive return
1486 from the read() function. If the client had disconnected by this time,
1487 the result was a log entry for a synchronization error with an empty
1488 string after "input=" when read() returned zero. If read() returned -1
1489 (an event I could not check), uninitialized data bytes were printed.
1490 There were reports of junk text (parts of files, etc) appearing after
1493 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1495 PH/22 Added support for macro redefinition, and (re)definition in between
1496 driver and ACL definitions.
1498 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1499 forgetting to use the resulting value; it was using the unexpanded value.
1501 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1502 hadn't been configured. The fix is from Juergen Kreileder, who
1503 understands it better than I do:
1505 "Here's what I see happening with three configured cyrus_sasl
1506 authenticators configured (plain, login, cram-md5):
1508 On startup auth_cyrus_sasl_init() gets called for each of these.
1509 This means three calls to sasl_listmech() without a specified mech_list.
1510 => SASL tests which mechs of all available mechs actually work
1511 => three warnings about OTP not working
1512 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1514 With the patch, sasl_listmech() also gets called three times. But now
1515 SASL's mech_list option is set to the server_mech specified in the the
1516 authenticator. Or in other words, the answer from sasl_listmech()
1517 gets limited to just the mech you're testing for (which is different
1519 => the return list contains just 'plain' or 'login', 'cram-md5' or
1520 nothing depending on the value of ob->server_mech.
1522 I've just tested the patch: Authentication still works fine,
1523 unavailable mechs specified in the exim configuration are still
1524 caught, and the auth.log warnings about OTP are gone."
1526 PH/25 When debugging is enabled, the contents of the command line are added
1527 to the debugging output, even when log_selector=+arguments is not
1530 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1531 answer is "GNU", and only if the return is "GNU/something" is the answer
1534 PH/27 $acl_verify_message is now set immediately after the failure of a
1535 verification in an ACL, and so is available in subsequent modifiers. In
1536 particular, the message can be preserved by coding like this:
1538 warn !verify = sender
1539 set acl_m0 = $acl_verify_message
1541 Previously, $acl_verify_message was set only while expanding "message"
1542 and "log_message" when a very denied access.
1544 PH/28 Modified OS/os.c-Linux with
1546 -#ifndef OS_LOAD_AVERAGE
1547 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1549 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1550 nomenclature these days.)
1552 PH/29 Installed patch from the Sieve maintainer that adds the options
1553 sieve_useraddress and sieve_subaddress to the redirect router.
1555 PH/30 In these circumstances:
1556 . Two addresses routed to the same list of hosts;
1557 . First host does not offer TLS;
1558 . First host accepts first address;
1559 . First host gives temporary error to second address;
1560 . Second host offers TLS and a TLS session is established;
1561 . Second host accepts second address.
1562 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1563 and peerdn, if requested) that were in fact used only for the second
1566 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1567 attention to any local part prefix or suffix that was matched by the
1568 router that accepted the address. It now behaves in the same way as it
1569 does for delivery: the affixes are removed from the local part unless
1570 rcpt_include_affixes is set on the transport.
1572 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1573 timeout during the DATA phase of an incoming message.
1575 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1576 applied a patch sanctioned by the Sieve maintainer.
1578 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1579 the uid or gid is negative. A case of a negative gid caused this to be
1580 noticed. The fix allows for either to be negative.
1582 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1583 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1586 PH/36 The change PH/12 above was broken. Fixed it.
1588 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1589 the grounds that routing the same address twice would always produce the
1590 same answer. This might have been true once, but it is certainly no
1591 longer true now. Routing a child address may depend on the previous
1592 routing that produced that child. Some complicated redirection strategies
1593 went wrong when messages had multiple recipients, and made Exim's
1594 behaviour dependent on the order in which the addresses were given.
1596 I have moved the duplicate checking until after the routing is complete.
1597 Exim scans the addresses that are assigned to local and remote
1598 transports, and removes any duplicates. This means that more work will be
1599 done, as duplicates will always all be routed, but duplicates are
1600 presumably rare, so I don't expect this is of any significance.
1602 For deliveries to pipes, files, and autoreplies, the duplicate checking
1603 still happens during the routing process, since they are not going to be
1606 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1607 It corrects a timeout issue with spamd. This is Ian's comment: "The
1608 background is that sometimes spamd either never reads data from a
1609 connection it has accepted, or it never writes response data. The exiscan
1610 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1611 blindly assumes that writes won't block so it may never time out."
1613 PH/39 Allow G after quota size as well as K and M.
1615 PH/40 The value set for $authenticated_id in an authenticator may not contain
1616 binary zeroes or newlines because the value is written to log lines and
1617 to spool files. There was no check on this. Now the value is run through
1618 the string_printing() function so that such characters are converted to
1619 printable escape sequences.
1621 PH/41 $message_linecount is a new variable that contains the total number of
1622 lines in the message. Compare $body_linecount, which is the count for the
1625 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1626 bounce and delay warning messages, except in certain special cases, which
1629 (a) An SMTP error message from a remote host;
1630 (b) A message specified in a :fail: redirection;
1631 (c) A message specified in a "fail" command in a system filter;
1632 (d) A message specified in a FAIL return from the queryprogram router;
1633 (e) A message specified by the cannot_route_message router option.
1635 In these cases only, Exim does include the error details in bounce and
1636 warning messages. There are also a few cases where bland messages such
1637 as "unrouteable address" or "local delivery error" are given.
1639 PH/43 $value is now also set for the "else" part of a ${run expansion.
1641 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1642 being worked on, but at least Exim now implements the latest version to
1645 PH/45 In a pipe transport, although a timeout while waiting for the pipe
1646 process to complete was treated as a delivery failure, a timeout while
1647 writing the message to the pipe was logged, but erroneously treated as a
1648 successful delivery. Such timeouts include transport filter timeouts. For
1649 consistency with the overall process timeout, these timeouts are now
1650 treated as errors, giving rise to delivery failures by default. However,
1651 there is now a new Boolean option for the pipe transport called
1652 timeout_defer, which, if set TRUE, converts the failures into defers for
1653 both kinds of timeout. A transport filter timeout is now identified in
1656 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1657 systems where "make" and "gmake" are different, calling "gmake" at top
1658 level broke things. I've arranged for the value of $(MAKE) to be passed
1659 from the Makefile to this script so that it can call the same version of
1663 A note about Exim versions 4.44 and 4.50
1664 ----------------------------------------
1666 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1667 changes of various kinds. As a consequence, a big documentation update was
1668 needed. This delayed the release for rather longer than seemed good, especially
1669 in the light of a couple of (minor) security issues. Therefore, the changes
1670 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1671 release. So 4.44 and 4.50 are in effect two different branches that both start
1674 I have left the 4.50 change log unchanged; it contains all the changes since
1675 4.43. The change log for 4.44 is below; many of its items are identical to
1676 those for 4.50. This seems to be the most sensible way to preserve the
1677 historical information.
1683 1. Minor wording change to the doc/README.SIEVE file.
1685 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1686 computation of the current number of files was incorrect.
1688 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1689 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1692 4. Give more explanation in the error message when the command for a transport
1693 filter fails to execute.
1695 5. There are several places where Exim runs a non-Exim command in a
1696 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1697 was being done only for the command run by the queryprogram router. It is
1698 now done for all such subprocesses. The other cases are: ${run, transport
1699 filters, and the commands run by the lmtp and pipe transports.
1701 6. Added CONFIGURE_GROUP build-time option.
1703 7. Some older OS have a limit of 256 on the maximum number of file
1704 descriptors. Exim was using setrlimit() to set 1000 as a large value
1705 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1706 systems. I've change it so that if it can't get 1000, it tries for 256.
1708 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1709 was an oversight, and furthermore, ever since the addition of extra
1710 controls (e.g. 4.43/32), the checks on when to allow different forms of
1711 "control" were broken. There should now be diagnostics for all cases when a
1712 control that does not make sense is encountered.
1714 9. Added the /retain_sender option to "control=submission".
1716 10. $recipients is now available in the predata ACL (oversight).
1718 11. Tidy the search cache before the fork to do a delivery from a message
1719 received from the command line. Otherwise the child will trigger a lookup
1720 failure and thereby defer the delivery if it tries to use (for example) a
1721 cached ldap connection that the parent has called unbind on.
1723 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1724 of $address_data from the recipient verification was clobbered by the
1725 sender verification.
1727 13. The value of address_data from a sender verification is now available in
1728 $sender_address_data in subsequent conditions in the ACL statement.
1730 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1732 15. Added a new option "connect=<time>" to callout options, to set a different
1735 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1736 was its contents. (It was OK if the option was not defined at all.)
1738 17. A "Completed" log line is now written for messages that are removed from
1739 the spool by the -Mrm option.
1741 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1742 information about exactly what failed.
1744 19. Added -dd to debug only the daemon process.
1746 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1747 handles timeouts, both on the server side and network timeouts. Renamed the
1748 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1750 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1751 It is now set to "smtps".
1753 22. $host_address is now set to the target address during the checking of
1754 ignore_target_hosts.
1756 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1757 being passed; this would have caused $sender_host_name to have been used if
1758 matching the list had actually called for a host name (not very likely,
1759 since this list is usually IP addresses). A host name is now passed as
1762 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1763 code that specifies a non-restarting handler (typically sigaction() in
1764 modern systems) in an attempt to fix a rare and obscure crash bug.
1766 25. Narrowed the window for a race in the daemon that could cause it to ignore
1767 SIGCHLD signals. This is not a major problem, because they are used only to
1768 wake it up if nothing else does.
1770 26. A malformed maildirsize file could cause Exim to calculate negative values
1771 for the mailbox size or file count. Odd effects could occur as a result.
1772 The maildirsize information is now recalculated if the size or filecount
1775 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1776 support for a long time. Removed HAVE_SYS_VFS_H.
1778 28. Installed the latest version of exipick from John Jetmore.
1780 29. In an address list, if the pattern was not a regular expression, an empty
1781 subject address (from a bounce message) matched only if the pattern was an
1782 empty string. Non-empty patterns were not even tested. This was the wrong
1783 because it is perfectly reasonable to use an empty address as part of a
1784 database query. An empty address is now tested by patterns that are
1785 lookups. However, all the other forms of pattern expect the subject to
1786 contain a local part and a domain, and therefore, for them, an empty
1787 address still always fails if the pattern is not itself empty.
1789 30. Exim went into a mad DNS loop when attempting to do a callout where the
1790 host was specified on an smtp transport, and looking it up yielded more
1791 than one IP address.
1793 31. Re-factored the code for checking spool and log partition space into a
1794 function that finds that data and another that does the check. The former
1795 is then used to implement four new variables: $spool_space, $log_space,
1796 $spool_inodes, and $log_inodes.
1798 32. The RFC2047 encoding function was originally intended for short strings
1799 such as real names; it was not keeping to the 75-character limit for
1800 encoded words that the RFC imposes. It now respects the limit, and
1801 generates multiple encoded words if necessary. To be on the safe side, I
1802 have increased the buffer size for the ${rfc2047: expansion operator from
1805 33. It is now permitted to omit both strings after an "if" condition; if the
1806 condition is true, the result is "true". As before, when the second string
1807 is omitted, a false condition yields an empty string. This makes it less
1808 cumbersome to write custom ACL and router conditions.
1810 34. Failure to deliver a bounce message always caused it to be frozen, even if
1811 there was an errors_to setting on the router. The errors_to setting is now
1814 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1815 canonical form (fully expanded) before being placed in
1816 $sender_host_address.
1818 36. The table in the code that translates DNS record types into text (T_A to
1819 "A" for instance) was missing entries for NS and CNAME. It is just possible
1820 that this could have caused confusion if both these types were looked up
1821 for the same domain, because the text type is used as part of Exim's
1822 per-process caching. But the chance of anyone hitting this buglet seems
1825 37. The dnsdb lookup has been extended in a number of ways.
1827 (1) There is a new type, "zns", which walks up the domain tree until it
1828 finds some nameserver records. It should be used with care.
1830 (2) There is a new type, "mxh", which is like "mx" except that it returns
1831 just the host names, not the priorities.
1833 (3) It is now possible to give a list of domains (or IP addresses) to be
1834 looked up. The behaviour when one of the lookups defers can be
1835 controlled by a keyword.
1837 (4) It is now possible to specify the separator character for use when
1838 multiple records are returned.
1840 38. The dnslists ACL condition has been extended: it is now possible to supply
1841 a list of IP addresses and/or domains to be looked up in a particular DNS
1844 39. Added log_selector=+queue_time_overall.
1846 40. When running the queue in the test harness, wait just a tad after forking a
1847 delivery process, to get repeatability of debugging output.
1849 41. Include certificate and key file names in error message when GnuTLS fails
1850 to set them up, because the GnuTLS error message doesn't include the name
1851 of the failing file when there is a problem reading it.
1853 42. Allow both -bf and -bF in the same test run.
1855 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1857 44. The "Exiscan patch" is now merged into the mainline Exim source.
1859 45. Sometimes the final signoff response after QUIT could fail to get
1860 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1861 < 0 before doing a fflush(). This bug looks as though it goes back to the
1862 introduction of TLS in release 3.20, but "sometimes" must have been rare
1863 because the tests only now provoked it.
1865 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1866 (this can affect the format of dates).
1868 47. exim_tidydb, when checking for the continued existence of a message for
1869 which it has found a message-specific retry record, was not finding
1870 messages that were in split spool directories. Consequently, it was
1871 deleting retry records that should have stayed in existence.
1873 48. Steve fixed some bugs in eximstats.
1875 49. The SPA authentication driver was not abandoning authentication and moving
1876 on to the next authenticator when an expansion was forced to fail,
1877 contradicting the general specification for all authenticators. Instead it
1878 was generating a temporary error. It now behaves as specified.
1880 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1881 (the order specifies the preference for clients). The order is now AES256,
1882 AES128, 3DES, ARCFOUR128.
1884 51. Small patch to Sieve code - explicitly set From: when generating an
1887 52. Exim crashed if a remote delivery caused a very long error message to be
1888 recorded - for instance if somebody sent an entire SpamAssassin report back
1889 as a large number of 550 error lines. This bug was coincidentally fixed by
1890 increasing the size of one of Exim's internal buffers (big_buffer) that
1891 happened as part of the Exiscan merge. However, to be on the safe side, I
1892 have made the code more robust (and fixed the comments that describe what
1895 53. Now that there can be additional text after "Completed" in log lines (if
1896 the queue_time_overall log selector is set), a one-byte patch to exigrep
1897 was needed to allow it to recognize "Completed" as not the last thing in
1900 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1901 patch that reportedly fixes this has been added. I am not expert enough to
1902 create a test for it. This is what the patch creator wrote:
1904 "I found a little strange behaviour of ldap code when working with
1905 Windows 2003 AD Domain, where users was placed in more than one
1906 Organization Units. When I tried to give exim partial DN, the exit code
1907 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1908 But simultaneously result of request was absolutely normal ldap result,
1909 so I produce this patch..."
1911 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1912 REFERENCE, so I have modified the code to exclude the patch when that macro
1915 55. Some experimental protocols are using DNS PTR records for new purposes. The
1916 keys for these records are domain names, not reversed IP addresses. The
1917 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1918 leaves it alone. Component reversal etc. now happens only for IP addresses.
1921 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1923 57. Double the size of the debug message buffer (to 2048) so that more of very
1924 long debug lines gets shown.
1926 58. The exicyclog utility now does better if the number of log files to keep
1927 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1929 59. Two changes related to the smtp_active_hostname option:
1931 (1) $smtp_active_hostname is now available as a variable.
1932 (2) The default for smtp_banner uses $smtp_active_hostname instead
1933 of $primary_hostname.
1935 60. The host_aton() function is supposed to be passed a string that is known
1936 to be a valid IP address. However, in the case of IPv6 addresses, it was
1937 not checking this. This is a hostage to fortune. Exim now panics and dies
1938 if the condition is not met. A case was found where this could be provoked
1939 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1940 components; fortuitously, this particular loophole had already been fixed
1941 by change 4.50/55 above.
1943 If there are any other similar loopholes, the new check in host_aton()
1944 itself should stop them being exploited. The report I received stated that
1945 data on the command line could provoke the exploit when Exim was running as
1946 exim, but did not say which command line option was involved. All I could
1947 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1948 running as the user.
1951 61. There was a buffer overflow vulnerability in the SPA authentication code
1952 (which came originally from the Samba project). I have added a test to the
1953 spa_base64_to_bits() function which I hope fixes it.
1956 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1957 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1959 63. The daemon start-up calls getloadavg() while still root for those OS that
1960 need the first call to be done as root, but it missed one case: when
1961 deliver_queue_load_max is set with deliver_drop_privilege. This is
1962 necessary for the benefit of the queue runner, because there is no re-exec
1963 when deliver_drop_privilege is set.
1965 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1966 This has been fixed.
1968 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1969 was in use, was not putting the data itself into the right store pool;
1970 consequently, it could be overwritten for a subsequent message in the same
1971 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1974 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1976 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1977 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1978 the function were treating the return as a boolean value, which happened to
1979 work because 0=false and not-0=true, but is not correct code.
1981 68. The host_aton() function was not handling scoped IPv6 addresses (those
1982 with, for example, "%eth0" on the end) correctly.
1984 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1985 negated items (that is, ~something) in unsigned ints. Some compilers
1986 apparently mutter when there is no cast.
1988 70. If an address verification called from an ACL failed, and did not produce a
1989 user-specific message (i.e. there was only a "system" message), nothing was
1990 put in $acl_verify_message. In this situation, it now puts the system
1993 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1994 session; change 4.31/43 added the unwanted input to the log line - except
1995 that it did not do this in the start of session case. It now does.
1997 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1998 This is wrong and can cause the other end to generate a synchronization
1999 error if it is another Exim or anything else that does the synchronization
2000 check. A QUIT command is no longer sent after a timeout.
2002 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2003 during host lookups.
2005 74. The defer_ok option of callout verification was not working if it was used
2006 when verifying addresses in header lines, that is, for this case:
2008 verify = header_sender/callout=defer_ok
2010 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2011 those file descriptors could be used for SMTP connections. If anything
2012 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2013 could be sent to the SMTP client, causing chaos. The daemon now opens
2014 stdin, stdout, and stderr to /dev/null when it puts itself into the
2017 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2018 log by default. The user can override this with suitable Perl magic.
2020 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2021 the log message when discard triggers, was not working for the DATA ACL or
2022 for the non-SMTP ACL.
2024 78. Error message wording change in sieve.c.
2026 79. If smtp_accept_max_per_host was set, the number of connections could be
2027 restricted to fewer than expected, because the daemon was trying to set up
2028 a new connection before checking whether the processes handling previous
2029 connections had finished. The check for completed processes is now done
2030 earlier. On busy systems, this bug wouldn't be noticed because something
2031 else would have woken the daemon, and it would have reaped the completed
2034 80. If a message was submitted locally by a user whose login name contained one
2035 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2036 It caused a spool format error. I have fixed the spool reading code. A
2037 related problem was that the "from" clause in the Received: line became
2038 illegal because of the space(s). It is now covered by ${quote_local_part.
2040 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2043 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2045 83. Patch from Sieve maintainer to fix unterminated string problem in
2046 "vacation" handling.
2048 84. Some minor changes to the Linux configuration files to help with other
2049 OS variants using glibc.
2051 85. One more patch for Sieve to update vacation handling to latest spec.
2054 ----------------------------------------------------
2055 See the note above about the 4.44 and 4.50 releases.
2056 ----------------------------------------------------
2062 1. Change 4.43/35 introduced a bug that caused file counts to be
2063 incorrectly computed when quota_filecount was set in an appendfile
2066 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2067 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2070 3. Give more explanation in the error message when the command for a transport
2071 filter fails to execute.
2073 4. There are several places where Exim runs a non-Exim command in a
2074 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2075 was being done only for the command run by the queryprogram router. It is
2076 now done for all such subprocesses. The other cases are: ${run, transport
2077 filters, and the commands run by the lmtp and pipe transports.
2079 5. Some older OS have a limit of 256 on the maximum number of file
2080 descriptors. Exim was using setrlimit() to set 1000 as a large value
2081 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2082 systems. I've change it so that if it can't get 1000, it tries for 256.
2084 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2085 was an oversight, and furthermore, ever since the addition of extra
2086 controls (e.g. 4.43/32), the checks on when to allow different forms of
2087 "control" were broken. There should now be diagnostics for all cases when a
2088 control that does not make sense is encountered.
2090 7. $recipients is now available in the predata ACL (oversight).
2092 8. Tidy the search cache before the fork to do a delivery from a message
2093 received from the command line. Otherwise the child will trigger a lookup
2094 failure and thereby defer the delivery if it tries to use (for example) a
2095 cached ldap connection that the parent has called unbind on.
2097 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2098 of $address_data from the recipient verification was clobbered by the
2099 sender verification.
2101 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2102 was its contents. (It was OK if the option was not defined at all.)
2104 11. A "Completed" log line is now written for messages that are removed from
2105 the spool by the -Mrm option.
2107 12. $host_address is now set to the target address during the checking of
2108 ignore_target_hosts.
2110 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2111 being passed; this would have caused $sender_host_name to have been used if
2112 matching the list had actually called for a host name (not very likely,
2113 since this list is usually IP addresses). A host name is now passed as
2116 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2117 code that specifies a non-restarting handler (typically sigaction() in
2118 modern systems) in an attempt to fix a rare and obscure crash bug.
2120 15. Narrowed the window for a race in the daemon that could cause it to ignore
2121 SIGCHLD signals. This is not a major problem, because they are used only to
2122 wake it up if nothing else does.
2124 16. A malformed maildirsize file could cause Exim to calculate negative values
2125 for the mailbox size or file count. Odd effects could occur as a result.
2126 The maildirsize information is now recalculated if the size or filecount
2129 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2130 support for a long time. Removed HAVE_SYS_VFS_H.
2132 18. Updated exipick to current release from John Jetmore.
2134 19. Allow an empty sender to be matched against a lookup in an address list.
2135 Previously the only cases considered were a regular expression, or an
2138 20. Exim went into a mad DNS lookup loop when doing a callout where the
2139 host was specified on the transport, if the DNS lookup yielded more than
2142 21. The RFC2047 encoding function was originally intended for short strings
2143 such as real names; it was not keeping to the 75-character limit for
2144 encoded words that the RFC imposes. It now respects the limit, and
2145 generates multiple encoded words if necessary. To be on the safe side, I
2146 have increased the buffer size for the ${rfc2047: expansion operator from
2149 22. Failure to deliver a bounce message always caused it to be frozen, even if
2150 there was an errors_to setting on the router. The errors_to setting is now
2153 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2154 canonical form (fully expanded) before being placed in
2155 $sender_host_address.
2157 24. Updated eximstats to version 1.33
2159 25. Include certificate and key file names in error message when GnuTLS fails
2160 to set them up, because the GnuTLS error message doesn't include the name
2161 of the failing file when there is a problem reading it.
2163 26. Expand error message when OpenSSL has problems setting up cert/key files.
2166 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2167 (this can affect the format of dates).
2169 28. exim_tidydb, when checking for the continued existence of a message for
2170 which it has found a message-specific retry record, was not finding
2171 messages that were in split spool directories. Consequently, it was
2172 deleting retry records that should have stayed in existence.
2174 29. eximstats updated to version 1.35
2175 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2176 1.35 - bugfix such that pie charts by volume are generated correctly
2178 30. The SPA authentication driver was not abandoning authentication and moving
2179 on to the next authenticator when an expansion was forced to fail,
2180 contradicting the general specification for all authenticators. Instead it
2181 was generating a temporary error. It now behaves as specified.
2183 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2184 (the order specifies the preference for clients). The order is now AES256,
2185 AES128, 3DES, ARCFOUR128.
2187 31. Small patch to Sieve code - explicitly set From: when generating an
2190 32. Exim crashed if a remote delivery caused a very long error message to be
2191 recorded - for instance if somebody sent an entire SpamAssassin report back
2192 as a large number of 550 error lines. This bug was coincidentally fixed by
2193 increasing the size of one of Exim's internal buffers (big_buffer) that
2194 happened as part of the Exiscan merge. However, to be on the safe side, I
2195 have made the code more robust (and fixed the comments that describe what
2198 33. Some experimental protocols are using DNS PTR records for new purposes. The
2199 keys for these records are domain names, not reversed IP addresses. The
2200 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2201 leaves it alone. Component reversal etc. now happens only for IP addresses.
2204 34. The host_aton() function is supposed to be passed a string that is known
2205 to be a valid IP address. However, in the case of IPv6 addresses, it was
2206 not checking this. This is a hostage to fortune. Exim now panics and dies
2207 if the condition is not met. A case was found where this could be provoked
2208 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2209 components; fortuitously, this particular loophole had already been fixed
2210 by change 4.50/55 or 4.44/33 above.
2212 If there are any other similar loopholes, the new check in host_aton()
2213 itself should stop them being exploited. The report I received stated that
2214 data on the command line could provoke the exploit when Exim was running as
2215 exim, but did not say which command line option was involved. All I could
2216 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2217 running as the user.
2220 35. There was a buffer overflow vulnerability in the SPA authentication code
2221 (which came originally from the Samba project). I have added a test to the
2222 spa_base64_to_bits() function which I hope fixes it.
2225 36. The daemon start-up calls getloadavg() while still root for those OS that
2226 need the first call to be done as root, but it missed one case: when
2227 deliver_queue_load_max is set with deliver_drop_privilege. This is
2228 necessary for the benefit of the queue runner, because there is no re-exec
2229 when deliver_drop_privilege is set.
2231 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2232 was in use, was not putting the data itself into the right store pool;
2233 consequently, it could be overwritten for a subsequent message in the same
2234 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2237 38. Sometimes the final signoff response after QUIT could fail to get
2238 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2239 < 0 before doing a fflush(). This bug looks as though it goes back to the
2240 introduction of TLS in release 3.20, but "sometimes" must have been rare
2241 because the tests only now provoked it.
2247 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2248 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2249 Now it returns an integer. A number of calls were still expecting a T/F
2250 return. Fortuitously, in all cases, the tests worked in OK situations,
2251 which is the norm. However, things would have gone wrong on any write
2252 failures on the smtp file descriptor. This function is used when sending
2253 messages over SMTP and also when doing verify callouts.
2255 2. When Exim is called to do synchronous delivery of a locally submitted
2256 message (the -odf or -odi options), it no longer closes stderr before doing
2259 3. Implemented the mua_wrapper option.
2261 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2263 5. Implemented the functions header_remove(), header_testname(),
2264 header_add_at_position(), and receive_remove_recipient(), and exported them
2267 6. If an ACL "warn" statement specified the addition of headers, Exim already
2268 inserted X-ACL-Warn: at the start if there was no header name. However, it
2269 was not making this test for the second and subsequent header lines if
2270 there were newlines in the string. This meant that an invalid header could
2271 be inserted if Exim was badly configured.
2273 7. Allow an ACL "warn" statement to add header lines at the start or after all
2274 the Received: headers, as well as at the end.
2276 8. Added the rcpt_4xx retry error code.
2278 9. Added postmaster_mailfrom=xxx to callout verification option.
2280 10. Added mailfrom=xxxx to the callout verification option, for verify=
2283 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2284 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2286 12. Inserted some casts to stop certain compilers warning when using pointer
2287 differences as field lengths or precisions in printf-type calls (mostly
2288 affecting debugging statements).
2290 13. Added optional readline() support for -be (dynamically loaded).
2292 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2293 same clock tick as a message's arrival, so that its received time was the
2294 same as the "first fail" time on the retry record, and that message
2295 remained on the queue past the ultimate address timeout, every queue runner
2296 would try a delivery (because it was past the ultimate address timeout) but
2297 after another failure, the ultimate address timeout, which should have then
2298 bounced the address, did not kick in. This was a "< instead of <=" error;
2299 in most cases the first failure would have been in the next clock tick
2300 after the received time, and all would be well.
2302 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2303 being recognized when the domain list was tested by the match_domain
2304 condition in an expansion string.
2306 16. Added the ${str2b64: operator.
2308 17. Exim was always calling setrlimit() to set a large limit for the number of
2309 processes, without checking whether the existing limit was already
2310 adequate. (It did check for the limit on file descriptors.) Furthermore,
2311 errors from getrlimit() and setrlimit() were being ignored. Now they are
2312 logged to the main and panic logs, but Exim does carry on, to try to do its
2313 job under whatever limits there are.
2315 18. Imported PCRE 5.0.
2317 19. Trivial typo in log message " temporarily refused connection" (the leading
2320 20. If the log selector return_path_on_delivery was set and an address was
2321 redirected to /dev/null, the delivery process crashed because it assumed
2322 that a return path would always be set for a "successful" delivery. In this
2323 case, the whole delivery is bypassed as an optimization, and therefore no
2326 21. Internal re-arrangement: the function for sending a challenge and reading
2327 a response while authentication was assuming a zero-terminated challenge
2328 string. It's now changed to take a pointer and a length, to allow for
2329 binary data in such strings.
2331 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2333 23. Exim was not respecting finduser_retries when seeking the login of the
2334 uid under which it was called; it was always trying 10 times. (The default
2335 setting of finduser_retries is zero.) Also, it was sleeping after the final
2336 failure, which is pointless.
2338 24. Implemented tls_on_connect_ports.
2340 25. Implemented acl_smtp_predata.
2342 26. If the domain in control=submission is set empty, Exim assumes that the
2343 authenticated id is a complete email address when it generates From: or
2344 Sender: header lines.
2346 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2347 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2348 chown and chgrp in /bin and hostname in /usr/bin.
2350 28. Exim was keeping the "process log" file open after each use, just as it
2351 does for the main log. This opens the possibility of it remaining open for
2352 long periods when the USR1 signal hits a daemon. Occasional processlog
2353 errors were reported, that could have been caused by this. Anyway, it seems
2354 much more sensible not to leave this file open at all, so that is what now
2357 29. The long-running daemon process does not normally write to the log once it
2358 has entered its main loop, and it closes the log before doing so. This is
2359 so that log files can straightforwardly be renamed and moved. However,
2360 there are a couple of unusual error situations where the daemon does write
2361 log entries, and I had neglected to close the log afterwards.
2363 30. The text of an SMTP error response that was received during a remote
2364 delivery was being truncated at 512 bytes. This is too short for some of
2365 the long messages that one sometimes sees. I've increased the limit to
2368 31. It is now possible to make retry rules that apply only when a message has a
2369 specific sender, in particular, an empty sender.
2371 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2372 it possible to be selective about when SMTP synchronization is enforced.
2374 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2376 32. Implemented hosts_connection_nolog.
2378 33. Added an ACL for QUIT.
2380 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2383 35. Added mailbox_size and mailbox_filecount to appendfile.
2385 36. Added control = no_multiline_responses to ACLs.
2387 37. There was a bug in the logic of the code that waits for the clock to tick
2388 in the case where the clock went backwards by a substantial amount such
2389 that the microsecond fraction of "now" was more than the microsecond
2390 fraction of "then" (but the whole seconds number was less).
2392 38. Added support for the libradius Radius client library this is found on
2393 FreeBSD (previously only the radiusclient library was supported).
2399 1. When certain lookups returned multiple values in the form name=value, the
2400 quoting of the values was not always being done properly. Specifically:
2401 (a) If the value started with a double quote, but contained no whitespace,
2403 (b) If the value contained whitespace other than a space character (i.e.
2404 tabs or newlines or carriage returns) it was not quoted.
2405 This fix has been applied to the mysql and pgsql lookups by writing a
2406 separate quoting function and calling it from the lookup code. The fix
2407 should probably also be applied to nisplus, ibase and oracle lookups, but
2408 since I cannot test any of those, I have not disturbed their existing code.
2410 2. A hit in the callout cache for a specific address caused a log line with no
2411 reason for rejecting RCPT. Now it says "Previous (cached) callout
2412 verification failure".
2414 3. There was an off-by-one bug in the queryprogram router. An over-long
2415 return line was truncated at 256 instead of 255 characters, thereby
2416 overflowing its buffer with the terminating zero. As well as fixing this, I
2417 have increased the buffer size to 1024 (and made a note to document this).
2419 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2420 when Exim is waiting for an SMTP response from a remote server, Exim
2421 restarts its select() call on the socket, thereby resetting its timeout.
2422 This is not a problem when such interrupts are rare. Somebody set up a cron
2423 job to run exiwhat every 2 minutes, which is less than the normal select()
2424 timeout (5 or 10 minutes). This meant that the select() timeout never
2425 kicked in because it was always reset. I have fixed this by comparing the
2426 time when an interrupt arrives with the time at the start of the first call
2427 to select(). If more time than the timeout has elapsed, the interrupt is
2428 treated as a timeout.
2430 5. Some internal re-factoring in preparation for the addition of Sieve
2431 extensions (by MH). In particular, the "personal" test is moved to a
2432 separate function, and given an option for scanning Cc: and Bcc: (which is
2433 not set for Exim filters).
2435 6. When Exim created an email address using the login of the caller as the
2436 local part (e.g. when creating a From: or Sender: header line), it was not
2437 quoting the local part when it contained special characters such as @.
2439 7. Installed new OpenBSD configuration files.
2441 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2442 try to make them clearer.
2444 9. Callout options, other than the timeout value, were being ignored when
2445 verifying sender addresses in header lines. For example, when using
2447 verify = header_sender/callout=no_cache
2449 the cache was (incorrectly) being used.
2451 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2452 only the Cygwin environment.
2454 11. When return_path_on_delivery was set as a log selector, if different remote
2455 addresses in the same message used different return paths and parallel
2456 remote delivery occurred, the wrong values would sometimes be logged.
2457 (Whenever a remote delivery process finished, the return path value from
2458 the most recently started remote delivery process was logged.)
2460 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2461 lines when AUTH and/or TLS are in use. This is the "received protocol"
2462 field. Exim used to use "asmtp" for authenticated SMTP, without any
2463 indication (in the protocol name) for TLS use. Now it follows the RFC and
2464 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2465 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2466 names appear in log lines as well as in Received: header lines.
2468 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2469 extensions, and comparison tests, and to fix some bugs.
2471 14. Changes to the "personal" filter test:
2473 (1) The test was buggy in that it was just doing the equivalent of
2474 "contains" tests on header lines. For example, if a user's address was
2475 anne@some.where, the "personal" test would incorrectly be true for
2477 To: susanne@some.where
2479 This test is now done by extracting each address from the header in turn,
2480 and checking the entire address. Other tests that are part of "personal"
2481 are now done using regular expressions (for example, to check local parts
2482 of addresses in From: header lines).
2484 (2) The list of non-personal local parts in From: addresses has been
2485 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2486 taken from the Sieve specification recommendations.
2488 (3) If the message contains any header line starting with "List-" it is
2489 treated as non-personal.
2491 (4) The test for "circular" in the Subject: header line has been removed
2492 because it now seems ill-conceived.
2494 15. Minor typos in src/EDITME comments corrected.
2496 16. Installed latest exipick from John Jetmore.
2498 17. If headers_add on a router specified a text string that was too long for
2499 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2500 of string_sprintf() is now avoided.
2502 18. $message_body_size was not set (it was always zero) when running the DATA
2503 ACL and the local_scan() function.
2505 19. For the "mail" command in an Exim filter, no default was being set for
2506 the once_repeat time, causing a random time value to be used if "once" was
2507 specified. (If the value happened to be <= 0, no repeat happened.) The
2508 default is now 0s, meaning "never repeat". The "vacation" command was OK
2509 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2510 (I found it when inspecting the code).
2512 20. There is now an overall timeout for performing a callout verification. It
2513 defaults to 4 times the callout timeout, which applies to individual SMTP
2514 commands during the callout. The overall timeout applies when there is more
2515 than one host that can be tried. The timeout is checked before trying the
2516 next host. This prevents very long delays if there are a large number of
2517 hosts and all are timing out (e.g. when the network connections are timing
2518 out). The value of the overall timeout can be changed by specifying an
2519 additional sub-option for "callout", called "maxwait". For example:
2521 verify = sender/callout=5s,maxwait=20s
2523 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2524 to the end before writing, but this should make it even safer).
2526 22. Exim was forgetting that it had advertised PIPELINING for the second and
2527 subsequent messages on an SMTP connection. It was also not resetting its
2528 memory on STARTTLS and an internal HELO.
2530 23. When Exim logs an SMTP synchronization error within a session, it now
2531 records whether PIPELINING has been advertised or not.
2533 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2534 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2535 rather than long int.
2537 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2539 26. Added the never_mail option to autoreply.
2545 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2546 crash if the getsockname() call failed; this can happen if a connection is
2547 closed very soon after it is established. The problem was simply in the
2548 order in which certain operations were done, causing Exim to try to write
2549 to the SMTP stream before it had set up the file descriptor. The bug has
2550 been fixed by making things happen in the correct order.
2556 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2557 before the connection was closed, thus losing the rejection response.
2559 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2560 some early Solaris releases, but causes trouble in current releases where
2561 socklen_t is defined.
2563 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2566 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2569 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2570 that releasing the top of it at the end releases what was used for sub-
2571 expansions (unless the block got too big). However, discard this block if
2572 the first thing is a variable or header, so that we can use its block when
2573 it is dynamic (useful for very large $message_headers, for example).
2575 6. Lookups now cache *every* query, not just the most recent. A new, separate
2576 store pool is used for this. It can be recovered when all lookup caches are
2577 flushed. Lookups now release memory at the end of their result strings.
2578 This has involved some general refactoring of the lookup sources.
2580 7. Some code has been added to the store_xxx() functions to reduce the amount
2581 of flapping under certain conditions.
2583 8. log_incoming_interface used to affect only the <= reception log lines. Now
2584 it causes the local interface and port to be added to several more SMTP log
2585 lines, for example "SMTP connection from", and rejection lines.
2587 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2589 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2591 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2592 could be overwritten at the end of the current message (or the start of a
2593 new message if it was set in a HELO ACL). The value is now preserved for
2594 the duration of the SMTP connection.
2596 12. If a transport had a headers_rewrite setting, and a matching header line
2597 contained an unqualified address, that address was qualified, even if it
2598 did not match any rewriting rules. The underlying bug was that the values
2599 of the flags that permit the existence of unqualified sender and recipient
2600 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2601 non-local messages, and by -bnq for local messages) were not being
2602 preserved with the message after it was received.
2604 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2605 "next input=" as part of the text comprising the host identity instead of
2606 the correct text. The code was using the same buffer for two different
2607 strings. However, depending on which order the printing function evaluated
2608 its arguments, the bug did not always show up. Under Linux, for example, my
2609 test suite worked just fine.
2611 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2612 doesn't work with some older versions of Perl. It has been changed to "my",
2613 which in any case is probably the better facility to use.
2615 15. A really picky compiler found some instances of statements for creating
2616 error messages that either had too many or two few arguments for the format
2619 16. The size of the buffer for calls to the DNS resolver has been increased
2620 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2621 for addresses that have a lot of PTR records. This alleviates a problem; it
2622 does not fully solve it.
2624 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2625 buffer now truncates the list and logs the incident, which is the same
2626 action as happens when Exim is looking up a host name and its aliases.
2627 Previously in this situation something unpredictable would happen;
2628 sometimes it was "internal error: store_reset failed".
2630 18. If a server dropped the connection unexpectedly when an Exim client was
2631 using GnuTLS and trying to read a response, the client delivery process
2632 crashed while trying to generate an error log message.
2634 19. If a "warn" verb in an ACL added multiple headers to a message in a single
2635 string, for example:
2637 warn message = H1: something\nH2: something
2639 the text was added as a single header line from Exim's point of view
2640 though it ended up OK in the delivered message. However, searching for the
2641 second and subsequent header lines using $h_h2: did not work. This has been
2642 fixed. Similarly, if a system filter added multiple headers in this way,
2643 the routers could not see them.
2645 20. Expanded the error message when iplsearch is called with an invalid key to
2646 suggest using net-iplsearch in a host list.
2648 21. When running tests using -bh, any delays imposed by "delay" modifiers in
2649 ACLs are no longer actually imposed (and a message to that effect is
2652 22. If a "gecos" field in a passwd entry contained escaped characters, in
2653 particular, if it contained a \" sequence, Exim got it wrong when building
2654 a From: or a Sender: header from that name. A second bug also caused
2655 incorrect handling when an unquoted " was present following a character
2656 that needed quoting.
2658 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2659 was not being matched caselessly.
2661 24. Arranged for all hyphens in the exim.8 source to be escaped with
2664 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2665 properly. Recipient callout cache records were still being keyed to include
2666 the sender, even when use_sender was set false. This led to far more
2667 callouts that were necessary. The sender is no longer included in the key
2668 when use_sender is false.
2670 26. Added "control = submission" modifier to ACLs.
2672 27. Added the ${base62d: operator to decode base 62 numbers.
2674 28. dnsdb lookups can now access SRV records.
2676 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2677 the configuration file.
2679 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2680 (-v) mode. This makes the output for a verbose queue run more intelligible.
2682 31. Added a use_postmaster feature to recipient callouts.
2684 32. Added the $body_zerocount variable, containing the number of binary zero
2685 bytes in the message body.
2687 33. The time of last modification of the "new" subdirectory is now used as the
2688 "mailbox time last read" when there is a quota error for a maildir
2691 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2693 35. Added +ignore_unknown as a special item in host lists.
2695 36. Code for decoding IPv6 addresses in host lists is now included, even if
2696 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2697 address was recognized as an IP address, but was then not correctly decoded
2698 into binary, causing unexpected and incorrect effects when compared with
2705 1. Very minor rewording of debugging text in manualroute to say "list of
2706 hosts" instead of "hostlist".
2708 2. If verify=header_syntax was set, and a header line with an unqualified
2709 address (no domain) and a large number of spaces between the end of the
2710 name and the colon was received, the reception process suffered a buffer
2711 overflow, and (when I tested it) crashed. This was caused by some obsolete
2712 code that should have been removed. The fix is to remove it!
2714 3. When running in the test harness, delay a bit after writing a bounce
2715 message to get a bit more predictability in the log output.
2717 4. Added a call to search_tidyup() just before forking a reception process. In
2718 theory, someone could use a lookup in the expansion of smtp_accept_max_
2719 per_host which, without the tidyup, could leave open a database connection.
2721 5. Added the variables $recipient_data and $sender_data which get set from a
2722 lookup success in an ACL "recipients" or "senders" condition, or a router
2723 "senders" option, similar to $domain_data and $local_part_data.
2725 6. Moved the writing of debug_print from before to after the "senders" test
2728 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2729 problems for message scanning, either using a data ACL, or using
2730 local_scan() because the Received: header was not generated till after they
2731 were called (in order to set the time as the time of reception completion).
2732 I have revised the way this works. The header is now generated after the
2733 body is received, but before the ACL or local_scan() are called. After they
2734 are run, the timestamp in the header is updated.
2740 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2741 before starting a queue runner without re-exec. This happened only when
2742 deliver_drop_privilege was set or when the Exim user was set to root. The
2743 effect of the bug was that timeouts during subsequent deliveries caused
2744 crashes instead of being properly handled. The handler is now left at its
2745 default (and expected) setting.
2747 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2748 message, again when deliver_drop_privilege is set or Exim is run as root.
2749 The bug described in (1) was not present in this case, but the tidying up
2750 of the other signals was missing. I have made the two cases consistent.
2752 3. The ignore_target_hosts setting on a manualroute router was being ignored
2753 for hosts that were looked up using the /MX notation.
2755 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2758 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2759 operated on the sender address. After changing the $sender_address to <>
2760 for the sender address verify, Exim was re-instated it as the original
2761 (before rewriting) address, but remembering that it had rewritten it, so it
2762 wasn't rewriting it again. This bug also had the effect of breaking the
2763 sender address verification caching when the sender address was rewritten.
2765 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2766 This has been changed so that if the ip literal address matches
2767 ignore_target_hosts, the router declines.
2769 7. Added expansion conditions match_domain, match_address, and match_local_
2770 part (NOT match_host).
2772 8. The placeholder for the Received: header didn't have a length field set.
2774 9. Added code to Exim itself and to exim_lock to test for a specific race
2775 condition that could lead to file corruption when using MBX delivery. The
2776 issue is with the lockfile that is created in /tmp. If this file is removed
2777 after a process has opened it but before that process has acquired a lock,
2778 there is the potential for a second process to recreate the file and also
2779 acquire a lock. This could lead to two Exim processes writing to the file
2780 at the same time. The added code performs the same test as UW imapd; it
2781 checks after acquiring the lock that its file descriptor still refers to
2782 the same named file.
2784 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2785 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2788 11. Added the smtp_active_hostname option. If used, this will typically be made
2789 to depend on the incoming interface address. Because $interface_address is
2790 not set up until the daemon has forked a reception process, error responses
2791 that can happen earlier (such as "too many connections") no longer contain
2794 12. If an expansion in a condition on a "warn" statement fails because a lookup
2795 defers, the "warn" statement is abandoned, and the next ACL statement is
2796 processed. Previously this caused the whole ACL to be aborted.
2798 13. Added the iplsearch lookup type.
2800 14. Added ident_timeout as a log selector.
2802 15. Added tls_certificate_verified as a log selector.
2804 16. Added a global option tls_require_ciphers (compare the smtp transport
2805 option of the same name). This controls incoming TLS connections.
2807 17. I finally figured out how to make tls_require_ciphers do a similar thing
2808 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2809 before starting the TLS session.
2811 18. Tabs are now shown as \t in -bP output.
2813 19. If the log selector return_path_on_delivery was set, Exim crashed when
2814 bouncing a message because it had too many Received: header lines.
2816 20. If two routers both had headers_remove settings, and the first one included
2817 a superfluous trailing colon, the final name in the first list and the
2818 first name in the second list were incorrectly joined into one item (with a
2819 colon in the middle).
2825 1. Added -C and -D options to the exinext utility, mainly to make it easier
2826 to include in the automated testing, but these could be helpful when
2827 multiple configurations are in use.
2829 2. The exinext utility was not formatting the output nicely when there was
2830 an alternate port involved in the retry record key, nor when there was a
2831 message id as well (for retries that were specific to a specific message
2832 and a specific host). It was also confused by IPv6 addresses, because of
2833 the additional colons they contain. I have fixed the IPv4 problem, and
2834 patched it up to do a reasonable job for IPv6.
2836 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2837 delivery, the log line now contains "pipelined" if PIPELINING was used.
2839 4. An SMTP transport process used to panic and die if the bind() call to set
2840 an explicit outgoing interface failed. This has been changed; it is now
2841 treated in the same way as a connect() failure.
2843 5. A reference to $sender_host_name in the part of a conditional expansion
2844 that was being skipped was still causing a DNS lookup. This no longer
2847 6. The def: expansion condition was not recognizing references to header lines
2848 that used bh_ and bheader_.
2850 7. Added the _cache feature to named lists.
2852 8. The code for checking quota_filecount in the appendfile transport was
2853 allowing one more file than it should have been.
2855 9. For compatibility with Sendmail, the command line option
2863 and sets the incoming protocol and host name (for trusted callers). The
2864 host name and its colon can be omitted when only the protocol is to be set.
2865 Note the Exim already has two private options, -pd and -ps, that refer to
2866 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2867 "s", but I don't think that's a major issue.
2869 10. A number of refactoring changes to the code, none of which should affect
2872 (a) The number of logging options was getting close to filling up the
2873 32-bit word that was used as a bit map. I have split them into two classes:
2874 those that are passed in the argument to log_write(), and those that are
2875 only ever tested independently outside of that function. These are now in
2876 separate 32-bit words, so there is plenty of room for expansion again.
2877 There is no change in the user interface or the logging behaviour.
2879 (b) When building, for example, log lines, the code previously used a
2880 macro that called string_cat() twice, in order to add two strings. This is
2881 not really sufficiently general. Furthermore, there was one instance where
2882 it was actually wrong because one of the argument was used twice, and in
2883 one call a function was used. (As it happened, calling the function twice
2884 did not affect the overall behaviour.) The macro has been replaced by a
2885 function that can join an arbitrary number of extra strings onto a growing
2888 (c) The code for expansion conditions now uses a table and a binary chop
2889 instead of a serial search (which was left over from when there were very
2890 few conditions). Also, it now recognizes conditions like "pam" even when
2891 the relevant support is not compiled in: a suitably worded error message is
2892 given if an attempt is made to use such a condition.
2894 11. Added ${time_interval:xxxxx}.
2896 12. A bug was causing one of the ddress fields not to be passed back correctly
2897 from remote delivery subprocesses. The field in question was not being
2898 subsequently used, so this caused to problems in practice.
2900 13. Added new log selectors queue_time and deliver_time.
2902 14. Might have fixed a bug in maildirsizefile handling that threw up
2903 "unexpected character" debug warnings, and recalculated the data
2904 unnecessarily. In any case, I expanded the warning message to give more
2907 15. Added the message "Restricted characters in address" to the statements in
2908 the default ACL that block characters like @ and % in local parts.
2910 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2911 Three changes have been made:
2913 (a) There was a serious bug; a negative response to MAIL caused the whole
2914 recipient domain to be cached as invalid, thereby blocking all messages
2915 to all local parts at the same domain, from all senders. This bug has
2916 been fixed. The domain is no longer cached after a negative response to
2917 MAIL if the sender used is not empty.
2919 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2922 (c) A new callout option, "use_sender" has been added for people who want
2923 the modified behaviour.
2929 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2932 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2933 indeed breaks things for older releases.
2935 3. Added additional logging to the case where there is a problem reading data
2936 from a filter that is running in a subprocess using a pipe, in order to
2937 try to track down a specific problem.
2939 4. Testing facility fudge: when running in the test harness and attempting
2940 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2941 getting "No route to host". Convert this to a timeout.
2943 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2946 6. Some OS don't have socklen_t but use size_t instead. This affects the
2947 fifth argument of getsockopt() amongst other things. This is now
2948 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2949 can be set for individual OS. I have set it for SunOS5, OSF1, and
2950 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2951 some earlier ones do not.
2953 7. Change 4.30/15 was not doing the test caselessly.
2955 8. The standard form for an IPv6 address literal was being rejected by address
2956 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2957 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2958 this, as well as the form without the "IPv6" on the front (but only when
2959 address literals are enabled, of course).
2961 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2963 10. Exim crashed if a message with an empty sender address specified by -f
2964 encountered a router with an errors_to setting. This could be provoked only
2965 by a command such as
2969 where an empty string was supplied; "<>" did not hit this bug.
2971 11. Installed PCRE release 4.5.
2973 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2974 remained set. It is now erased.
2976 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2977 times from message ids (which are base 36 rather than the normal 62).
2979 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2980 were being counted as actual protocol errors, and logged if the log
2981 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2982 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2983 and DATA following a set of rejected RCPTs do not count as protocol errors.
2984 In other words, Exim assumes they were pipelined, though this may not
2985 actually be the case. Of course, in all cases the client gets an
2986 appropriate error code.
2988 15. If a lookup fails in an ACL condition, a message about the failure may
2989 be available; it is used if testing the ACL cannot continue, because most
2990 such messages specify what the cause of the deferral is. However, some
2991 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2992 that caused an old message to be retained and used if a later statement
2993 caused a defer, replacing the real cause of the deferral.
2995 16. If an IP address had so many PTR records that the DNS lookup buffer
2996 was not large enough to hold them, Exim could crash while trying to process
2997 the truncated data. It now detects and logs this case.
2999 17. Further to 4.21/58, another change has been made: if (and only if) the
3000 first line of a message (the first header line) ends with CRLF, a bare LF
3001 in a subsequent header line has a space inserted after it, so as not to
3002 terminate the header.
3004 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3005 unnecessarily, used atoi() instead of strtol(), and didn't check the
3006 termination when getting file sizes from file names by regex.
3008 19. Completely re-implemented the support for maildirsize files, in the light
3009 of a number of problems with the previous contributed implementation
3010 (4.30/29). In particular:
3012 . If the quota is zero, the maildirsize file is maintained, but no quota is
3015 . If the maildir directory does not exist, it is created before any attempt
3016 to write a maildirsize file.
3018 . The quota value in the file is just a cache; if the quota is changed in
3019 the transport, the new value overrides.
3021 . A regular expression is available for excluding directories from the
3024 20. The autoreply transport checks the characters in options that define the
3025 message's headers; it allows continued headers, but it was checking with
3026 isspace() after an embedded newline instead of explicitly looking for a
3029 21. If all the "regular" hosts to which an address was routed had passed their
3030 expiry times, and had not reached their retry times, the address was
3031 bounced, even if fallback hosts were defined. Now Exim should go on to try
3034 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3035 equivalent code in the SMTP transport. Some hosts send humungous responses
3036 to HELO/EHLO, more than 1024 it seems.
3038 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3039 gives compiler warnings in some environments. I've now done it "properly",
3042 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3043 (because of problems with the built-in one) was declared to return uschar *
3044 instead of char *, causing compiler failure.
3046 25. Fixed a file descriptor leak when processing alias/forward files.
3048 26. Fixed a minor format string issue in dbfn.c.
3050 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3052 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3053 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3056 29. When a sender address is verified, it is cached, to save repeating the test
3057 when there is more than one recipient in a message. However, when the
3058 verification involves a callout, it is possible for different callout
3059 options to be set for different recipients. It is too complicated to keep
3060 track of this in the cache, so now Exim always runs a verification when a
3061 callout is required, relying on the callout cache for the optimization.
3062 The overhead is duplication of the address routing, but this should not be
3065 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3066 to be verified with callout=postmaster, and the main callout worked but the
3067 postmaster check failed, the verification correctly failed. However, if a
3068 subsequent RCPT command asked for sender verification *without* the
3069 postmaster check, incorrect caching caused this verification also to fail,
3072 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3073 it was not caching the DNS options (qualify_single, search_parents) that
3074 were used when the lookup failed. A subsequent lookup with different
3075 options therefore always gave the same answer, though there were cases
3076 where it should not have. (Example: a "domains = !$mx_any" option on a
3077 dnslookup router: the "domains" option is always processed without any
3078 widening, but the router might have qualify_single set.) Now Exim uses the
3079 cached value only when the same options are set.
3081 32. Added John Jetmore's "exipick" utility to the distribution.
3083 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3084 than a timeout (e.g. a certificate is required, and is not provided), an
3085 Exim server now closes the connection immediately. Previously it waited for
3086 the client to close - but if the client is SSL, it seems that they each
3087 wait for each other, leading to a delay before one of them times out.
3089 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3090 maintained 0.8.x compatibility because I don't think many are using it, and
3091 it is clearly obsolete.
3093 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3096 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3097 client certificate was expired. A simple patch fixes this, though I don't
3098 understand the full logic of why the verify callback is called multiple
3101 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3102 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3103 which causes problems with some clients (such as the Certicom SSL Plus
3104 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3105 disables the coutermeasure allowing Eudora to connect."
3107 38. Exim was not checking that a write() to a log file succeeded. This could
3108 lead to Bad Things if a log got too big, in particular if it hit a file
3109 size limit. Exim now panics and dies if it cannot write to a log file, just
3110 as it does if it cannot open a log file.
3112 39. Modified OS/Makefile-Linux so that it now contains
3114 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3116 The two -D definitions ensure that Exim is compiled with large file
3117 support, which makes it possible to handle log files that are bigger than
3120 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3121 instance) a domain was checked against a named list that involved a lookup,
3122 causing $domain_data to be set, then another domain was checked against the
3123 same list, then the first domain was re-checked, the value of $domain_data
3124 after the final check could be wrong. In particular, if the second check
3125 failed, it could be set empty. This bug probably also applied to
3128 41. The strip_trailing_dot option was not being applied to the address given
3129 with the -f command-line option.
3131 42. The code for reading a message's header from the spool was incrementing
3132 $received_count, but never initializing it. This meant that the value was
3133 incorrect (doubled) while delivering a message in the same process in which
3134 it was received. In the most common configuration of Exim, this never
3135 happens - a fresh exec is done - but it can happen when
3136 deliver_drop_privilege is set.
3138 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3139 it now includes up to 150 characters of the unexpected data in the log
3142 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3143 and building data strings. The size of both of these buffers was 10 000
3144 bytes - far larger than anybody would *ever* want, thought I. Needless to
3145 say, somebody hit the limit. I have increased the maximum line length to
3146 20 000 and the maximum data length of concatenated lines to 100 000. I have
3147 also fixed two bugs, because there was no checking on these buffers. Tsk,
3148 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3149 buffer is too small.
3151 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3152 lsearch lookups. Now it does.
3154 46. When parsing a route_list item in a manualroute router, a fixed-length
3155 buffer was used for the list of hosts. I made this 1024 bytes long,
3156 thinking that nobody would ever have a list of hosts that long. Wrong.
3157 Somebody had a whole pile of complicated expansion conditions, and the
3158 string was silently truncated, leading to an expansion error. It turns out
3159 that it is easier to change to an unlimited length (owing to other changes
3160 that have happened since this code was originally written) than to build
3161 structure for giving a limitation error. The length of the item that
3162 expands into the list of hosts is now unlimited.
3164 47. The lsearch lookup could not handle data where the length of text line was
3165 more than 4095 characters. Such lines were truncated, leading to shortened
3166 data being returned. It should now handle lines of any length.
3168 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3169 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3172 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3173 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3174 no longer happens in comment lines. A list of replacements is now placed
3175 at the head of all of the source files, except those whose only change is
3176 to replace PERL_COMMAND in the very first #! line.
3178 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3179 messages on the queue, with a bottom-up merge sort, using code contributed
3180 by Michael Haardt. This should make operations like -bp somewhat faster on
3181 large queues. It won't affect queue runners, except when queue_run_in_order
3184 51. Installed eximstats 1.31 in the distribution.
3186 52. Added support for SRV lookups to the dnslookup router.
3188 53. If an ACL referred to $message_body or $message_body_end, the value was not
3189 reset for any messages that followed in the same SMTP session.
3191 54. The store-handling optimization for building very long strings was not
3192 differentiating between the different store pools. I don't think this
3193 actually made any difference in practice, but I've tidied it.
3195 55. While running the routers to verify a sender address, $sender_address
3196 was still set to the sender address. This is wrong, because when routing to
3197 send a bounce to the sender, it would be empty. Therefore, I have changed
3198 it so that, while verifying a sender address, $sender_address is set to <>.
3199 (There is no change to what happens when verifying a recipient address.)
3201 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3202 target A or AAAA records (if not already returned) without resetting the
3203 qualify_single or search_parents options of the DNS resolver. These are
3204 inappropriate in this case because the targets of MX and SRV records must
3205 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3206 target that, when qualified, matched something in the local domain. These
3207 two options are now turned off when doing these lookups.
3209 57. It seems that at least some releases of Reiserfs (which does not have the
3210 concept of a fixed number of inodes) returns zero and not -1 for the
3211 number of available inodes. This interacted badly with check_spool_inodes,
3212 which assumed that -1 was the "no such thing" setting. What I have done is
3213 to check that the total number of inodes is greater than zero before doing
3214 the test of how many are available.
3216 58. When a "warn" ACL statement has a log_message modifier, the message is
3217 remembered, and not repeated. This is to avoid a lot of repetition when a
3218 message has many recipients that cause the same warning to be written.
3219 Howewer, Exim was preserving the list of already written lines for an
3220 entire SMTP session, which doesn't seem right. The memory is now reset if a
3221 new message is started.
3223 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3224 debugging output unless log_rewrite was also set.
3226 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3227 of (int)(handle) when we know that handle contains (void *)(-1).
3229 61. The Exim daemon panic-logs an error return when it closes the incoming
3230 connection. However "connection reset by peer" seems to be common, and
3231 isn't really an error worthy of noting specially, so that particular error
3234 62. When Exim is trying to find all the local interfaces, it used to panic and
3235 die if the ioctl to get the interface flags failed. However, it seems that
3236 on at least one OS (Solaris 9) it is possible to have an interface that is
3237 included in the list of interfaces, but for which you get a failure error
3238 for this call. This happens when the interface is not "plumbed" into a
3239 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3240 failure of the "get flags" call assumes that the interface is down.
3242 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3243 makes life easier for people who are doing arithmetic on fields extracted
3244 from dates, where you often get leading zeros that should not be
3245 interpreted as octal.
3247 64. Added qualify_domain to the redirect router, to override the global
3250 65. If a pathologically long header line contained very many addresses (the
3251 report of this problem mentioned 10 000) and each of them was rewritten,
3252 Exim could use up a very large amount of memory. (It kept on making new
3253 copies of the header line as it rewrote, and never released the old ones.)
3254 At the expense of a bit more processing, the header rewriting function has
3255 been changed so that it no longer eats memory in this way.
3257 66. The generation of the Received: header has been moved from the time that a
3258 message starts to be received, to the time that it finishes. The timestamp
3259 in the Received: header should now be very close to that of the <= log
3260 line. There are two side-effects of this change:
3262 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3263 logged header lines no longer include the local Received: line, because
3264 it has not yet been created. The same applies to a copy of the message
3265 that is returned to a non-SMTP sender when a message is rejected.
3267 (b) When a filter file is tested using -bf, no additional Received: header
3268 is added to the test message. After some thought, I decided that this
3271 This change does not affect the value of $received_for. It is still set
3272 after address rewriting, but before local_scan() is called.
3274 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3276 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3277 gave an unhelpful panic error message, and a defer error. I have managed to
3278 change this behaviour so that it now rejects any supplied certificate,
3279 which seems right, as the list of acceptable certificates is empty.
3281 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3282 gave an unhelpful defer error. I have not managed to make this reject any
3283 supplied certificates, but the error message it gives is "no certificate
3284 supplied", which is not helpful.
3286 70. exigrep's output now also includes lines that are not associated with any
3287 message, but which match the given pattern. Implemented by a patch from
3288 Martin Sluka, which also tidied up the Perl a bit.
3290 71. Recipient callout verification, like sender verification, was using <> in
3291 the MAIL FROM command. This isn't really the right thing, since the actual
3292 sender may affect whether the remote host accepts the recipient or not. I
3293 have changed it to use the actual sender in the callout; this means that
3294 the cache record is now keyed on a recipient/sender pair, not just the
3295 recipient address. There doesn't seem to be a real danger of callout loops,
3296 since a callout by the remote host to check the sender would use <>.
3297 [SEE ABOVE: changed after hitting problems.]
3299 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3300 temporary errors. However, in the case of such a code being given after
3301 the end of a data transmission (i.e. after ".") Exim was failing to write
3302 a retry record for the message. (Yes, there was some broken host that was
3303 actually sending 8xx at this point.)
3305 73. An unknown lookup type in a host list could cause Exim to panic-die when
3306 the list was checked. (An example that provoked this was putting <; in the
3307 middle of a list instead of at the start.) If this happened during a DATA
3308 ACL check, a -D file could be left lying around. This kind of configuration
3309 error no longer causes Exim to die; instead it causes a defer errror. The
3310 incident is still logged to the main and panic logs.
3312 74. Buglet left over from Exim 3 conversion. The message "too many messages
3313 in one connection" was written to the rejectlog but not the mainlog, except
3314 when address rewriting (yes!) was being logged.
3316 75. Added write_rejectlog option.
3318 76. When a system filter was run not as root (that is, when system_filter_user
3319 was set), the values of the $n variables were not being returned to the
3320 main process; thus, they were not subsequently available in the $sn
3323 77. Added +return_path_on_delivery log selector.
3325 78. A connection timeout was being treated differently from recipients deferred
3326 when testing hosts_max_try with a message that was older than the host's
3327 retry timeout. (The host should not be counted, thus allowing all hosts to
3328 be tried at least once before bouncing.) This may have been the cause of an
3329 occasionally reported bug whereby a message would remain on the queue
3330 longer than the retry timeout, but would be bounced if a delivery was
3331 forced. I say "may" because I never totally pinned down the problem;
3332 setting up timeout/retry tests is difficult. See also the next item.
3334 79. The ultimate address timeout was not being applied to errors that involved
3335 a combination of host plus message (for example, a timeout on a MAIL
3336 command). When an address resolved to a number of possible hosts, and they
3337 were not all tried for each delivery (e.g. because of hosts_max_try), a
3338 message could remain on the queue longer than the retry timeout.
3340 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3343 81. Fixed an obscure SMTP outgoing bug which required at least the following
3344 conditions: (a) there was another message waiting for the same server;
3345 (b) the server returned 5xx to all RCPT commands in the first message so
3346 that the message was not completed; (c) the server dropped the connection
3347 or gave a negative response to the RSET that Exim sends to abort the
3348 transaction. The observed case was a dropped connection after DATA that had
3349 been sent in pipelining mode. That is, the server had advertised PIPELINING
3350 but was not implementing it correctly. The effect of the bug was incorrect
3351 behaviour, such as trying another host, and this could lead to a crash.
3357 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3358 and daemon.c were passed as pointers to ints; they should have been
3359 pointers to socklen_t variables (which are typically unsigned ints).
3361 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3364 3. Fixed a really odd bug that affected only the testing scheme; patching a
3365 certain fixed string in the binary changed the value of another string that
3366 happened to be identical to the end of the original first string.
3368 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3369 name", it returns that address as the IP address. On some operating
3370 systems (e.g. Solaris), it also passes back the IP address string as the
3371 "host name". However, on others (e.g. Linux), it passes back an empty
3372 string. Exim wasn't checking for this, and was changing the host name to an
3373 empty string, assuming it had been canonicized.
3375 5. Although rare, it is permitted to have more than one PTR record for a given
3376 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3377 all the names associated with an address, because they do in Solaris.
3378 However, it seems that they do not in Linux for data that comes from the
3379 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3380 I found this out when I moved to a new Linux workstation and tried to run
3381 the Exim test suite.
3383 To get round this problem I have changed the code so that it now does its
3384 own call to the DNS to look up PTR records when searching for a host name.
3385 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3386 addresses that are only in /etc/hosts are still found.
3388 This behaviour is, however, controlled by an option called host_lookup_
3389 order, which defaults to "bydns:byaddr". If people want to use the other
3390 order, or indeed, just use one or the other means of lookup, they can
3391 specify it in this variable.
3393 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3394 some operating systems, this comes back from gethostbyaddr() as an empty
3395 string, and this is what Exim used to test for. However, it seems that in
3396 other systems, "." is yielded. Exim now tests for this case too.
3398 7. The values of check_spool_space and check_log_space are now held internally
3399 as a number of kilobytes instead of an absolute number of bytes. If a
3400 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3401 kilobyte. This means that much larger values can be stored.
3403 8. Exim monitor: an attempt to get the action menu when not actually pointing
3404 at a message produces an empty menu entitled "No message selected". This
3405 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3406 no entries in it ("Shell widget menu has zero width and/or height"). So I
3407 have added a single, blank menu entry in this case.
3409 9. Added ${quote_local_part.
3411 10. MIME decoding is now applied to the contents of Subject: header lines when
3414 11. Now that a reference to $sender_host_address automatically causes a reverse
3415 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3416 host lookup before query-style lookups in lists that might use this
3417 variable. This has therefore been abolished, and the "net-" prefix is no
3418 longer necessary for query-style lookups.
3420 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3421 have been a typo for LFLAGS, so it has been changed.
3423 13. The install script calls Exim with "-C /dev/null" in order to find the
3424 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3425 to be output. Howeve, since Exim outputs its version number before the
3426 error, it didn't break the script. It just looked ugly. I fixed this by
3427 always allowing "-C /dev/null" if the caller is root.
3429 14. Ignore overlarge ACL variable number when reading spool file - insurance
3430 against a later release with more variables having written the file.
3432 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3433 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3434 this, as well as the form without the "IPv6" on the front.
3436 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3437 OS/Makefile-Darwin file.
3439 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3440 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3442 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3443 variable. However, some OS use ints for inodes. I've added cast to long int
3444 to get rid of the compiler warning.
3446 19. I had forgotten to lock out "/../" in configuration file names when
3447 ALT_CONFIG_PREFIX was set.
3449 20. Routers used for verification do not need to specify transports. However,
3450 if such a router generated a host list, and callout was configured, Exim
3451 crashed, because it could not find a port number from the (non-existent)
3452 transport. It now assumes port 25 in this circumstance.
3454 21. Added the -t option to exigrep.
3456 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3457 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3458 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3459 them is set, LOOKUP_LSEARCH is forced.
3461 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3463 24. Added sender and host information to the "rejected by local_scan()" log
3464 line; previously there was no indication of these.
3466 25. Added .include_if_exists.
3468 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3469 Linux. It turns out that not all file systems support this. Apparently some
3470 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3471 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3472 not supported on the file descriptor, is now ignored when Exim is trying to
3473 sync a directory. This applies only to Linux.
3475 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3477 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3478 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3481 29. Added support for maildirsize files from supplied patch (modified a bit).
3483 30. The use of :fail: followed by an empty string could lead Exim to respond to
3484 sender verification failures with (e.g.):
3486 550 Verification failed for <xxx>
3487 550 Sender verify failed
3489 where the first response line was missing the '-' that indicates it is not
3490 the final line of the response.
3492 31. The loop for finding the name of the user that called Exim had a hardwired
3493 limit of 10; it now uses the value of finduser_retries, which is used for
3494 all other user lookups.
3496 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3499 33. Exim was neglecting to zero errno before one call of strtol() when
3500 expanding a string and expecting an integer value. On some systems this
3501 resulted in spurious "integer overflow" errors. Also, it was casting the
3502 result into an int without checking.
3504 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3505 did not work. The code looks as if it has *never* worked, though it appears
3506 to have been documented since at least releast 1.62. I have made it work.
3508 35. The "timeout_DNS" error in retry rules, also documented since at least
3509 1.62, also never worked. As it isn't clear exactly what this means, and
3510 clearly it isn't a major issue, I have abolished the feature by treating it
3511 as "timeout", and writing a warning to the main and panic logs.
3513 36. The display of retry rules for -brt wasn't always showing the error code
3516 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3517 timeout_connect_A, timeout_connect_MX.
3519 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3520 to the empty sender.
3522 39. The daemon was not analysing the content of -oX till after it had closed
3523 stderr and disconnected from the controlling terminal. This meant that any
3524 syntax errors were only noted on the panic log, and the return code from
3525 the command was 0. By re-arranging the code a little, I've made the
3526 decoding happen first, so such errors now appear on stderr, and the return
3527 code is 1. However, the actual setting up of the sockets still happens in
3528 the disconnected process, so errors there are still only recorded on the
3531 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3532 connections (as happens on some IP stacks) was logged at start up time as
3533 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3534 it from "IPv6 and IPv4", which means that two separate sockets are being
3537 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3538 says whether AF_INET or AF_INET6 was passed as an argument.
3540 42. Exiwhat output was messed up when time zones were included in log
3543 43. Exiwhat now gives more information about the daemon's listening ports,
3544 and whether -tls-on-connect was used.
3546 44. The "port" option of the smtp transport is now expanded.
3548 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3549 silently ignored. Now an error message is written to the main and panic
3552 46. There's a new ACL modifier called "logwrite" which writes to a log file
3553 as soon as it is encountered.
3555 47. Added $local_user_uid and $local_user_gid at routing time.
3557 48. Exim crashed when trying to verify a sender address that was being
3560 49. Exim was recognizing only a space character after ".include". It now also
3561 recognizes a tab character.
3563 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3564 extracting the relevant information from the specification. The man page no
3565 longer contains scrambled data for the -d option, and I've added a section
3566 at the front about calling Exim under different names.
3568 51. Added "extra_headers" argument to the "mail" command in filter files.
3570 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3573 53. Installed eximstats 1.29.
3575 54. Added transport_filter_timeout as a generic transport option.
3577 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3578 Cc: header lines. This was required by RFC 822, but it not required by RFC
3581 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3582 incoming messages that did not have them. Now it does so only if the
3583 message originates locally, that is, if there is no associated remote host
3584 address. When Resent- header lines are present, this applies to the Resent-
3585 lines rather than the non-Resent- lines.
3587 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3588 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3590 58. Messages for configuration errors now include the name of the main
3591 configuration file - useful now that there may be more than one file in a
3592 list (.included file names were always shown).
3594 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3595 for those rare installations that do not start the daemon as root or run it
3596 setuid root. I've cut out the call to initgroups() if the daemon is not
3599 60. The Exim user and group can now be bound into the binary as text strings
3600 that are looked up at the start of Exim's processing.
3602 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3604 62. Added $mailstore_basename variable.
3606 63. Installed patch to sieve.c from Michael Haardt.
3608 64. When Exim failed to open the panic log after failing to open the main log,
3609 the original message it was trying to log was written to stderr and debug
3610 output, but if they were not available (the usual case in production), it
3611 was lost. Now it is written to syslog before the two lines that record the
3612 failures to open the logs.
3614 65. Users' Exim filters run in subprocesses under the user's uid. It is
3615 possible for a "deliver" command or an alias in a "personal" command to
3616 provoke an address rewrite. If logging of address rewriting is configured,
3617 this fails because the process is not running as root or exim. There may be
3618 a better way of dealing with this, but for the moment (because 4.30 needs
3619 to be released), I have disabled address rewrite logging when running a
3620 filter in a non-root, non-exim process.
3626 1. The buildconfig auxiliary program wasn't quoting the value set for
3627 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3628 not defined. This bug was masked in 4.22 by the effect that was fixed in
3631 2. Some messages that were rejected after a message id was allocated were
3632 shown as "incomplete" by exigrep. It no longer does this for messages that
3633 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3635 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3636 have allow_domain_literals set, the ID did not get logged in the <= line.
3637 Domain literals are now always recognized in Message-ID: header lines.
3639 4. The first argument for a ${extract expansion item is the key name or field
3640 number. Leading and trailing spaces in this item were not being ignored,
3641 causing some misleading effects.
3643 5. When deliver_drop_privilege was set, single queue runner processes started
3644 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3645 same command in the process it spins off) were not dropping privilege.
3647 6. When the daemon running as "exim" started a queue runner, it always
3648 re-executed Exim in the spun-off process. This is a waste of effort when
3649 deliver_drop_privilege is set. The new process now just calls the
3650 queue-runner function directly.
3656 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3659 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3660 ignored. Though the use of -oP was forcing the writing of a pid file, it
3661 was always written to the default place.
3663 3. If the message "no IP address found for host xxxx" is generated during
3664 incoming verification, it is now followed by identification of the incoming
3665 connection (so you can more easily find what provoked it).
3667 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3669 5. Added some features to "harden" Exim a bit more against certain attacks:
3671 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3672 be put in Local/Makefile. This is like the never_users runtime option,
3673 but it cannot be overridden. The default setting is "root".
3675 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3676 prefix string with which any file named in a -C command line option
3679 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3680 is retained for -C and -D only if the caller of Exim is root. Without
3681 it, the exim user may also use -C and -D and retain privilege.
3683 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3684 command line option is disabled.
3686 6. Macro names set by the -D option must start with an upper case letter, just
3687 like macro names defined in the configuration file.
3689 7. Added "dereference=" facility to LDAP.
3691 8. Two instances of the typo "uknown" in the source files are fixed.
3693 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3694 the Configure-Makefile script screwed up while processing it.
3696 10. Incorporated PCRE 4.4.
3698 11. The SMTP synchronization check was not operating right at the start of an
3699 SMTP session. For example, it could not catch a HELO sent before the client
3700 waited for the greeting. There is now a check for outstanding input at the
3701 point when the greeting is written. Because of the duplex, asynchronous
3702 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3703 way, but not yet received, when the check is performed.
3705 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3706 on TCP/IP sockets, because this apparently causes some broken clients to
3709 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3710 unchanged) from the Cygwin maintainer.
3712 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3713 was supported instead of testing for mbx. Effectively a typo.
3715 15. The spa authenticator server code was not checking that the input it
3716 received was valid base64.
3718 16. The debug output line for the "set" modifier in ACLs was not showing the
3719 name of the variable that was being set.
3721 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3723 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3724 lookup on its own. Something else was needed to trigger the lookup. For
3725 example, a match in host_lookup or the need for a host name in a host list.
3726 Now, if $sender_host_name is referenced and the host name has not yet been
3727 looked up, a lookup is performed. If the lookup fails, the variable remains
3728 empty, and $host_lookup_failed is set to "1".
3730 19. Added "eqi" as a case-independent comparison operator.
3732 20. The saslauthd authentication condition could segfault if neither service
3733 nor realm was specified.
3735 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3736 error message that was logged was misleading, and incoming SMTP
3737 connections were dropped. The message is now more accurate, and temporary
3738 errors are given to SMTP connections.
3740 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3741 (except RSET) with a 421 error, until QUIT is received. However, it was
3742 failing to send a response to QUIT.
3744 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3745 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3746 ACL is now run after the helo_try_verify_hosts code.
3748 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3749 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3750 case-independent, so other case variants are also recognized). Apparently
3751 some systems use these upper case variants.
3753 25. If more than two messages were waiting for the same host, and a transport
3754 filter was specified for the transport, Exim sent two messages over the
3755 same TCP/IP connection, and then failed with "socket operation on non-
3756 socket" when it tried to send the third.
3758 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3760 27. The extern definition of crypt16() in expand.c was not being excluded when
3761 the OS had its own crypt16() function.
3763 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3764 as a preferred synonym for return_size_limit, both as an option and as an
3767 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3769 30. Changed the default configuration ACL to relax the local part checking rule
3770 for addresses that are not in any local domains. For these addresses,
3771 slashes and pipe symbols are allowed within local parts, but the sequence
3772 /../ is explicitly forbidden.
3774 31. SPA server authentication was not clearing the challenge buffer before
3777 32. log_message in a "warn" ACL statement was writing to the reject log as
3778 well as to the main log, which contradicts the documentation and doesn't
3779 seem right (because no rejection is happening). So I have stopped it.
3781 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3782 However, I am unable to do any testing of this.
3784 34. Fixed an infelicity in the appendfile transport. When checking directories
3785 for a mailbox, to see if any needed to be created, it was accidentally
3786 using path names with one or more superfluous leading slashes; tracing
3787 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3789 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3790 contents are added to the log line that is written for every discarded
3791 recipient. (Previously a log_message setting was ignored.)
3793 36. The ${quote: operator now quotes the string if it is empty.
3795 37. The install script runs exim in order to find its version number. If for
3796 some reason other than non-existence or emptiness, which it checks, it
3797 could not run './exim', it was installing it with an empty version number,
3798 i.e. as "exim-". This error state is now caught, and the installation is
3801 38. An argument was missing from the function that creates an error message
3802 when Exim fails to connect to the socket for saslauthd authentication.
3803 This could cause Exim to crash, or give a corrupted message.
3805 39. Added isip, isip4, and isip6 to ${if conditions.
3807 40. The ACL variables $acl_xx are now saved with the message, and can be
3808 accessed later in routers, transports, and filters.
3810 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3811 strings in the file are not string-expanded.
3813 42. If a MAIL command specified a SIZE value that was too large to fit into an
3814 int variable, the check against message_size_limit failed. Such values are
3815 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3816 day this will have to be increased, but I don't think I want to be around
3817 when emails are that large.
3824 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3825 iconv() is not standard in FreeBSD.
3827 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3828 IPv6 enabled. The observed symptom was a segmentation fault on return from
3829 the function os_common_find_running_interfaces() in src/os.c.
3831 3. In the check_special_case() function in daemon.c I had used "errno" as an
3832 argument name, which causes warnings on some systems. This was basically a
3833 typo, since it was named "eno" in the comments!
3835 4. The code that waits for the clock to tick (at a resolution of some fraction
3836 of a second) so as to ensure message-id uniqueness was always waiting for
3837 at least one whole tick, when it could have waited for less. [This is
3838 almost certainly not relevant at current processor speeds, where it is
3839 unlikely to ever wait at all. But we try to future-proof.]
3841 5. The function that sleeps for a time interval that includes fractions of a
3842 second contained a race. It did not block SIGALRM between setting the
3843 timer, and suspending (a couple of lines later). If the interval was short
3844 and the sigsuspend() was delayed until after it had expired, the suspension
3845 never ended. On busy systems this could lead to processes getting stuck for
3848 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3849 process, before it forks any delivery processes. The open lookup caching
3850 mechanism meant that the open file or database connection was passed into
3851 the delivery process. The problem was that delivery processes always tidy
3852 up cached lookup data. This could cause a problem for the next delivery
3853 process started by the queue runner, because the external queue runner
3854 process does not know about the closure. So the next delivery process
3855 still has data in the lookup cache. In the case of a file lookup, there was
3856 no problem because closing a file descriptor in a subprocess doesn't affect
3857 the parent. However, if the lookup was caching a connection to a database,
3858 the connection was closed, and the second delivery process was likely to
3859 see errors such as "PGSQL: query failed: server closed the connection
3860 unexpectedly". The problem has been fixed by closing all cached lookups
3861 in a queue runner before running a delivery process.
3863 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3864 seem to have the "const" qualifier which it has on other OS. I've
3867 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3868 *of the same type* (client or server) with the same public name that an
3869 error should be diagnosed.
3871 9. When Exim looked up a host name for an IP address, but failed to find the
3872 original IP address when looking up the host name (a safety check), it
3873 output the message "<ip address> does not match any IP for NULL", which was
3874 confusing, to say the least. The bug was that the host name should have
3875 appeared instead of "NULL".
3877 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3878 user that is built into the binary, and the -C or -D options is used, root
3879 privilege is dropped before the configuration file is read. In addition,
3880 logging is switched to stderr instead of the normal log files. If the
3881 configuration then re-defines the Exim user, the unprivileged environment
3882 is probably not what is expected, so Exim logs a panic warning message (but
3885 However, if deliver_drop_privilege is set, the unprivileged state may well
3886 be exactly what is intended, so the warning has been cut out in that case,
3887 and Exim is allowed to try to write to its normal log files.
3893 1. smtp_return_error_details was not giving details for temporary sender
3894 or receiver verification errors.
3896 2. Diagnose a configuration error if two authenticators have the same public
3899 3. Exim used not to create the message log file for a message until the first
3900 delivery attempt. This could be confusing when incoming messages were held
3901 for policy or load reasons. The message log file is now created at the time
3902 the message is received, and an initial "Received" line is written to it.
3904 4. The automatically generated man page for command line options had a minor
3905 bug that caused no ill effects; however, a more serious problem was that
3906 the procedure for building the man page automatically didn't always
3907 operate. Consequently, release 4.20 contains an out-of-date version. This
3908 shouldn't happen again.
3910 5. When building Exim with embedded Perl support, the script that builds the
3911 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3912 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3914 6. The freeze_tell option was not being used for messages that were frozen on
3915 arrival, either by an ACL or by local_scan().
3917 7. Added the smtp_incomplete_transaction log selector.
3919 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3920 was accepting AUTH without a new EHLO.
3922 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3923 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3926 10. Logging of TCP/IP connections (when configured) now happens in the main
3927 daemon process instead of the child process, so that the TCP/IP connection
3928 count is more accurate (but it can never be perfect).
3930 11. The use of "drop" in a nested ACL was not being handled correctly in the
3931 outer ACL. Now, if condition failure induced by the nested "drop" causes
3932 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3933 or "require"), the connection is dropped.
3935 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3936 that yield "discard" can now be used with an "accept" or a "discard" verb,
3937 but an error is generated for any others (because I can't see a useful way
3938 to define what should happen).
3940 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3941 are now processed in the same way as lines in the Exim configuration file.
3942 In particular, continuation lines are supported.
3944 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3946 15. Added -ti meaning -t -i.
3948 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3949 domains, and warn by logging if others are found.
3951 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3952 returned from GIFCONF when Exim is trying to find the list of interfaces on
3953 a host. The code in os.c has been modified to copy each ifreq to an aligned
3954 structure in all cases.
3956 Also, in some cases, the returned ifreq's were being copied to a 'struct
3957 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3958 means the last couple of bytes of an IPv6 address could be chopped if the
3959 ifreq contained only a normal sockaddr (14 bytes storage).
3961 18. Named domain lists were not supported in the hosts_treat_as_local option.
3962 An entry such as +xxxx was not recognized, and was treated as a literal
3965 19. Ensure that header lines added by a DATA ACL are included in the reject log
3966 if the ACL subsequently rejects the message.
3968 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3969 MD5 (which is deprecated).
3971 21. When testing a filter file using -bf, Exim was writing a message when it
3972 took the sender from a "From " line in the message, but it was not doing so
3973 when it took $return_path from a Return-Path: header line. It now does.
3975 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3976 with a valid header line field name (a series of printing characters
3977 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3979 23. Changed "disc" in the source to "disk" to conform to the documentation and
3980 the book and for uniformity.
3982 24. Ignore Sendmail's -Ooption=value command line item.
3984 25. When execve() failed while trying to run a command in a pipe transport,
3985 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3986 could be confused with a return value of 69 from the command itself. This
3987 has been changed to 127, the value the shell returns if it is asked to run
3988 a non-existent command. The wording for the related log line suggests a
3989 non-existent command as the problem.
3991 26. If received_header_text expands to an empty string, do not add a Received:
3992 header line to the message. (Well, it adds a token one on the spool, but
3993 marks it "old" so that it doesn't get used or transmitted.)
3995 27. Installed eximstats 1.28 (addition of -nt option).
3997 28. There was no check for failure on the call to getsockname() in the daemon
3998 code. This can fail if there is a shortage of resources on the system, with
3999 ENOMEM, for example. A temporary error is now given on failure.
4001 29. Contrary to the C standard, it seems that in some environments, the
4002 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4003 program. Exim now does this explicitly; it affects the formatting of
4004 timestamps using strftime().
4006 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4007 complaints. I've now initialized all the variables, to avoid this.
4009 32. Header lines added by a system filter were not being "seen" during
4010 transport-time rewrites.
4012 33. The info_callback() function passed to OpenSSL is set up with type void
4013 (*)(SSL *, int, int), as described somewhere. However, when calling the
4014 function (actually a macro) that sets it up, the type void(*)() is
4015 expected. I've put in a cast to prevent warnings from picky compilers.
4017 34. If a DNS black list lookup found a CNAME record, but there were no A
4018 records associated with the domain it pointed at, Exim crashed.
4020 35. If a DNS black list lookup returned more than one A record, Exim ignored
4021 all but the first. It now scans all returned addresses if a particular IP
4022 value is being sought. In this situation, the contents of the
4023 $dnslist_value variable are a list of all the addresses, separated by a
4026 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4027 to accept a host name and all its aliases if the forward lookup for any of
4028 them yielded the IP address of the incoming connection. Now it accepts only
4029 those names whose forward lookup yields the correct IP address. Any other
4030 names are discarded. This closes a loophole whereby a rogue DNS
4031 administrator could create reverse DNS records to break through a
4032 wildcarded host restriction in an ACL.
4034 37. If a user filter or a system filter that ran in a subprocess used any of
4035 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4036 the wrong values were passed to the pipe command ($thisaddress had the
4037 value of $0, $0 had the value of $1, etc). This bug was introduced by
4038 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4040 38. Improved the line breaking for long SMTP error messages from ACLs.
4041 Previously, if there was no break point between 40 and 75 characters, Exim
4042 left the rest of the message alone. Two changes have been made: (a) I've
4043 reduced the minimum length to 35 characters; (b) if it can't find a break
4044 point between 35 and 75 characters, it looks ahead and uses the first one
4045 that it finds. This may give the occasional overlong line, but at least the
4046 remaining text gets split now.
4048 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4049 file descriptors might be low, and that setting 1000 would always raise it.
4050 It turns out that in some environments, the limit is already over 1000 and
4051 that lowering it causes trouble. So now Exim takes care not to decrease it.
4053 40. When delivering a message, the value of $return_path is set to $sender_
4054 address at the start of routing (routers may change the value). By an
4055 oversight, this default was not being set up when an address was tested by
4056 -bt or -bv, which affected the outcome if any router or filter referred to
4059 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4060 log only when "message" or "log_message" are set. However, if one of the
4061 conditions was an address verification, or a call to a nested ACL, the
4062 messages generated by the underlying test were being passed through. This
4063 no longer happens. The underlying message is available in $acl_verify_
4064 message for both "message" and "log_message" expansions, so it can be
4065 passed through if needed.
4067 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4068 new expansion $bh_ to give the encoded byte string without charset
4069 translation. Translation happens only if iconv() is available; HAVE_ICONV
4070 indicates this at build time. HEADERS_CHARSET gives the charset to
4071 translate to; headers_charset can change it in the configuration, and
4072 "headers charset" can change it in an individual filter file.
4074 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4075 that creates RFC 2047 encoded "words" labels them as that charset instead
4076 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4077 expansion operator; (ii) when Exim creates a From: line for a local
4078 message; (iii) when a header line is rewritten to include a "phrase" part.
4080 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4081 buggy, causing it to skip the first lines of messages whose message ID
4082 ended in 'D'. This would not have bitten before Exim release 4.14, because
4083 message IDs were unlikely to end in 'D' before then. The effect was to have
4084 incorrect size information for certain domains.
4086 45. #include "config.h" was missing at the start of the crypt16.c module. This
4087 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4090 46. If there was a timeout during a "random" callout check, Exim treated it as
4091 a failure of the random address, and carried on sending RSET and the real
4092 address. If the delay was just some slowness somewhere, the response to the
4093 original RCPT would be taken as a response to RSET and so on, causing
4094 mayhem of various kinds.
4096 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4097 when I implemented it. It didn't allow for the fact that some option values
4098 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4099 the right test for positive values.
4101 48. Domain names in DNS records are case-independent. Exim always looks them up
4102 in lower case. Some resolvers return domain names in exactly the case they
4103 appear in the zone file, that is, they may contain uppercase letters. Not
4104 all resolvers do this - some return always lower case. Exim was treating a
4105 change of case by a resolver as a change of domain, similar to a widening
4106 of a domain abbreviation. This triggered its re-routing code and so it was
4107 trying to route what was effectively the same domain again. This normally
4108 caused routing to fail (because the router wouldn't handle the domain
4109 twice). Now Exim checks for this case specially, and just changes the
4110 casing of the domain that it ultimately uses when it transmits the message
4113 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4116 50. If a filter generated a file delivery with a non-absolute name (possible if
4117 no home directory exists for the router), the forbid_file option was not
4120 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4121 the existing equality matching.
4123 52. Exim was using ints instead of ino_t variables in some places where it was
4124 dealing with inode numbers.
4126 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4127 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4128 called TMPDIR, and if it finds it is different, it changes its value.
4130 54. The smtp_printf() function is now made available to local_scan() so
4131 additional output lines can be written before returning. There is also an
4132 smtp_fflush() function to enable the detection of a dropped connection.
4133 The variables smtp_input and smtp_batched_input are exported to
4136 55. Changed the default runtime configuration: the message "Unknown user"
4137 has been removed from the ACL, and instead placed on the localuser router,
4138 using the cannot_route_message feature. This means that any verification
4139 failures that generate their own messages won't get overridden. Similarly,
4140 the "Unrouteable address" message that was in the ACL for unverifiable
4141 relay addresses has also been removed.
4143 56. Added hosts_avoid_esmtp to the smtp transport.
4145 57. The exicyclog script was not checking for the esoteric option
4146 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4147 will work only if exicyclog is run under the appropriate euid.
4149 58. Following a discussion on the list, the rules by which Exim recognises line
4150 endings on incoming messages have been changed. The -dropcr and drop_cr
4151 options are now no-ops, retained only for backwards compatibility. The
4152 following line terminators are recognized: LF CRLF CR. However, special
4153 processing applies to CR:
4155 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4156 nor a local message in the state where . is a terminator.
4158 (ii) If a bare CR is encountered in a header line, an extra space is added
4159 after the line terminator so as not to end the header. The reasoning
4160 behind this is that bare CRs in header lines are most likely either
4161 to be mistakes, or people trying to play silly games.
4163 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4164 was being incorrectly given as 18 bytes larger than it should have been.
4165 This is a VOB (very old bug).
4167 60. This may never have affected anything current, but just in case it has:
4168 When the local host is found other than at the start of a list of hosts,
4169 the local host, those with the same MX, and any that follow, are discarded.
4170 When the list in question was part of a longer list of hosts, the following
4171 hosts (not currently being processed) were also being discarded. This no
4172 longer happens. I'm not sure if this situation could ever has previously
4175 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4178 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4179 header. This is something that is recommended in a new Internet Draft, and
4180 is something that is documented as being done by Sendmail. There are two
4181 possible values. For messages generated by the autoreply transport, Exim
4184 Auto-Submitted: auto-replied
4186 whereas for all other generated messages (e.g. bounces) it adds
4188 Auto-Submitted: auto-generated
4190 63. The "personal" condition in filters now includes a test for the
4191 Auto-Submitted: header. If it contains the string "auto-" the message it
4192 not considered personal.
4194 64. Added rcpt_include_affixes as a generic transport option.
4196 65. Added queue_only_override (default true).
4198 66. Added the syslog_duplication option.
4200 67. If what should have been the first header line of a message consisted of
4201 a space followed by a colon, Exim was mis-interpreting it as a header line.
4202 It isn't of course - it is syntactically invalid and should therefore be
4203 treated as the start of the message body. The misbehaviour could have
4204 caused a number of strange effects, including loss of data in subsequent
4205 header lines, and spool format errors.
4207 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4208 client host had authenticated. This control can now be exercised by an ACL
4209 for more flexibility.
4211 69. By default, callouts do not happen when testing with -bh. There is now a
4212 variant, -bhc, which does actually run the callout code, including
4213 consulting and updating the callout cache.
4215 70. Added support for saslauthd authentication, courtesy of Alexander
4218 71. If statvfs() failed on the spool or log directories while checking their
4219 size for availability, Exim confusingly gave the error "space shortage".
4220 Furthermore, in debugging mode it crashed with a floating point exception.
4221 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4222 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4223 serious problem, Exim now writes to the main and panic logs when this
4224 happens, with details of the failure. It then refuses to accept the
4225 incoming message, giving the message "spool directory problem" or "log
4226 directory problem" with a 421 code for SMTP messages.
4228 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4229 0, 1, and 2 exist, because some OS complain for execs without them (see
4230 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4231 descriptors. However, the code omitted to check that the open succeeded,
4232 causing mysterious errors if for some reason the permissions on /dev/null
4233 got screwed. Now Exim writes a message to the main and panic logs, and
4234 bombs out if it can't open /dev/null.
4236 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4237 interact so that it is all more flexible. It is supposed to remain
4238 backwards compatible. Also added extra_local_interfaces.
4240 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4241 to bomb out with an assertion failure - to the client this appears as a
4242 connection drop. This problem occurs in the part of the code that was taken
4243 from the Samba project. Fortunately, the assertion is in a very simple
4244 function, so I have fixed this by reproducing the function inline in the
4245 one place where it is called, and arranging for authentication to fail
4246 instead of killing the process with assert().
4248 75. The SPA client code was not working when the server requested OEM rather
4249 than Unicode encoding.
4251 76. Added code to make require_files with a specific uid setting more usable in
4252 the case where statting the file as root fails - usually a non-root-mounted
4253 NFS file system. When this happens and the failure is EACCES, Exim now
4254 forks a subprocess and does the per-uid checking as the relevant uid.
4256 77. Added process_log_path.
4258 78. If log_file_path was not explicitly set, a setting of check_log_space or
4259 check_log_inodes was ignored.
4261 79. If a space check for the spool or log partitions fails, the incident is now
4262 logged. Of course, in the latter case the data may get lost...
4264 80. Added the %p formatting code to string_format() so that it can be used to
4265 print addresses in debug_print(). Adjusted all the address printing in the
4266 debugging in store.c to use %p rather than %d.
4268 81. There was a concern that a line of code in smtp_in.c could overflow a
4269 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4270 initially expressed, the concern was not well-founded, because trailing
4271 spaces are removed early. However, if the trailing spaces were followed by
4272 a NULL, they did not get removed, so the overflow was possible. Two fixes
4275 (a) I re-wrote the offending code in a cleaner fashion.
4276 (b) If an incoming SMTP command contains a NULL character, it is rejected
4279 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4280 runs initgroups(), so that if the Exim user is in any additional groups,
4281 they will be used during message reception.
4287 The change log for 4.20 and earlier releases has been archived.