1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.418 2006/10/30 22:22:50 tom Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
8 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
9 leftover -K file (the existence of which was triggered by #402).
10 While we were at it, introduced process PID as part of the -K
11 filename. This should rule out race conditions when creating
14 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
15 processing considerably. Previous code took too long for large mails,
16 triggering a timeout which in turn triggers #401.
18 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
19 in the DK code in transports.c. sendfile() is not really portable,
20 hence the _LINUX specificness.
22 TF/01 In the add_headers option to the mail command in an Exim filter,
23 there was a bug that Exim would claim a syntax error in any
24 header after the first one which had an odd number of characters
27 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
28 callout verification, Exim cached a "reject" for the entire domain. This
29 is correct for most verifications, but it is not correct for a recipient
30 verification with use_sender or use_postmaster set, because in that case
31 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
32 case of MAIL FROM:<> rejection from other early rejections (e.g.
33 rejection of HELO). When verifying a recipient using a non-null MAIL
34 address, the cache is ignored if it shows MAIL FROM:<> rejection.
35 Whatever the result of the callout, the value of the domain cache is
36 left unchanged (for any other kind of callout, getting as far as trying
37 RCPT means that the domain itself is ok).
39 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
42 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
43 manifest itself as EPIPE rather than ECONNECT. When tidying away a
44 session, the daemon ignores ECONNECT errors and logs others; it now
45 ignores EPIPE as well.
47 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
48 (quoted-printable decoding).
50 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
51 later the small subsequent patch to fix an introduced bug.
53 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
55 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
57 PH/08 An error is now given if message_size_limit is specified negative.
59 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
60 to be given (somewhat) arbitrary names.
62 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
65 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
66 miscellaneous code fixes
68 PH/10 Added the log_reject_target ACL modifier to specify where to log
71 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
72 hostname. This is wrong, because it relates to the incoming message (and
73 probably the interface on which it is arriving) and not to the outgoing
74 callout (which could be using a different interface). This has been
75 changed to use the value of the helo_data option from the smtp transport
76 instead - this is what is used when a message is actually being sent. If
77 there is no remote transport (possible with a router that sets up host
78 addresses), $smtp_active_hostname is used.
80 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
81 tweaks were necessary in order to get it to work (see also 21 below):
82 (a) The code assumed that strncpy() returns a negative number on buffer
83 overflow, which isn't the case. Replaced with Exim's string_format()
85 (b) There were several signed/unsigned issues. I just did the minimum
86 hacking in of casts. There is scope for a larger refactoring.
87 (c) The code used strcasecmp() which is not a standard C function.
88 Replaced with Exim's strcmpic() function.
89 (d) The code set only $1; it now sets $auth1 as well.
90 (e) A simple test gave the error "authentication client didn't specify
91 service in request". It would seem that Dovecot has changed its
92 interface. Fortunately there's a specification; I followed it and
93 changed what the client sends and it appears to be working now.
95 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
98 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
99 address A is aliased to B and C, where B exists and C does not. Without
100 -v the output is "A verified" because verification stops after a
101 successful redirection if more than one address is generated. However,
102 with -v the child addresses are also verified. Exim was outputting "A
103 failed to verify" and then showing the successful verification for C,
104 with its parentage. It now outputs "B failed to verify", showing B's
105 parentage before showing the successful verification of C.
107 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
108 look up a TXT record in a specific list after matching in a combined
111 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
112 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
113 they consult the DNS. I had assumed they would set it the way they
114 wanted; and indeed my experiments on Linux seem to show that in some
115 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
116 To be on the safe side, however, I have now made the interface to
117 host_find_byname() similar to host_find_bydns(), with an argument
118 containing the DNS resolver options. The host_find_byname() function now
119 sets these options at its start, just as host_find_bydns() does. The smtp
120 transport options dns_qualify_single and dns_search_parents are passed to
121 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
122 of host_find_byname() use the default settings of RES_DEFNAMES
123 (qualify_single) but not RES_DNSRCH (search_parents).
125 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
126 spool_read_header() do less string testing, by means of a preliminary
127 switch on the second character of optional "-foo" lines. (This is
128 overdue, caused by the large number of possibilities that now exist.
129 Originally there were few.) While I was there, I also converted the
130 str(n)cmp tests so they don't re-test the leading "-" and the first
131 character, in the hope this might squeeze out yet more improvement.
133 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
134 flag allowing group syntax was set by the header_syntax check but not
135 turned off, possible causing trouble later; (2) The flag was not being
136 set at all for the header_verify test, causing "group"-style headers to
137 be rejected. I have now set it in this case, and also caused header_
138 verify to ignore an empty address taken from a group. While doing this, I
139 came across some other cases where the code for allowing group syntax
140 while scanning a header line wasn't quite right (mostly, not resetting
141 the flag correctly in the right place). These bugs could have caused
142 trouble for malformed header lines. I hope it is now all correct.
144 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
145 with the "reply" argument non-NULL. The code, however (which originally
146 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
147 but it didn't always do it. This confused somebody who was copying the
148 code for some other use. I have removed all the tests.
150 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
151 feature that was used to support insecure browsers during the U.S. crypto
152 embargo. It requires special client support, and Exim is probably the
153 only MTA that supported it -- and would never use it because real RSA is
154 always available. This code has been removed, because it had the bad
155 effect of slowing Exim down by computing (never used) parameters for the
156 RSA_EXPORT functionality.
158 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
159 authenticator to fail if there's a tab character in the incoming data
160 (there should never be unless someone is messing about, as it's supposed
161 to be base64-encoded). Also added, on Timo's advice, the "secured" option
162 if the connection is using TLS or if the remote IP is the same as the
163 local IP, and the "valid-client-cert option" if a client certificate has
166 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
167 authenticators. This can be used for authorization after authentication
168 succeeds. (In the case of plaintext, it servers for both authentication
171 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
172 if any retry times were supplied.
174 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
175 connection, where there is no client IP address to check. In this
176 situation, the verify now always succeeds.
178 PH/25 Applied John Jetmore's -Mset patch.
180 PH/26 Added -bem to be like -Mset, but loading a message from a file.
182 PH/27 In a string expansion for a processed (not raw) header when multiple
183 headers of the same name were present, leading whitespace was being
184 removed from all of them, but trailing whitespace was being removed only
185 from the last one. Now trailing whitespace is removed from each header
186 before concatenation. Completely empty headers in a concatenation (as
189 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
190 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
192 PH/29 After an address error (typically a 4xx response from a server), Exim
193 always tries the failing address if it appears in a new message, but
194 respects the retry time otherwise. This was implemented by checking for
195 being in a queue run, which isn't quite right. Now it checks the
196 "first_delivery" flag instead.
198 PH/30 Exim was sometimes attempting to deliver messages that had suffered
199 address errors (4xx response to RCPT) over the same connection as other
200 messages routed to the same hosts. Such deliveries are always "forced",
201 so retry times are not inspected. This resulted in far too many retries
202 for the affected addresses. The effect occurred only when there were more
203 hosts than the hosts_max_try setting in the smtp transport when it had
204 the 4xx errors. Those hosts that it had tried were not added to the list
205 of hosts for which the message was waiting, so if all were tried, there
206 was no problem. Two fixes have been applied:
208 (i) If there are any address or message errors in an SMTP delivery, none
209 of the hosts (tried or untried) are now added to the list of hosts
210 for which the message is waiting, so the message should not be a
211 candidate for sending over the same connection that was used for a
212 successful delivery of some other message. This seems entirely
213 reasonable: after all the message is NOT "waiting for some host".
214 This is so "obvious" that I'm not sure why it wasn't done
215 previously. Hope I haven't missed anything, but it can't do any
216 harm, as the worst effect is to miss an optimization.
218 (ii) If, despite (i), such a delivery is accidentally attempted, the
219 routing retry time is respected, so at least it doesn't keep
220 hammering the server.
226 SC/01 Use a glob alias rather than an array ref in eximstats generated
227 parser. This improves both readability and performance.
229 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
230 Don't display local sender or destination tables in eximstats unless
231 there is data to show.
232 Added average volumes into the eximstats top table text output.
234 SC/03 Collect data on the number of addresses (recipients) as well
235 as the number of messages in eximstats.
237 TF/01 Correct an error in the documentation for the redirect router. Exim
238 does (usually) call initgroups() when daemonizing.
240 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
241 with consistent privilege compared to when running as a daemon.
243 TF/03 Note in the spec that $authenticated_id is not set for local
244 submissions from trusted users.
246 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
247 Thanks to Dean Brooks <dean@iglou.com> for the patch.
249 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
250 by adding some example configuration directives to the default
251 configuration file. A little bit of work is required to uncomment the
252 directives and define how usernames and passwords are checked, but
253 there is now a framework to start from.
255 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
256 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
257 without this. I don't know how relevant this is to other LDAP libraries.
259 PH/02 Add the verb name to the "unknown ACL verb" error.
261 PH/03 Magnus Holmgren's patch for filter_prepend_home.
263 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
265 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
266 directory not expanded when it should be if an expanded home directory
267 was set for the address (which is overridden by the transport).
269 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
272 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
273 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
274 because it is too late at that time, and has no effect.
276 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
277 security issue with \' (bugzilla #107). I could not use the
278 PQescapeStringConn() function, because it needs a PGconn value as one of
281 PH/08 When testing addresses using -bt, indicate those final addresses that
282 are duplicates that would not cause an additional delivery. At least one
283 person was confused, thinking that -bt output corresponded to deliveries.
284 (Suppressing duplicates isn't a good idea as you lose the information
285 about possibly different redirections that led to the duplicates.)
287 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
288 systems where poll() doesn't work, in particular OS X.
290 PH/10 Added more information to debugging output for retry time not reached.
292 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
293 operations in malware.c.
295 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
298 PH/13 If write_rejectlog was set false when logging was sent to syslog with
299 syslog_duplication set false, log lines that would normally be written
300 both the the main log and to the reject log were not written to syslog at
303 PH/14 In the default configuration, change the use of "message" in ACL warn
304 statements to "add_header".
306 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
307 not followed by a command (e.g. "seen endif").
309 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
310 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
313 PH/17 Added extra conditions to the default value of delay_warning_condition
317 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
318 { match{$h_precedence:}{(?i)bulk|list|junk} } \
319 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
322 The Auto-Submitted: and various List- headers are standardised, whereas I
323 don't think Precedence: ever was.
325 PH/18 Refactored debugging code in route_finduser() to show more information,
326 in particular, the error code if getpwnam() issues one.
328 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
329 This is apparently needed in addition to the PH/07 change above to avoid
330 any possible encoding problems.
332 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
333 but not after initializing Perl.
335 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
336 output them only if debugging. By default they are written stderr,
337 apparently, which is not desirable.
339 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
342 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
345 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
347 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
348 authenticated or an ident call has been made. Suppress the default
349 values for $authenticated_id and $authenticated_sender (but permit -oMai
350 and -oMas) when testing with -bh.
352 PH/24 Re-jigged the order of the tests in the default configuration so that the
353 tests for valid domains and recipients precede the DNS black list and CSA
354 tests, on the grounds that those ones are more expensive.
356 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
357 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
358 This bug exists in every version of Exim that I still have, right back to
361 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
362 However, an attempt to turn on case-sensitivity in a regex key by
363 including (?-i) didn't work because the subject string was already
364 lowercased, and the effects were non-intuitive. It turns out that a
365 one-line patch can be used to allow (?-i) to work as expected.
371 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
372 other effects) broke the use of negated acl sub-conditions.
374 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
377 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
378 "Deny" causes Exim to reject the incoming connection with a 554 error.
379 Unfortunately, if there is a major crisis, such as a disk failure,
380 tcp-wrappers gives "deny", whereas what one would like would be some
381 kind of temporary error. A kludge has been added to help with this.
382 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
383 554 error is used if errno is still zero or contains ENOENT (which occurs
384 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
387 PH/03 Add -lutil to the default FreeBSD LIBS setting.
389 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
390 errors. Otherwise a message that provokes a temporary error (when other
391 messages do not) can cause a whole host to time out.
393 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
394 addresses were routed directly to files or pipes from a redirect router.
395 File deliveries just didn't batch; pipe deliveries might have suffered
398 PH/06 A failure to get a lock for a hints database would erroneously always say
399 "Failed to get write lock", even when it was really a read lock.
401 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
402 of 0600. This has been changed to use the value of the lockfile_mode
403 option (which defaults to 0600).
405 PH/08 Applied small patch from the Sieve maintainer.
407 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
408 folder from quota calculations, a direct delivery into this folder messed
409 up the contents of the maildirsize file. This was because the regex was
410 used only to exclude .Trash (or whatever) when the size of the mailbox
411 was calculated. There was no check that a delivery was happening into an
412 excluded directory. This bug has been fixed by ignoring all quota
413 processing for deliveries into excluded directories.
415 PH/10 Added the maildirfolder_create_regex option to appendfile.
421 PH/01 The code for finding all the local interface addresses on a FreeBSD
422 system running IPv6 was broken. This may well have applied to all BSD
423 systems, as well as to others that have similar system calls. The broken
424 code found IPv4 interfaces correctly, but gave incorrect values for the
425 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
426 that it would not match correctly against @[] and not recognize the IPv6
429 PH/02 The ipliteral router was not recognizing addresses of the form user@
430 [ipv6:....] because it didn't know about the "ipv6:" prefix.
432 PH/03 Added disable_ipv6.
434 PH/04 Changed $reply_address to use the raw form of the headers instead of the
435 decoded form, because it is most often used to construct To: headers
436 lines in autoreplies, and the decoded form may well be syntactically
437 invalid. However, $reply_address has leading white space removed, and all
438 newlines turned into spaces so that the autoreply transport does not
441 PH/05 If group was specified without a user on a router, and no group or user
442 was specified on a transport, the group from the router was ignored.
444 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
445 for visible compile-time settings that can be used to change these
446 numbers, for those that want even more. Backwards compatibility with old
447 spool files has been maintained. However, going back to a previous Exim
448 release will lost any variables that are in spool files.
450 PH/07 Two small changes when running in the test harness: increase delay when
451 passing a TCP/IP connection to a new process, in case the original
452 process has to generate a bounce, and remove special handling of
453 127.0.0.2 (sic), which is no longer necessary.
455 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
456 be the same on different OS.
458 PH/09 Moved a debug statement in filter processing to avoid a race problem when
461 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
462 whether --show-vars was specified or not
464 JJ/02 exipick: Added support for new ACL variable spool format introduced
467 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
468 syntactically invalid From: or Reply-to: line, and a filter used this to
469 generate an autoreply, and therefore failed to obtain an address for the
470 autoreply, Exim could try to deliver to a non-existent relative file
471 name, causing unrelated and misleading errors. What now happens is that
472 it logs this as a hard delivery error, but does not attempt to create a
475 PH/11 The exinext utility has a -C option for testing purposes, but although
476 the given file was scanned by exinext itself; it wasn't being passed on
477 when Exim was called.
479 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
480 an end-of-file indication when reading a command response.
482 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
483 compiled. In many other places in Exim, IPv6 addresses are always
484 recognized, so I have changed this. It also means that IPv4 domain
485 literals of the form [IPV4:n.n.n.n] are now always recognized.
487 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
488 used if the router is not running as root, for example, when verifying at
489 ACL time, or when using -bh. The debugging output from this situation was
490 non-existent - all you got was a failure to exec. I have made two
493 (a) Failures to set uid/gid, the current directory, or a process leader
494 in a subprocess such as that created by queryprogram now generate
495 suitable debugging ouput when -d is set.
497 (b) The queryprogram router detects when it is not running as root,
498 outputs suitable debugging information if -d is set, and then runs
499 the subprocess without attempting to change uid/gid.
501 PH/15 Minor change to Makefile for building test_host (undocumented testing
504 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
505 additional section of a DNS packet that returns MX or SRV records.
506 Instead, it always explicitly searches for A/AAAA records. This avoids
507 major problems that occur when a DNS server includes only records of one
508 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
509 fixed another bug: if SRV records were looked up and the corresponding
510 address records were *not* found in the additional section, the port
511 values from the SRV records were lost.
513 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
514 using the correct key (the original address) when searching the retry
515 rules in order to find which one to use for generating the retry hint.
517 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
518 adding the default one. Similarly, if it contains a Reply-To: header, the
519 errors_reply_to option, if set, is not used.
521 PH/19 When calculating a retry time, Exim used to measure the "time since
522 failure" by looking at the "first failed" field in the retry record. Now
523 it does not use this if it is later than than the arrival time of the
524 message. Instead it uses the arrival time. This makes for better
525 behaviour in cases where some deliveries succeed, thus re-setting the
526 "first failed" field. An example is a quota failure for a huge message
527 when small messages continue to be delivered. Without this change, the
528 "time since failure" will always be short, possible causing more frequent
529 delivery attempts for the huge message than are intended.
530 [Note: This change was subsequently modified - see PH/04 for 4.62.]
532 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
533 $1, $2, $3) because the numerical variables can be reset during some
534 expansion items (e.g. "match"), thereby losing the authentication data.
536 PH/21 Make -bV show the size of off_t variables so that the test suite can
537 decide whether to run tests for quotas > 2G.
539 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
540 mailbox_size, and mailbox_filecount in the appendfile transport. If a
541 filecount value is greater than 2G or if a quota value is greater than 2G
542 on a system where the size of off_t is not greater than 4, a panic error
545 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
546 never match. The debug and -bh output now contains an explicit error
547 message indicating a malformed IPv4 address or mask.
549 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
550 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
553 PH/25 Do not write to syslog when running in the test harness. The only
554 occasion when this arises is a failure to open the main or panic logs
555 (for which there is an explicit test).
557 PH/26 Added the /no_tell option to "control=freeze".
559 PH/27 If a host name lookup failed very early in a connection, for example, if
560 the IP address matched host_lookup and the reverse lookup yielded a name
561 that did not have a forward lookup, an error message of the form "no IP
562 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
563 could be logged. Now it outputs the IP address instead of "NULL".
565 PH/28 An enabling patch from MH: add new function child_open_exim2() which
566 allows the sender and the authenticated sender to be set when
567 submitting a message from within Exim. Since child_open_exim() is
568 documented for local_scan(), the new function should be too.
570 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
571 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
572 results in an empty string is now treated as unset.
574 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
576 PH/31 Added sender_verify_fail logging option.
578 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
579 needed by RFC 822 but not by RFC 2822 was commented out. I have now
580 tidied the source and removed it altogether.
582 PH/33 When a queue run was abandoned because the load average was too high, a
583 log line was always written; now it is written only if the queue_run log
584 selector is set. In addition, the log line for abandonment now contains
585 information about the queue run such as the pid. This is always present
586 in "start" and "stop" lines but was omitted from the "abandon" line.
588 PH/34 Omit spaces between a header name and the colon in the error message that
589 is given when verify = headers_syntax fails (if there are lots of them,
590 the message gets confusing).
592 PH/35 Change the default for dns_check_names_pattern to allow slashes within
593 names, as there are now some PTR records that contain slashes. This check
594 is only to protect against broken name servers that fall over on strange
595 characters, so the fact that it applies to all lookups doesn't matter.
597 PH/36 Now that the new test suite is complete, we can remove some of the
598 special code in Exim that was needed for the old test suite. For example,
599 sorting DNS records because real resolvers return them in an arbitrary
600 order. The new test suite's fake resolver always returns records in the
603 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
604 bounces) except when queue_only is set, to avoid logging races between
605 the different processes.
607 PH/38 Panic-die if .include specifies a non-absolute path.
609 PH/39 A tweak to the "H" retry rule from its user.
611 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
612 a label. They prevented compilation on older perls.
614 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
615 a warning to be raised on newish perls.
617 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
618 on queue. Changes to match documented behaviour of showing count of
619 messages matching specified criteria.
621 PH/40 Changed the default ident timeout from 30s to 5s.
623 PH/41 Added support for the use of login_cap features, on those BSD systems
624 that have them, for controlling the resources used by pipe deliveries.
626 PH/42 The content-scanning code uses fopen() to create files in which to put
627 message data. Previously it was not paying any attention to the mode of
628 the files. Exim runs with umask(0) because the rest of the code creates
629 files with open(), and sets the required mode explicitly. Thus, these
630 files were ending up world-writeable. This was not a big issue, because,
631 being within the spool directory, they were not world-accessible. I have
632 created a function called modefopen, which takes an additional mode
633 argument. It sets umask(777), creates the file, chmods it to the required
634 mode, then resets the umask. All the relevant calls to fopen() in the
635 content scanning code have been changed to use this function.
637 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
638 to 24 hours. This avoids potential overflow problems when processing G
639 and H retry rules. I suspect nobody ever tinkers with this value.
641 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
643 PH/45 When the plaintext authenticator is running as a client, the server's
644 challenges are checked to ensure they are valid base64 strings. By
645 default, the authentication attempt is cancelled if an invalid string is
646 received. Setting client_ignore_invalid_base64 true ignores these errors.
647 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
648 they are received. Thus, the responses can be made to depend on the
649 challenges. If an invalid string is ignored, an empty string is placed in
652 PH/46 Messages that are created by the autoreply transport now contains a
653 References: header, in accordance with RFCs 2822 and 3834.
655 PH/47 Added authenticated_sender_force to the smtp transport.
657 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
659 PH/49 Installed latest patch from the Sieve maintainer.
661 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
662 was also set, the appendfile transport was unnecessarily scanning a
663 directory of message files (e.g. for maildir delivery) to find the count
664 of files (along with the size), even though it did not need this
665 information. It now does the scan only if it needs to find either the
666 size of the count of files.
668 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
670 PH/52 Two bugs concerned with error handling when the smtp transport is
673 (i) Exim was not creating retry information for temporary errors given
674 for individual recipients after the DATA command when the smtp transport
675 was used in LMTP mode. This meant that they could be retried too
676 frequently, and not timed out correctly.
678 (ii) Exim was setting the flag that allows error details to be returned
679 for LMTP errors on RCPT commands, but not for LMTP errors for individual
680 recipients that were returned after the DATA command.
682 PH/53 This is related to PH/52, but is more general: for any failing address,
683 when detailed error information was permitted to be returned to the
684 sender, but the error was temporary, then after the final timeout, only
685 "retry timeout exceeded" was returned. Now it returns the full error as
686 well as "retry timeout exceeded".
688 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
689 do this, and (what is worse) MTAs that accept it.
691 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
692 will now be deprecated.
694 PH/56 New os.c-cygwin from the Cygwin maintainer.
696 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
697 formats (previously only available in exim formats via -bpr, -bpru,
698 and -bpra. Now also available in native and exiqgrep formats)
700 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
701 with very large, slow to parse queues
703 JJ/08 exipick: added ! as generic prefix to negate any criteria format
705 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
707 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
708 responses to authentication challenges, though it was showing the
709 challenges; (ii) I've removed the CR characters from the debug output for
712 PH/58 Allow for the insertion of a newline as well as a space when a string
713 is turned into more than one encoded-word during RFC 2047 encoding. The
714 Sieve code now uses this.
716 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
717 data_4xx, lost_connection, tls_required.
719 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
720 message was being sent as an SMTP response.
722 PH/61 Add -l and -k options to exicyclog.
724 PH/62 When verifying, if an address was redirected to one new address, so that
725 verification continued, and the new address failed or deferred after
726 having set something in $address_data, the value of $address_data was not
727 passed back to the ACL. This was different to the case when no
728 redirection occurred. The value is now passed back in both cases.
730 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
731 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
732 use login_cap.h, so on its own it isn't the distinguishing feature. The
733 new name refers directly to the setclassresources() function.
735 PH/65 Added configuration files for NetBSD3.
737 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
739 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
740 is preferred over IPv4.
742 PH/68 The bounce_return_message and bounce_return_body options were not being
743 honoured for bounces generated during the reception of non-SMTP messages.
744 In particular, this applied to messages rejected by the ACL. This bug has
745 been fixed. However, if bounce_return_message is true and bounce_return_
746 body is false, the headers that are returned for a non-SMTP message
747 include only those that have been read before the error was detected.
748 (In the case of an ACL rejection, they have all been read.)
750 PH/69 The HTML version of the specification is now built in a directory called
751 spec_html instead of spec.html, because the latter looks like a path with
752 a MIME-type, and this confuses some software.
754 PH/70 Catch two compiler warnings in sieve.c.
756 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
757 function verify_get_ident() calls ip_connect() to connect a socket, but
758 if the "connect()" function timed out, ip_connect() used to close the
759 socket. However, verify_get_ident() also closes the socket later, and in
760 between Exim writes to the log, which may get opened at this point. When
761 the socket was closed in ip_connect(), the log could get the same file
762 descriptor number as the socket. This naturally causes chaos. The fix is
763 not to close the socket in ip_connect(); the socket should be closed by
764 the function that creates it. There was only one place in the code where
765 this was missing, in the iplookup router, which I don't think anybody now
766 uses, but I've fixed it anyway.
768 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
769 well as to direct DNS lookups. Otherwise the handling of names in host
770 lists is inconsistent and therefore confusing.
776 PH/01 Two changes to the default runtime configuration:
778 (1) Move the checks for relay_from_hosts and authenticated clients from
779 after to before the (commented out) DNS black list checks.
781 (2) Add control=submission to the relay_from_hosts and authenticated
782 clients checks, on the grounds that messages accepted by these
783 statements are most likely to be submissions.
785 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
787 (1) Generate an error if the third argument for the ${prvs expansion is
790 (2) Treat a missing third argument of ${prvscheck as if it were an empty
793 (3) Reset the variables that are obtained from the first argument of
794 ${prvscheck and used in the second argument before leaving the code,
795 because their memory is reclaimed, so using them afterwards may do
798 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
799 one (it's much easier than Tom thought :-).
801 (5) Because of (4), we can now allow for the use of $prvscheck_result
802 inside the third argument.
804 PH/03 For some reason, the default setting of PATH when running a command from
805 a pipe transport was just "/usr/bin". I have changed it to
808 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
809 anything to be listed in the output from -bV.
811 PH/05 When a filter generated an autoreply, the entire To: header line was
812 quoted in the delivery log line, like this:
814 => >A.N.Other <ano@some.domain> <original@ddress> ...
816 This has been changed so that it extracts the operative address. There
817 may be more than one such address. If so, they are comma-separated, like
820 => >ano@some.domain,ona@other.domain <original@ddress> ...
822 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
823 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
824 not being looked up in the rDNS to get a host name, Exim was showing the
825 IP address twice in Received: lines, even though the IP addresses were
826 identical. For example:
828 Received: from [1.2.3.4] (helo=[1.2.3.4])
830 However, if the real host name was known, it was omitting the HELO data
831 if it matched the actual IP address. This has been tidied up so that it
832 doesn't show the same IP address twice.
834 PH/07 When both +timestamp and +memory debugging was on, the value given by
835 $tod_xxx expansions could be wrong, because the tod_stamp() function was
836 called by the debug printing, thereby overwriting the timestamp buffer.
837 Debugging no longer uses the tod_stamp() function when +timestamp is set.
839 PH/08 When the original message was included in an autoreply transport, it
840 always said "this is a copy of the message, including all the headers",
841 even if body_only or headers_only was set. It now gives an appropriate
844 PH/09 Applied a patch from the Sieve maintainer which:
846 o fixes some comments
847 o adds the (disabled) notify extension core
848 o adds some debug output for the result of if/elsif tests
849 o points to the current vacation draft in the documentation
850 and documents the missing references header update
854 o fixes a bug in processing the envelope test (when testing
855 multiple envelope elements, the last element determinted the
858 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
859 Electronic Mail") by including:
861 Auto-submitted: auto-generated
863 in the messages that it generates (bounce messages and others, such as
864 warnings). In the case of bounce messages for non-SMTP mesages, there was
865 also a typo: it was using "Auto_submitted" (underscore instead of
866 hyphen). Since every message generated by Exim is necessarily in response
867 to another message, thes have all been changed to:
869 Auto-Submitted: auto-replied
871 in accordance with these statements in the RFC:
873 The auto-replied keyword:
875 - SHOULD be used on messages sent in direct response to another
876 message by an automatic process,
878 - MUST NOT be used on manually-generated messages,
880 - MAY be used on Delivery Status Notifications (DSNs) and Message
881 Disposition Notifications (MDNs),
883 - MUST NOT be used on messages generated by automatic or periodic
884 processes, except for messages which are automatic responses to
887 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
888 to the default Received: header definition.
890 PH/12 Added log selector acl_warn_skipped (default on).
892 PH/13 After a successful wildlsearch lookup, discard the values of numeric
893 variables because (a) they are in the wrong storage pool and (b) even if
894 they were copied, it wouldn't work properly because of the caching.
896 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
897 checking when decoding. Apparently there are clients that generate
898 overlong encoded strings. Why am I not surprised?
900 PH/15 If the first argument of "${if match_address" was not empty, but did not
901 contain an "@" character, Exim crashed. Now it writes a panic log message
902 and treats the condition as false.
904 PH/16 In autoreply, treat an empty string for "once" the same as unset.
906 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
907 extension "envelope-auth". The code is finished and in agreement with
908 other implementations, but there is no documentation so far and in fact,
909 nobody wrote the draft yet. This extension is currently #undef'ed, thus
910 not changing the active code.
912 Print executed "if" and "elsif" statements when debugging is used. This
913 helps a great deal to understand what a filter does.
915 Document more things not specified clearly in RFC3028. I had all this
916 sorted out, when out of a sudden new issues came to my mind. Oops."
918 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
921 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
922 canonical form (as documented). However, after a host name lookup from
923 the IP address, check_host() was doing a simple string comparison with
924 addresses acquired from the DNS when checking that the found name did
925 have the original IP as one of its addresses. Since any found IPv6
926 addresses are likely to be in abbreviated form, the comparison could
927 fail. Luckily, there already exists a function for doing the comparison
928 by converting both addresses to binary, so now that is used instead of
931 PH/20 There was another similar case to PH/19, when a complete host name was
932 given in a host list; looking up its IP address could give an abbreviated
933 form, whereas the current host's name might or might not be abbreviated.
934 The same fix has been applied.
940 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
941 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
944 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
945 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
947 PH/03 Typo: missing ".o" in src/pcre/Makefile.
949 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
950 header line, restrict the check to what is listed in RFCs 2369 and 2929.
951 Also, for "Auto-Submitted", treat anything other than "no" as
952 non-personal, in accordance with RFC 3834. (Previously it treated
953 anything starting "auto-" as non-personal.)
955 TF/01 The control=submission/name=... option had a problem with syntax
956 errors if the name included a slash character. The /name= option
957 now slurps the rest of the string, so it can include any characters
958 but it must come last in the list of options (after /sender_retain
961 PH/05 Some modifications to the interface to the fake nameserver for the new
969 TK/01 Added the "success_on_redirect" address verification option. See
970 NewStuff for rationale and an example.
972 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
974 PH/02 Patch to exigrep to allow it to work on syslog lines.
976 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
977 fread() to skip over the body file's header line, because in Cygwin the
978 header line is locked and is inaccessible.
980 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
981 co-exist for some time) to make it clear that it is the Exim ID that is
982 referenced, not the Message-ID: header line.
984 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
985 string_format() function, because snprintf() does not exist on all
988 PH/06 The use of forbid_filter_existstest now also locks out the use of the
989 ${stat: expansion item.
991 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
992 protocol synchronization error", to keep the pedants happy.
994 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
995 well as for IRIX systems, when gcc is being used. See the host.c source
998 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1000 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1003 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1004 transport and to the smtp transport in LMTP mode.
1006 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1008 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1009 run a filter in a subprocess. This could lead to confusion in subsequent
1010 lookups in the parent process. There should also be a search_tidyup() at
1011 the end of the subprocess.
1013 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1014 only if the host matched helo_try_verify_hosts, which caused the
1015 verification to occur when the EHLO/HELO command was issued. The ACL just
1016 tested the remembered result. Now, if a previous verification attempt has
1017 not happened, "verify = helo" does it there and then.
1019 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1021 TK/03 Fix log output including CR from clamd.
1023 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1024 exist provoked a memory error which could cause a segfault.
1026 PH/15 Installed PCRE 6.2
1028 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1030 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1031 of the problem. Specifically, suggested +O2 rather than +O1 for the
1034 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1036 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1037 record, the port was not correctly logged when the outgoing_port log
1038 selector was set (it logged the transort's default port).
1040 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1041 fallback_hosts, and "hosts" in the smtp transport.
1043 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1044 host errors such as "Connection refused".
1046 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1047 authentication with radiusclient 0.4.9:
1049 - Error returned from rc_read_config was caught wrongly
1050 - Username/password not passed on to radius server due to wrong length.
1052 The presumption is that some radiusclient API changes for 4.51/PH/17
1053 were not taken care of correctly. The code is still untested by me (my
1054 Linux distribution still has 0.3.2 of radiusclient), but it was
1055 contributed by a Radius user.
1057 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1058 expanding the "port" option of the smtp transport.
1060 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1061 while reading a MIME header. Thanks to Tom Hughes for a patch.
1063 PH/24 Include config.h inside local_scan.h so that configuration settings are
1066 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1067 that in an ACL for RCPT (for example), you can examine exactly what was
1070 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1071 commands, but it was not correctly comparing the address with the actual
1072 client host address. Thus, it would show the EHLO address in Received:
1073 header lines when this was not necessary.
1075 PH/27 Added the % operator to ${eval:}.
1077 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1078 it should be ignoring failures (because with -C, for example, it has lost
1079 privilege). It wasn't ignoring creation failures other than "already
1082 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1085 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1086 because an input line was too long, either on its own, or by virtue of
1087 too many continuations, the temporary file was not being removed, and the
1088 return code was incorrect.
1090 PH/31 Missing "BOOL" in function definition in filtertest.c.
1092 PH/32 Applied Sieve patches from the maintainer.
1094 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1096 PH/33 Added "verify = not_blind".
1098 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1099 Local/Makefile (with some defaults set). These are used in built scripts
1100 such as exicyclog, but they have never been used in the exim_install
1101 script (though there are many overriding facilities there). I have
1102 arranged that the exim_install script now takes note of these two
1105 PH/35 Installed configuration files for Dragonfly.
1107 PH/36 When a locally submitted message by a trusted user did not contain a
1108 From: header, and the sender address was obtained from -f or from an SMTP
1109 MAIL command, and the trusted user did not use -F to supply a sender
1110 name, $originator_name was incorrectly used when constructing a From:
1111 header. Furthermore, $originator_name was used for submission mode
1112 messages from external hosts without From: headers in a similar way,
1113 which is clearly wrong.
1115 PH/37 Added control=suppress_local_fixups.
1117 PH/38 When log_selector = +received_sender was set, and the addition of the
1118 sender made the log line's construction buffer exactly full, or one byte
1119 less than full, an overflow happened when the terminating "\n" was
1122 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1123 when the result of a list match is failure because a DNS lookup failed.
1125 PH/40 RM_COMMAND is now used in the building process.
1127 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1128 the "build-* directories that it finds.
1130 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1131 address in a domain literal was a prefix of an interface address.
1133 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1134 when verifying a sender address, unless rewrite_headers is false.
1136 PH/44 Wrote a long comment about why errors_to addresses are verified as
1137 recipients, not senders.
1139 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1140 the ratelimit ACL was added.
1142 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1144 PH/46 Added extra information about PostgreSQL errors to the error string.
1146 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1147 avoiding the need to install special zones in a real server. This is
1148 backwards compatible; if it can't find the fake resolver, it drops back.
1149 Thus, both old and new test suites can be run.
1151 TF/02 Added util/ratelimit.pl
1153 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1154 clock is set back in time.
1156 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1157 Candler <B.Candler@pobox.com>.
1159 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1160 OK for addresses that are the result of redirections.
1162 PH/48 A number of further additions for the benefit of the new test suite,
1163 including a fake gethostbyname() that interfaces to the fake DNS resolver
1166 TF/06 The fix for widen_domains has also been applied to qualify_single and
1167 search_parents which are the other dnslookup options that can cause
1170 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1173 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1175 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1176 service" if it encountered an MX record with an empty target hostname.
1177 The message is now "an MX or SRV record indicated no SMTP service".
1179 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1180 if the DNS of the sending site is misconfigured. This is quite a
1181 common situation. This change restores the behaviour of treating a
1182 helo verification defer as a failure.
1184 PH/51 If self=fail was set on a router, the bounce message did not include the
1185 actual error message.
1191 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1193 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1194 command itself ended in error, the underlying message about the transport
1195 filter timeout was being overwritten with the pipe command error. Now the
1196 underlying error message should be appended to the second error message.
1198 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1200 PH/02 Reduce the amount of output that "make" produces by default. Full output
1201 can still be requested.
1203 PH/03 The warning log line about a condition test deferring for a "warn" verb
1204 was being output only once per connection, rather than after each
1205 occurrence (because it was using the same function as for successful
1206 "warn" verbs). This seems wrong, so I have changed it.
1208 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1209 it should not have, which might have caused a crash in the right
1210 circumstances, but probably never did.
1212 PH/04 Installed a modified version of Tony Finch's patch to make submission
1213 mode fix the return path as well as the Sender: header line, and to
1214 add a /name= option so that you can make the user's friendly name appear
1217 TF/03 Added the control = fakedefer ACL modifier.
1219 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1220 Mark Lowes for thorough testing.
1222 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1224 TK/03 Merged latest SRS patch from Miles Wilton.
1226 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1227 with the definition in sysexits.h (which is #included earlier).
1228 Fortunately, Exim does not actually use EX_OK. The code used to try to
1229 preserve the sysexits.h value, by assumimg that macro definitions were
1230 scanned for macro replacements. I have been disabused of this notion,
1231 so now the code just undefines EX_OK before #including unistd.h.
1233 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1234 in the smtp transport. When a block could not be written in a single
1235 write() function, the timeout was being re-applied to each part-write.
1236 This seems wrong - if the receiver was accepting one byte at a time it
1237 would take for ever. The timeout is now adjusted when this happens. It
1238 doesn't have to be particularly precise.
1240 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1241 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1243 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1244 without a domain if the check to <postmaster@domain> fails.
1246 SC/01 Eximstats: added -xls and the ability to specify output files
1247 (patch written by Frank Heydlauf).
1249 SC/02 Eximstats: use FileHandles for outputing results.
1251 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1253 SC/04 Eximstats: fixed display of large numbers with -nvr option
1255 SC/05 Eximstats: fixed merging of reports with empty tables.
1257 SC/06 Eximstats: added the -include_original_destination flag
1259 SC/07 Eximstats: removed tabs and trailing whitespace.
1261 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1263 TK/06 MBOX spool code: Add real "From " MBOX separator line
1264 so the .eml file is really in mbox format (even though
1265 most programs do not really care). Patch from Alex Miller.
1267 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1268 The latter is generated from $received_to and is only set if the
1269 message has one envelope recipient. SA can use these headers,
1270 obviously out-of-the-box. Patch from Alex Miller.
1272 PH/08 The ${def test on a variable was returning false if the variable's
1273 value was "0", contrary to what the specification has always said!
1274 The result should be true unless the variable is empty.
1276 PH/09 The syntax error of a character other than { following "${if
1277 def:variable_name" (after optional whitespace) was not being diagnosed.
1278 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1279 accidental colon was present, for example, could give incorrect results.
1281 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1282 result is used (not including appendfile, where other changes are about
1285 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1286 This involved changing a lot of size variables from int to off_t. It
1287 should work with maildirs and everything.
1289 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1290 spamd dying while we are connected to it.
1292 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1295 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1296 function for BDB 4.3.
1298 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1300 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1301 the sources for PCRE have been re-arranged and more files are now
1304 PH/15 The code I had for printing potentially long long variables in PH/11
1305 above was not the best (it lost precision). The length of off_t variables
1306 is now inspected at build time, and an appropriate printing format (%ld
1307 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1308 to be "long long int" or "long int". This is needed for the internal
1309 formatting function string_vformat().
1311 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1312 the configuration file to be ":syslog", then the script "guesses" where
1313 the logs files are, rather than using the compiled in default. In our
1314 case the guess is not the same as the compiled default, so the script
1315 suddenly stopped working when I started to use syslog. The patch checks
1316 to see if log_file_path is "". If so, it attempts to read it from exim
1317 with no configuration file to get the compiled in version, before it
1318 falls back to the previous guessing code."
1320 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1321 implementing BATV in an Exim configuration. See NewStuff for the gory
1324 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1325 Makefile that are specific to HP-UX.
1327 PH/18 If the "use_postmaster" option was set for a recipient callout together
1328 with the "random" option, the postmaster address was used as the MAIL
1329 FROM address for the random test, but not for the subsequent recipient
1330 test. It is now used for both.
1332 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1333 patch removes a few documentation additions to RFC 3028, because the
1334 latest draft now contains them. It adds the new en;ascii-case comparator
1335 and a new error check for 8bit text in MIME parts. Comparator and
1336 require names are now matched exactly. I enabled the subaddress
1337 extension, but it is not well tested yet (read: it works for me)."
1339 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1340 rework some of the code of TK/09 above to avoid the hardwired use of
1341 "%lld" and "long long". Replaced the call to snprintf() with a call to
1344 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1345 records point to non-existent hosts", "retry timeout exceeded", and
1346 "retry time not reached for any host after a long failure period".
1348 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1349 experimental DomainKeys support:
1351 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1352 (2) On an error such as an illegally used "control", the wrong name for
1353 the control was given.
1355 These problems did NOT occur unless DomainKeys support was compiled.
1357 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1359 PH/24 Added ${if match_ip condition.
1361 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1362 databases so that it will be absolutely obvious if a crash occurs in the
1363 DB library. This is a regular occurrence (often caused by mis-matched
1366 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1367 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1368 picked up on a user's system that detects such things. There doesn't seem
1369 to be a gcc warning option for this - only an attribute that has to be
1370 put on the function's prototype. It seems that in Fedora Core 4 they have
1371 set this on a number of new functions. No doubt there will be more in due
1374 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1375 specify a transport. However, if an address that was verified by such a
1376 router was the subject of a callout, Exim crashed because it tried to
1377 read the rcpt_include_affixes from the non-existent transport. Now it
1378 just assumes that the setting of that option is false. This bug was
1379 introduced by 4.51/PH/31.
1381 PH/28 Changed -d+all to exclude +memory, because that information is very
1382 rarely of interest, but it makes the output a lot bigger. People tend to
1383 do -d+all out of habit.
1385 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1386 code in os-type was giving problems when libc.so lives in lib64, like on
1389 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1390 aren't the modern standard, and it seems that some systems' include files
1391 don't always have them. Exim was already checking for some of the newer
1392 ones like T_AAAA, and defining it itself. I've added checks for all the
1393 record types that Exim uses.
1395 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1396 not automatically generating a new one, as it is supposed to. This
1397 prevented TLS from working. If the file did exist, but contained invalid
1398 data, a new version was generated, as expected. It was only the case of a
1399 non-existent file that was broken.
1401 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1402 with a change in libdomainkeys > 0.64.
1404 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1405 from DNS. If the selector record carries the flag, it now has
1406 precedence over the domain-wide flag.
1408 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1410 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1411 the use of a port name that isn't defined in /etc/services) occurred, the
1412 message was deferred as in a normal delivery, and thus remained on the
1413 spool, instead of being failed because of the mua_wrapper setting. This
1414 is now fixed, and I tidied up some of the mua_wrapper messages at the
1417 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1418 the messages in a hash of arrays rather than using individual hashes.
1419 This is a bit cleaner and results in dramatic memory savings, albeit
1420 at a slight CPU cost.
1422 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1423 as requested by Marc Sherman.
1425 SC/10 Eximstats: added histograms for user specified patterns as requested
1428 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1430 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1431 fopen() in the content-scanning modules that did not already have it.
1437 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1438 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1440 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1442 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1444 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1445 to test Sieve filters that use "vacation".
1447 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1448 that changes the way the GnuTLS parameters are stored in the cache file.
1449 The new format can be generated externally. For backward compatibility,
1450 if the data in the cache doesn't make sense, Exim assumes it has read an
1451 old-format file, and it generates new data and writes a new file. This
1452 means that you can't go back to an older release without removing the
1455 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1456 work if there are any delivery delays because "one_time" forces the
1457 parent to be marked "delivered", so its unseen clone is never tried
1458 again. For this reason, Exim now forbids the simultaneous setting of
1461 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1462 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1463 entry if you want to know the details. The fix, however, neglected to
1464 consider the case when local delivery batching is involved. The test for
1465 "previously delivered" was not happening when checking to see if an
1466 address could be batched with a previous (undelivered) one; under
1467 certain circumstances this could lead to multiple deliveries to the same
1470 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1471 in its include files, and this causes problems building Exim.
1473 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1474 header_syntax) but Exim was just ignoring anything given after a slash.
1475 In particular, this caused confusion with an attempt to use "verify =
1476 reverse_host_lookup/defer_ok". An error is now given when options are
1477 supplied for verify items that do not have them. (Maybe reverse_host_
1478 lookup should have a defer_ok option, but that's a different point.)
1480 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1481 defined by RFC 821) to 2048, because there were problems with some AUTH
1482 commands, and RFC 1869 says the size should be increased for extended
1483 SMTP commands that take arguments.
1485 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1488 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1489 "unknown" error; now it says that the functionality isn't in the binary.
1491 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1492 an address' error message when a string expansion fails (syntax or
1493 whatever). Otherwise the password may appear in the log. Following change
1494 PH/42 below, there is no longer a chance of it appearing in a bounce
1497 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1499 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1500 of its addresses was ever tried. (Bugzilla bug #2.)
1502 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1503 the result incorrectly in the debug output. (It correctly added a newline
1504 to what was transported.)
1506 TF/01 Added $received_time.
1508 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1509 commented out examples of how to interface to a virus scanner and to
1510 SpamAssassin. Also added commented examples of av_scanner and
1511 spamd_address settings.
1513 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1514 and controls are allowed in which ACLs. There were a couple of minor
1515 errors. Some of the entries in the conditions table (which is a table of
1516 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1517 negation of where the condition IS allowed.
1519 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1521 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1522 header file does not have a version number, so I've had to invent a new
1523 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1524 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1525 radiusclient), but it was contributed by a Radius user.
1527 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1528 files or directories, for OpenSSL.
1530 PH/19 When an Exim process that is running as root has to create an Exim log
1531 file, it does so in a subprocess that runs as exim:exim so as to get the
1532 ownership right at creation (otherwise, other Exim processes might see
1533 the file with the wrong ownership). There was no test for failure of this
1534 fork() call, which would lead to the process getting stuck as it waited
1535 for a non-existent subprocess. Forks do occasionally fail when resources
1536 run out. I reviewed all the other calls to fork(); they all seem to check
1539 PH/20 When checking for unexpected SMTP input at connect time (before writing
1540 the banner), Exim was not dealing correctly with a non-positive return
1541 from the read() function. If the client had disconnected by this time,
1542 the result was a log entry for a synchronization error with an empty
1543 string after "input=" when read() returned zero. If read() returned -1
1544 (an event I could not check), uninitialized data bytes were printed.
1545 There were reports of junk text (parts of files, etc) appearing after
1548 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1550 PH/22 Added support for macro redefinition, and (re)definition in between
1551 driver and ACL definitions.
1553 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1554 forgetting to use the resulting value; it was using the unexpanded value.
1556 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1557 hadn't been configured. The fix is from Juergen Kreileder, who
1558 understands it better than I do:
1560 "Here's what I see happening with three configured cyrus_sasl
1561 authenticators configured (plain, login, cram-md5):
1563 On startup auth_cyrus_sasl_init() gets called for each of these.
1564 This means three calls to sasl_listmech() without a specified mech_list.
1565 => SASL tests which mechs of all available mechs actually work
1566 => three warnings about OTP not working
1567 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1569 With the patch, sasl_listmech() also gets called three times. But now
1570 SASL's mech_list option is set to the server_mech specified in the the
1571 authenticator. Or in other words, the answer from sasl_listmech()
1572 gets limited to just the mech you're testing for (which is different
1574 => the return list contains just 'plain' or 'login', 'cram-md5' or
1575 nothing depending on the value of ob->server_mech.
1577 I've just tested the patch: Authentication still works fine,
1578 unavailable mechs specified in the exim configuration are still
1579 caught, and the auth.log warnings about OTP are gone."
1581 PH/25 When debugging is enabled, the contents of the command line are added
1582 to the debugging output, even when log_selector=+arguments is not
1585 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1586 answer is "GNU", and only if the return is "GNU/something" is the answer
1589 PH/27 $acl_verify_message is now set immediately after the failure of a
1590 verification in an ACL, and so is available in subsequent modifiers. In
1591 particular, the message can be preserved by coding like this:
1593 warn !verify = sender
1594 set acl_m0 = $acl_verify_message
1596 Previously, $acl_verify_message was set only while expanding "message"
1597 and "log_message" when a very denied access.
1599 PH/28 Modified OS/os.c-Linux with
1601 -#ifndef OS_LOAD_AVERAGE
1602 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1604 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1605 nomenclature these days.)
1607 PH/29 Installed patch from the Sieve maintainer that adds the options
1608 sieve_useraddress and sieve_subaddress to the redirect router.
1610 PH/30 In these circumstances:
1611 . Two addresses routed to the same list of hosts;
1612 . First host does not offer TLS;
1613 . First host accepts first address;
1614 . First host gives temporary error to second address;
1615 . Second host offers TLS and a TLS session is established;
1616 . Second host accepts second address.
1617 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1618 and peerdn, if requested) that were in fact used only for the second
1621 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1622 attention to any local part prefix or suffix that was matched by the
1623 router that accepted the address. It now behaves in the same way as it
1624 does for delivery: the affixes are removed from the local part unless
1625 rcpt_include_affixes is set on the transport.
1627 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1628 timeout during the DATA phase of an incoming message.
1630 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1631 applied a patch sanctioned by the Sieve maintainer.
1633 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1634 the uid or gid is negative. A case of a negative gid caused this to be
1635 noticed. The fix allows for either to be negative.
1637 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1638 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1641 PH/36 The change PH/12 above was broken. Fixed it.
1643 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1644 the grounds that routing the same address twice would always produce the
1645 same answer. This might have been true once, but it is certainly no
1646 longer true now. Routing a child address may depend on the previous
1647 routing that produced that child. Some complicated redirection strategies
1648 went wrong when messages had multiple recipients, and made Exim's
1649 behaviour dependent on the order in which the addresses were given.
1651 I have moved the duplicate checking until after the routing is complete.
1652 Exim scans the addresses that are assigned to local and remote
1653 transports, and removes any duplicates. This means that more work will be
1654 done, as duplicates will always all be routed, but duplicates are
1655 presumably rare, so I don't expect this is of any significance.
1657 For deliveries to pipes, files, and autoreplies, the duplicate checking
1658 still happens during the routing process, since they are not going to be
1661 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1662 It corrects a timeout issue with spamd. This is Ian's comment: "The
1663 background is that sometimes spamd either never reads data from a
1664 connection it has accepted, or it never writes response data. The exiscan
1665 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1666 blindly assumes that writes won't block so it may never time out."
1668 PH/39 Allow G after quota size as well as K and M.
1670 PH/40 The value set for $authenticated_id in an authenticator may not contain
1671 binary zeroes or newlines because the value is written to log lines and
1672 to spool files. There was no check on this. Now the value is run through
1673 the string_printing() function so that such characters are converted to
1674 printable escape sequences.
1676 PH/41 $message_linecount is a new variable that contains the total number of
1677 lines in the message. Compare $body_linecount, which is the count for the
1680 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1681 bounce and delay warning messages, except in certain special cases, which
1684 (a) An SMTP error message from a remote host;
1685 (b) A message specified in a :fail: redirection;
1686 (c) A message specified in a "fail" command in a system filter;
1687 (d) A message specified in a FAIL return from the queryprogram router;
1688 (e) A message specified by the cannot_route_message router option.
1690 In these cases only, Exim does include the error details in bounce and
1691 warning messages. There are also a few cases where bland messages such
1692 as "unrouteable address" or "local delivery error" are given.
1694 PH/43 $value is now also set for the "else" part of a ${run expansion.
1696 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1697 being worked on, but at least Exim now implements the latest version to
1700 PH/45 In a pipe transport, although a timeout while waiting for the pipe
1701 process to complete was treated as a delivery failure, a timeout while
1702 writing the message to the pipe was logged, but erroneously treated as a
1703 successful delivery. Such timeouts include transport filter timeouts. For
1704 consistency with the overall process timeout, these timeouts are now
1705 treated as errors, giving rise to delivery failures by default. However,
1706 there is now a new Boolean option for the pipe transport called
1707 timeout_defer, which, if set TRUE, converts the failures into defers for
1708 both kinds of timeout. A transport filter timeout is now identified in
1711 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1712 systems where "make" and "gmake" are different, calling "gmake" at top
1713 level broke things. I've arranged for the value of $(MAKE) to be passed
1714 from the Makefile to this script so that it can call the same version of
1718 A note about Exim versions 4.44 and 4.50
1719 ----------------------------------------
1721 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1722 changes of various kinds. As a consequence, a big documentation update was
1723 needed. This delayed the release for rather longer than seemed good, especially
1724 in the light of a couple of (minor) security issues. Therefore, the changes
1725 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1726 release. So 4.44 and 4.50 are in effect two different branches that both start
1729 I have left the 4.50 change log unchanged; it contains all the changes since
1730 4.43. The change log for 4.44 is below; many of its items are identical to
1731 those for 4.50. This seems to be the most sensible way to preserve the
1732 historical information.
1738 1. Minor wording change to the doc/README.SIEVE file.
1740 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1741 computation of the current number of files was incorrect.
1743 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1744 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1747 4. Give more explanation in the error message when the command for a transport
1748 filter fails to execute.
1750 5. There are several places where Exim runs a non-Exim command in a
1751 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1752 was being done only for the command run by the queryprogram router. It is
1753 now done for all such subprocesses. The other cases are: ${run, transport
1754 filters, and the commands run by the lmtp and pipe transports.
1756 6. Added CONFIGURE_GROUP build-time option.
1758 7. Some older OS have a limit of 256 on the maximum number of file
1759 descriptors. Exim was using setrlimit() to set 1000 as a large value
1760 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1761 systems. I've change it so that if it can't get 1000, it tries for 256.
1763 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1764 was an oversight, and furthermore, ever since the addition of extra
1765 controls (e.g. 4.43/32), the checks on when to allow different forms of
1766 "control" were broken. There should now be diagnostics for all cases when a
1767 control that does not make sense is encountered.
1769 9. Added the /retain_sender option to "control=submission".
1771 10. $recipients is now available in the predata ACL (oversight).
1773 11. Tidy the search cache before the fork to do a delivery from a message
1774 received from the command line. Otherwise the child will trigger a lookup
1775 failure and thereby defer the delivery if it tries to use (for example) a
1776 cached ldap connection that the parent has called unbind on.
1778 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1779 of $address_data from the recipient verification was clobbered by the
1780 sender verification.
1782 13. The value of address_data from a sender verification is now available in
1783 $sender_address_data in subsequent conditions in the ACL statement.
1785 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1787 15. Added a new option "connect=<time>" to callout options, to set a different
1790 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1791 was its contents. (It was OK if the option was not defined at all.)
1793 17. A "Completed" log line is now written for messages that are removed from
1794 the spool by the -Mrm option.
1796 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1797 information about exactly what failed.
1799 19. Added -dd to debug only the daemon process.
1801 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1802 handles timeouts, both on the server side and network timeouts. Renamed the
1803 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1805 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1806 It is now set to "smtps".
1808 22. $host_address is now set to the target address during the checking of
1809 ignore_target_hosts.
1811 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1812 being passed; this would have caused $sender_host_name to have been used if
1813 matching the list had actually called for a host name (not very likely,
1814 since this list is usually IP addresses). A host name is now passed as
1817 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1818 code that specifies a non-restarting handler (typically sigaction() in
1819 modern systems) in an attempt to fix a rare and obscure crash bug.
1821 25. Narrowed the window for a race in the daemon that could cause it to ignore
1822 SIGCHLD signals. This is not a major problem, because they are used only to
1823 wake it up if nothing else does.
1825 26. A malformed maildirsize file could cause Exim to calculate negative values
1826 for the mailbox size or file count. Odd effects could occur as a result.
1827 The maildirsize information is now recalculated if the size or filecount
1830 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1831 support for a long time. Removed HAVE_SYS_VFS_H.
1833 28. Installed the latest version of exipick from John Jetmore.
1835 29. In an address list, if the pattern was not a regular expression, an empty
1836 subject address (from a bounce message) matched only if the pattern was an
1837 empty string. Non-empty patterns were not even tested. This was the wrong
1838 because it is perfectly reasonable to use an empty address as part of a
1839 database query. An empty address is now tested by patterns that are
1840 lookups. However, all the other forms of pattern expect the subject to
1841 contain a local part and a domain, and therefore, for them, an empty
1842 address still always fails if the pattern is not itself empty.
1844 30. Exim went into a mad DNS loop when attempting to do a callout where the
1845 host was specified on an smtp transport, and looking it up yielded more
1846 than one IP address.
1848 31. Re-factored the code for checking spool and log partition space into a
1849 function that finds that data and another that does the check. The former
1850 is then used to implement four new variables: $spool_space, $log_space,
1851 $spool_inodes, and $log_inodes.
1853 32. The RFC2047 encoding function was originally intended for short strings
1854 such as real names; it was not keeping to the 75-character limit for
1855 encoded words that the RFC imposes. It now respects the limit, and
1856 generates multiple encoded words if necessary. To be on the safe side, I
1857 have increased the buffer size for the ${rfc2047: expansion operator from
1860 33. It is now permitted to omit both strings after an "if" condition; if the
1861 condition is true, the result is "true". As before, when the second string
1862 is omitted, a false condition yields an empty string. This makes it less
1863 cumbersome to write custom ACL and router conditions.
1865 34. Failure to deliver a bounce message always caused it to be frozen, even if
1866 there was an errors_to setting on the router. The errors_to setting is now
1869 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1870 canonical form (fully expanded) before being placed in
1871 $sender_host_address.
1873 36. The table in the code that translates DNS record types into text (T_A to
1874 "A" for instance) was missing entries for NS and CNAME. It is just possible
1875 that this could have caused confusion if both these types were looked up
1876 for the same domain, because the text type is used as part of Exim's
1877 per-process caching. But the chance of anyone hitting this buglet seems
1880 37. The dnsdb lookup has been extended in a number of ways.
1882 (1) There is a new type, "zns", which walks up the domain tree until it
1883 finds some nameserver records. It should be used with care.
1885 (2) There is a new type, "mxh", which is like "mx" except that it returns
1886 just the host names, not the priorities.
1888 (3) It is now possible to give a list of domains (or IP addresses) to be
1889 looked up. The behaviour when one of the lookups defers can be
1890 controlled by a keyword.
1892 (4) It is now possible to specify the separator character for use when
1893 multiple records are returned.
1895 38. The dnslists ACL condition has been extended: it is now possible to supply
1896 a list of IP addresses and/or domains to be looked up in a particular DNS
1899 39. Added log_selector=+queue_time_overall.
1901 40. When running the queue in the test harness, wait just a tad after forking a
1902 delivery process, to get repeatability of debugging output.
1904 41. Include certificate and key file names in error message when GnuTLS fails
1905 to set them up, because the GnuTLS error message doesn't include the name
1906 of the failing file when there is a problem reading it.
1908 42. Allow both -bf and -bF in the same test run.
1910 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1912 44. The "Exiscan patch" is now merged into the mainline Exim source.
1914 45. Sometimes the final signoff response after QUIT could fail to get
1915 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1916 < 0 before doing a fflush(). This bug looks as though it goes back to the
1917 introduction of TLS in release 3.20, but "sometimes" must have been rare
1918 because the tests only now provoked it.
1920 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1921 (this can affect the format of dates).
1923 47. exim_tidydb, when checking for the continued existence of a message for
1924 which it has found a message-specific retry record, was not finding
1925 messages that were in split spool directories. Consequently, it was
1926 deleting retry records that should have stayed in existence.
1928 48. Steve fixed some bugs in eximstats.
1930 49. The SPA authentication driver was not abandoning authentication and moving
1931 on to the next authenticator when an expansion was forced to fail,
1932 contradicting the general specification for all authenticators. Instead it
1933 was generating a temporary error. It now behaves as specified.
1935 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1936 (the order specifies the preference for clients). The order is now AES256,
1937 AES128, 3DES, ARCFOUR128.
1939 51. Small patch to Sieve code - explicitly set From: when generating an
1942 52. Exim crashed if a remote delivery caused a very long error message to be
1943 recorded - for instance if somebody sent an entire SpamAssassin report back
1944 as a large number of 550 error lines. This bug was coincidentally fixed by
1945 increasing the size of one of Exim's internal buffers (big_buffer) that
1946 happened as part of the Exiscan merge. However, to be on the safe side, I
1947 have made the code more robust (and fixed the comments that describe what
1950 53. Now that there can be additional text after "Completed" in log lines (if
1951 the queue_time_overall log selector is set), a one-byte patch to exigrep
1952 was needed to allow it to recognize "Completed" as not the last thing in
1955 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1956 patch that reportedly fixes this has been added. I am not expert enough to
1957 create a test for it. This is what the patch creator wrote:
1959 "I found a little strange behaviour of ldap code when working with
1960 Windows 2003 AD Domain, where users was placed in more than one
1961 Organization Units. When I tried to give exim partial DN, the exit code
1962 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1963 But simultaneously result of request was absolutely normal ldap result,
1964 so I produce this patch..."
1966 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1967 REFERENCE, so I have modified the code to exclude the patch when that macro
1970 55. Some experimental protocols are using DNS PTR records for new purposes. The
1971 keys for these records are domain names, not reversed IP addresses. The
1972 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1973 leaves it alone. Component reversal etc. now happens only for IP addresses.
1976 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1978 57. Double the size of the debug message buffer (to 2048) so that more of very
1979 long debug lines gets shown.
1981 58. The exicyclog utility now does better if the number of log files to keep
1982 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1984 59. Two changes related to the smtp_active_hostname option:
1986 (1) $smtp_active_hostname is now available as a variable.
1987 (2) The default for smtp_banner uses $smtp_active_hostname instead
1988 of $primary_hostname.
1990 60. The host_aton() function is supposed to be passed a string that is known
1991 to be a valid IP address. However, in the case of IPv6 addresses, it was
1992 not checking this. This is a hostage to fortune. Exim now panics and dies
1993 if the condition is not met. A case was found where this could be provoked
1994 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1995 components; fortuitously, this particular loophole had already been fixed
1996 by change 4.50/55 above.
1998 If there are any other similar loopholes, the new check in host_aton()
1999 itself should stop them being exploited. The report I received stated that
2000 data on the command line could provoke the exploit when Exim was running as
2001 exim, but did not say which command line option was involved. All I could
2002 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2003 running as the user.
2006 61. There was a buffer overflow vulnerability in the SPA authentication code
2007 (which came originally from the Samba project). I have added a test to the
2008 spa_base64_to_bits() function which I hope fixes it.
2011 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2012 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2014 63. The daemon start-up calls getloadavg() while still root for those OS that
2015 need the first call to be done as root, but it missed one case: when
2016 deliver_queue_load_max is set with deliver_drop_privilege. This is
2017 necessary for the benefit of the queue runner, because there is no re-exec
2018 when deliver_drop_privilege is set.
2020 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2021 This has been fixed.
2023 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2024 was in use, was not putting the data itself into the right store pool;
2025 consequently, it could be overwritten for a subsequent message in the same
2026 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2029 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2031 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2032 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2033 the function were treating the return as a boolean value, which happened to
2034 work because 0=false and not-0=true, but is not correct code.
2036 68. The host_aton() function was not handling scoped IPv6 addresses (those
2037 with, for example, "%eth0" on the end) correctly.
2039 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2040 negated items (that is, ~something) in unsigned ints. Some compilers
2041 apparently mutter when there is no cast.
2043 70. If an address verification called from an ACL failed, and did not produce a
2044 user-specific message (i.e. there was only a "system" message), nothing was
2045 put in $acl_verify_message. In this situation, it now puts the system
2048 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2049 session; change 4.31/43 added the unwanted input to the log line - except
2050 that it did not do this in the start of session case. It now does.
2052 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2053 This is wrong and can cause the other end to generate a synchronization
2054 error if it is another Exim or anything else that does the synchronization
2055 check. A QUIT command is no longer sent after a timeout.
2057 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2058 during host lookups.
2060 74. The defer_ok option of callout verification was not working if it was used
2061 when verifying addresses in header lines, that is, for this case:
2063 verify = header_sender/callout=defer_ok
2065 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2066 those file descriptors could be used for SMTP connections. If anything
2067 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2068 could be sent to the SMTP client, causing chaos. The daemon now opens
2069 stdin, stdout, and stderr to /dev/null when it puts itself into the
2072 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2073 log by default. The user can override this with suitable Perl magic.
2075 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2076 the log message when discard triggers, was not working for the DATA ACL or
2077 for the non-SMTP ACL.
2079 78. Error message wording change in sieve.c.
2081 79. If smtp_accept_max_per_host was set, the number of connections could be
2082 restricted to fewer than expected, because the daemon was trying to set up
2083 a new connection before checking whether the processes handling previous
2084 connections had finished. The check for completed processes is now done
2085 earlier. On busy systems, this bug wouldn't be noticed because something
2086 else would have woken the daemon, and it would have reaped the completed
2089 80. If a message was submitted locally by a user whose login name contained one
2090 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2091 It caused a spool format error. I have fixed the spool reading code. A
2092 related problem was that the "from" clause in the Received: line became
2093 illegal because of the space(s). It is now covered by ${quote_local_part.
2095 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2098 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2100 83. Patch from Sieve maintainer to fix unterminated string problem in
2101 "vacation" handling.
2103 84. Some minor changes to the Linux configuration files to help with other
2104 OS variants using glibc.
2106 85. One more patch for Sieve to update vacation handling to latest spec.
2109 ----------------------------------------------------
2110 See the note above about the 4.44 and 4.50 releases.
2111 ----------------------------------------------------
2117 1. Change 4.43/35 introduced a bug that caused file counts to be
2118 incorrectly computed when quota_filecount was set in an appendfile
2121 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2122 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2125 3. Give more explanation in the error message when the command for a transport
2126 filter fails to execute.
2128 4. There are several places where Exim runs a non-Exim command in a
2129 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2130 was being done only for the command run by the queryprogram router. It is
2131 now done for all such subprocesses. The other cases are: ${run, transport
2132 filters, and the commands run by the lmtp and pipe transports.
2134 5. Some older OS have a limit of 256 on the maximum number of file
2135 descriptors. Exim was using setrlimit() to set 1000 as a large value
2136 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2137 systems. I've change it so that if it can't get 1000, it tries for 256.
2139 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2140 was an oversight, and furthermore, ever since the addition of extra
2141 controls (e.g. 4.43/32), the checks on when to allow different forms of
2142 "control" were broken. There should now be diagnostics for all cases when a
2143 control that does not make sense is encountered.
2145 7. $recipients is now available in the predata ACL (oversight).
2147 8. Tidy the search cache before the fork to do a delivery from a message
2148 received from the command line. Otherwise the child will trigger a lookup
2149 failure and thereby defer the delivery if it tries to use (for example) a
2150 cached ldap connection that the parent has called unbind on.
2152 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2153 of $address_data from the recipient verification was clobbered by the
2154 sender verification.
2156 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2157 was its contents. (It was OK if the option was not defined at all.)
2159 11. A "Completed" log line is now written for messages that are removed from
2160 the spool by the -Mrm option.
2162 12. $host_address is now set to the target address during the checking of
2163 ignore_target_hosts.
2165 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2166 being passed; this would have caused $sender_host_name to have been used if
2167 matching the list had actually called for a host name (not very likely,
2168 since this list is usually IP addresses). A host name is now passed as
2171 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2172 code that specifies a non-restarting handler (typically sigaction() in
2173 modern systems) in an attempt to fix a rare and obscure crash bug.
2175 15. Narrowed the window for a race in the daemon that could cause it to ignore
2176 SIGCHLD signals. This is not a major problem, because they are used only to
2177 wake it up if nothing else does.
2179 16. A malformed maildirsize file could cause Exim to calculate negative values
2180 for the mailbox size or file count. Odd effects could occur as a result.
2181 The maildirsize information is now recalculated if the size or filecount
2184 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2185 support for a long time. Removed HAVE_SYS_VFS_H.
2187 18. Updated exipick to current release from John Jetmore.
2189 19. Allow an empty sender to be matched against a lookup in an address list.
2190 Previously the only cases considered were a regular expression, or an
2193 20. Exim went into a mad DNS lookup loop when doing a callout where the
2194 host was specified on the transport, if the DNS lookup yielded more than
2197 21. The RFC2047 encoding function was originally intended for short strings
2198 such as real names; it was not keeping to the 75-character limit for
2199 encoded words that the RFC imposes. It now respects the limit, and
2200 generates multiple encoded words if necessary. To be on the safe side, I
2201 have increased the buffer size for the ${rfc2047: expansion operator from
2204 22. Failure to deliver a bounce message always caused it to be frozen, even if
2205 there was an errors_to setting on the router. The errors_to setting is now
2208 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2209 canonical form (fully expanded) before being placed in
2210 $sender_host_address.
2212 24. Updated eximstats to version 1.33
2214 25. Include certificate and key file names in error message when GnuTLS fails
2215 to set them up, because the GnuTLS error message doesn't include the name
2216 of the failing file when there is a problem reading it.
2218 26. Expand error message when OpenSSL has problems setting up cert/key files.
2221 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2222 (this can affect the format of dates).
2224 28. exim_tidydb, when checking for the continued existence of a message for
2225 which it has found a message-specific retry record, was not finding
2226 messages that were in split spool directories. Consequently, it was
2227 deleting retry records that should have stayed in existence.
2229 29. eximstats updated to version 1.35
2230 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2231 1.35 - bugfix such that pie charts by volume are generated correctly
2233 30. The SPA authentication driver was not abandoning authentication and moving
2234 on to the next authenticator when an expansion was forced to fail,
2235 contradicting the general specification for all authenticators. Instead it
2236 was generating a temporary error. It now behaves as specified.
2238 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2239 (the order specifies the preference for clients). The order is now AES256,
2240 AES128, 3DES, ARCFOUR128.
2242 31. Small patch to Sieve code - explicitly set From: when generating an
2245 32. Exim crashed if a remote delivery caused a very long error message to be
2246 recorded - for instance if somebody sent an entire SpamAssassin report back
2247 as a large number of 550 error lines. This bug was coincidentally fixed by
2248 increasing the size of one of Exim's internal buffers (big_buffer) that
2249 happened as part of the Exiscan merge. However, to be on the safe side, I
2250 have made the code more robust (and fixed the comments that describe what
2253 33. Some experimental protocols are using DNS PTR records for new purposes. The
2254 keys for these records are domain names, not reversed IP addresses. The
2255 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2256 leaves it alone. Component reversal etc. now happens only for IP addresses.
2259 34. The host_aton() function is supposed to be passed a string that is known
2260 to be a valid IP address. However, in the case of IPv6 addresses, it was
2261 not checking this. This is a hostage to fortune. Exim now panics and dies
2262 if the condition is not met. A case was found where this could be provoked
2263 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2264 components; fortuitously, this particular loophole had already been fixed
2265 by change 4.50/55 or 4.44/33 above.
2267 If there are any other similar loopholes, the new check in host_aton()
2268 itself should stop them being exploited. The report I received stated that
2269 data on the command line could provoke the exploit when Exim was running as
2270 exim, but did not say which command line option was involved. All I could
2271 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2272 running as the user.
2275 35. There was a buffer overflow vulnerability in the SPA authentication code
2276 (which came originally from the Samba project). I have added a test to the
2277 spa_base64_to_bits() function which I hope fixes it.
2280 36. The daemon start-up calls getloadavg() while still root for those OS that
2281 need the first call to be done as root, but it missed one case: when
2282 deliver_queue_load_max is set with deliver_drop_privilege. This is
2283 necessary for the benefit of the queue runner, because there is no re-exec
2284 when deliver_drop_privilege is set.
2286 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2287 was in use, was not putting the data itself into the right store pool;
2288 consequently, it could be overwritten for a subsequent message in the same
2289 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2292 38. Sometimes the final signoff response after QUIT could fail to get
2293 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2294 < 0 before doing a fflush(). This bug looks as though it goes back to the
2295 introduction of TLS in release 3.20, but "sometimes" must have been rare
2296 because the tests only now provoked it.
2302 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2303 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2304 Now it returns an integer. A number of calls were still expecting a T/F
2305 return. Fortuitously, in all cases, the tests worked in OK situations,
2306 which is the norm. However, things would have gone wrong on any write
2307 failures on the smtp file descriptor. This function is used when sending
2308 messages over SMTP and also when doing verify callouts.
2310 2. When Exim is called to do synchronous delivery of a locally submitted
2311 message (the -odf or -odi options), it no longer closes stderr before doing
2314 3. Implemented the mua_wrapper option.
2316 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2318 5. Implemented the functions header_remove(), header_testname(),
2319 header_add_at_position(), and receive_remove_recipient(), and exported them
2322 6. If an ACL "warn" statement specified the addition of headers, Exim already
2323 inserted X-ACL-Warn: at the start if there was no header name. However, it
2324 was not making this test for the second and subsequent header lines if
2325 there were newlines in the string. This meant that an invalid header could
2326 be inserted if Exim was badly configured.
2328 7. Allow an ACL "warn" statement to add header lines at the start or after all
2329 the Received: headers, as well as at the end.
2331 8. Added the rcpt_4xx retry error code.
2333 9. Added postmaster_mailfrom=xxx to callout verification option.
2335 10. Added mailfrom=xxxx to the callout verification option, for verify=
2338 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2339 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2341 12. Inserted some casts to stop certain compilers warning when using pointer
2342 differences as field lengths or precisions in printf-type calls (mostly
2343 affecting debugging statements).
2345 13. Added optional readline() support for -be (dynamically loaded).
2347 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2348 same clock tick as a message's arrival, so that its received time was the
2349 same as the "first fail" time on the retry record, and that message
2350 remained on the queue past the ultimate address timeout, every queue runner
2351 would try a delivery (because it was past the ultimate address timeout) but
2352 after another failure, the ultimate address timeout, which should have then
2353 bounced the address, did not kick in. This was a "< instead of <=" error;
2354 in most cases the first failure would have been in the next clock tick
2355 after the received time, and all would be well.
2357 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2358 being recognized when the domain list was tested by the match_domain
2359 condition in an expansion string.
2361 16. Added the ${str2b64: operator.
2363 17. Exim was always calling setrlimit() to set a large limit for the number of
2364 processes, without checking whether the existing limit was already
2365 adequate. (It did check for the limit on file descriptors.) Furthermore,
2366 errors from getrlimit() and setrlimit() were being ignored. Now they are
2367 logged to the main and panic logs, but Exim does carry on, to try to do its
2368 job under whatever limits there are.
2370 18. Imported PCRE 5.0.
2372 19. Trivial typo in log message " temporarily refused connection" (the leading
2375 20. If the log selector return_path_on_delivery was set and an address was
2376 redirected to /dev/null, the delivery process crashed because it assumed
2377 that a return path would always be set for a "successful" delivery. In this
2378 case, the whole delivery is bypassed as an optimization, and therefore no
2381 21. Internal re-arrangement: the function for sending a challenge and reading
2382 a response while authentication was assuming a zero-terminated challenge
2383 string. It's now changed to take a pointer and a length, to allow for
2384 binary data in such strings.
2386 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2388 23. Exim was not respecting finduser_retries when seeking the login of the
2389 uid under which it was called; it was always trying 10 times. (The default
2390 setting of finduser_retries is zero.) Also, it was sleeping after the final
2391 failure, which is pointless.
2393 24. Implemented tls_on_connect_ports.
2395 25. Implemented acl_smtp_predata.
2397 26. If the domain in control=submission is set empty, Exim assumes that the
2398 authenticated id is a complete email address when it generates From: or
2399 Sender: header lines.
2401 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2402 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2403 chown and chgrp in /bin and hostname in /usr/bin.
2405 28. Exim was keeping the "process log" file open after each use, just as it
2406 does for the main log. This opens the possibility of it remaining open for
2407 long periods when the USR1 signal hits a daemon. Occasional processlog
2408 errors were reported, that could have been caused by this. Anyway, it seems
2409 much more sensible not to leave this file open at all, so that is what now
2412 29. The long-running daemon process does not normally write to the log once it
2413 has entered its main loop, and it closes the log before doing so. This is
2414 so that log files can straightforwardly be renamed and moved. However,
2415 there are a couple of unusual error situations where the daemon does write
2416 log entries, and I had neglected to close the log afterwards.
2418 30. The text of an SMTP error response that was received during a remote
2419 delivery was being truncated at 512 bytes. This is too short for some of
2420 the long messages that one sometimes sees. I've increased the limit to
2423 31. It is now possible to make retry rules that apply only when a message has a
2424 specific sender, in particular, an empty sender.
2426 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2427 it possible to be selective about when SMTP synchronization is enforced.
2429 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2431 32. Implemented hosts_connection_nolog.
2433 33. Added an ACL for QUIT.
2435 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2438 35. Added mailbox_size and mailbox_filecount to appendfile.
2440 36. Added control = no_multiline_responses to ACLs.
2442 37. There was a bug in the logic of the code that waits for the clock to tick
2443 in the case where the clock went backwards by a substantial amount such
2444 that the microsecond fraction of "now" was more than the microsecond
2445 fraction of "then" (but the whole seconds number was less).
2447 38. Added support for the libradius Radius client library this is found on
2448 FreeBSD (previously only the radiusclient library was supported).
2454 1. When certain lookups returned multiple values in the form name=value, the
2455 quoting of the values was not always being done properly. Specifically:
2456 (a) If the value started with a double quote, but contained no whitespace,
2458 (b) If the value contained whitespace other than a space character (i.e.
2459 tabs or newlines or carriage returns) it was not quoted.
2460 This fix has been applied to the mysql and pgsql lookups by writing a
2461 separate quoting function and calling it from the lookup code. The fix
2462 should probably also be applied to nisplus, ibase and oracle lookups, but
2463 since I cannot test any of those, I have not disturbed their existing code.
2465 2. A hit in the callout cache for a specific address caused a log line with no
2466 reason for rejecting RCPT. Now it says "Previous (cached) callout
2467 verification failure".
2469 3. There was an off-by-one bug in the queryprogram router. An over-long
2470 return line was truncated at 256 instead of 255 characters, thereby
2471 overflowing its buffer with the terminating zero. As well as fixing this, I
2472 have increased the buffer size to 1024 (and made a note to document this).
2474 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2475 when Exim is waiting for an SMTP response from a remote server, Exim
2476 restarts its select() call on the socket, thereby resetting its timeout.
2477 This is not a problem when such interrupts are rare. Somebody set up a cron
2478 job to run exiwhat every 2 minutes, which is less than the normal select()
2479 timeout (5 or 10 minutes). This meant that the select() timeout never
2480 kicked in because it was always reset. I have fixed this by comparing the
2481 time when an interrupt arrives with the time at the start of the first call
2482 to select(). If more time than the timeout has elapsed, the interrupt is
2483 treated as a timeout.
2485 5. Some internal re-factoring in preparation for the addition of Sieve
2486 extensions (by MH). In particular, the "personal" test is moved to a
2487 separate function, and given an option for scanning Cc: and Bcc: (which is
2488 not set for Exim filters).
2490 6. When Exim created an email address using the login of the caller as the
2491 local part (e.g. when creating a From: or Sender: header line), it was not
2492 quoting the local part when it contained special characters such as @.
2494 7. Installed new OpenBSD configuration files.
2496 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2497 try to make them clearer.
2499 9. Callout options, other than the timeout value, were being ignored when
2500 verifying sender addresses in header lines. For example, when using
2502 verify = header_sender/callout=no_cache
2504 the cache was (incorrectly) being used.
2506 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2507 only the Cygwin environment.
2509 11. When return_path_on_delivery was set as a log selector, if different remote
2510 addresses in the same message used different return paths and parallel
2511 remote delivery occurred, the wrong values would sometimes be logged.
2512 (Whenever a remote delivery process finished, the return path value from
2513 the most recently started remote delivery process was logged.)
2515 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2516 lines when AUTH and/or TLS are in use. This is the "received protocol"
2517 field. Exim used to use "asmtp" for authenticated SMTP, without any
2518 indication (in the protocol name) for TLS use. Now it follows the RFC and
2519 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2520 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2521 names appear in log lines as well as in Received: header lines.
2523 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2524 extensions, and comparison tests, and to fix some bugs.
2526 14. Changes to the "personal" filter test:
2528 (1) The test was buggy in that it was just doing the equivalent of
2529 "contains" tests on header lines. For example, if a user's address was
2530 anne@some.where, the "personal" test would incorrectly be true for
2532 To: susanne@some.where
2534 This test is now done by extracting each address from the header in turn,
2535 and checking the entire address. Other tests that are part of "personal"
2536 are now done using regular expressions (for example, to check local parts
2537 of addresses in From: header lines).
2539 (2) The list of non-personal local parts in From: addresses has been
2540 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2541 taken from the Sieve specification recommendations.
2543 (3) If the message contains any header line starting with "List-" it is
2544 treated as non-personal.
2546 (4) The test for "circular" in the Subject: header line has been removed
2547 because it now seems ill-conceived.
2549 15. Minor typos in src/EDITME comments corrected.
2551 16. Installed latest exipick from John Jetmore.
2553 17. If headers_add on a router specified a text string that was too long for
2554 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2555 of string_sprintf() is now avoided.
2557 18. $message_body_size was not set (it was always zero) when running the DATA
2558 ACL and the local_scan() function.
2560 19. For the "mail" command in an Exim filter, no default was being set for
2561 the once_repeat time, causing a random time value to be used if "once" was
2562 specified. (If the value happened to be <= 0, no repeat happened.) The
2563 default is now 0s, meaning "never repeat". The "vacation" command was OK
2564 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2565 (I found it when inspecting the code).
2567 20. There is now an overall timeout for performing a callout verification. It
2568 defaults to 4 times the callout timeout, which applies to individual SMTP
2569 commands during the callout. The overall timeout applies when there is more
2570 than one host that can be tried. The timeout is checked before trying the
2571 next host. This prevents very long delays if there are a large number of
2572 hosts and all are timing out (e.g. when the network connections are timing
2573 out). The value of the overall timeout can be changed by specifying an
2574 additional sub-option for "callout", called "maxwait". For example:
2576 verify = sender/callout=5s,maxwait=20s
2578 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2579 to the end before writing, but this should make it even safer).
2581 22. Exim was forgetting that it had advertised PIPELINING for the second and
2582 subsequent messages on an SMTP connection. It was also not resetting its
2583 memory on STARTTLS and an internal HELO.
2585 23. When Exim logs an SMTP synchronization error within a session, it now
2586 records whether PIPELINING has been advertised or not.
2588 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2589 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2590 rather than long int.
2592 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2594 26. Added the never_mail option to autoreply.
2600 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2601 crash if the getsockname() call failed; this can happen if a connection is
2602 closed very soon after it is established. The problem was simply in the
2603 order in which certain operations were done, causing Exim to try to write
2604 to the SMTP stream before it had set up the file descriptor. The bug has
2605 been fixed by making things happen in the correct order.
2611 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2612 before the connection was closed, thus losing the rejection response.
2614 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2615 some early Solaris releases, but causes trouble in current releases where
2616 socklen_t is defined.
2618 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2621 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2624 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2625 that releasing the top of it at the end releases what was used for sub-
2626 expansions (unless the block got too big). However, discard this block if
2627 the first thing is a variable or header, so that we can use its block when
2628 it is dynamic (useful for very large $message_headers, for example).
2630 6. Lookups now cache *every* query, not just the most recent. A new, separate
2631 store pool is used for this. It can be recovered when all lookup caches are
2632 flushed. Lookups now release memory at the end of their result strings.
2633 This has involved some general refactoring of the lookup sources.
2635 7. Some code has been added to the store_xxx() functions to reduce the amount
2636 of flapping under certain conditions.
2638 8. log_incoming_interface used to affect only the <= reception log lines. Now
2639 it causes the local interface and port to be added to several more SMTP log
2640 lines, for example "SMTP connection from", and rejection lines.
2642 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2644 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2646 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2647 could be overwritten at the end of the current message (or the start of a
2648 new message if it was set in a HELO ACL). The value is now preserved for
2649 the duration of the SMTP connection.
2651 12. If a transport had a headers_rewrite setting, and a matching header line
2652 contained an unqualified address, that address was qualified, even if it
2653 did not match any rewriting rules. The underlying bug was that the values
2654 of the flags that permit the existence of unqualified sender and recipient
2655 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2656 non-local messages, and by -bnq for local messages) were not being
2657 preserved with the message after it was received.
2659 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2660 "next input=" as part of the text comprising the host identity instead of
2661 the correct text. The code was using the same buffer for two different
2662 strings. However, depending on which order the printing function evaluated
2663 its arguments, the bug did not always show up. Under Linux, for example, my
2664 test suite worked just fine.
2666 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2667 doesn't work with some older versions of Perl. It has been changed to "my",
2668 which in any case is probably the better facility to use.
2670 15. A really picky compiler found some instances of statements for creating
2671 error messages that either had too many or two few arguments for the format
2674 16. The size of the buffer for calls to the DNS resolver has been increased
2675 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2676 for addresses that have a lot of PTR records. This alleviates a problem; it
2677 does not fully solve it.
2679 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2680 buffer now truncates the list and logs the incident, which is the same
2681 action as happens when Exim is looking up a host name and its aliases.
2682 Previously in this situation something unpredictable would happen;
2683 sometimes it was "internal error: store_reset failed".
2685 18. If a server dropped the connection unexpectedly when an Exim client was
2686 using GnuTLS and trying to read a response, the client delivery process
2687 crashed while trying to generate an error log message.
2689 19. If a "warn" verb in an ACL added multiple headers to a message in a single
2690 string, for example:
2692 warn message = H1: something\nH2: something
2694 the text was added as a single header line from Exim's point of view
2695 though it ended up OK in the delivered message. However, searching for the
2696 second and subsequent header lines using $h_h2: did not work. This has been
2697 fixed. Similarly, if a system filter added multiple headers in this way,
2698 the routers could not see them.
2700 20. Expanded the error message when iplsearch is called with an invalid key to
2701 suggest using net-iplsearch in a host list.
2703 21. When running tests using -bh, any delays imposed by "delay" modifiers in
2704 ACLs are no longer actually imposed (and a message to that effect is
2707 22. If a "gecos" field in a passwd entry contained escaped characters, in
2708 particular, if it contained a \" sequence, Exim got it wrong when building
2709 a From: or a Sender: header from that name. A second bug also caused
2710 incorrect handling when an unquoted " was present following a character
2711 that needed quoting.
2713 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2714 was not being matched caselessly.
2716 24. Arranged for all hyphens in the exim.8 source to be escaped with
2719 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2720 properly. Recipient callout cache records were still being keyed to include
2721 the sender, even when use_sender was set false. This led to far more
2722 callouts that were necessary. The sender is no longer included in the key
2723 when use_sender is false.
2725 26. Added "control = submission" modifier to ACLs.
2727 27. Added the ${base62d: operator to decode base 62 numbers.
2729 28. dnsdb lookups can now access SRV records.
2731 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2732 the configuration file.
2734 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2735 (-v) mode. This makes the output for a verbose queue run more intelligible.
2737 31. Added a use_postmaster feature to recipient callouts.
2739 32. Added the $body_zerocount variable, containing the number of binary zero
2740 bytes in the message body.
2742 33. The time of last modification of the "new" subdirectory is now used as the
2743 "mailbox time last read" when there is a quota error for a maildir
2746 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2748 35. Added +ignore_unknown as a special item in host lists.
2750 36. Code for decoding IPv6 addresses in host lists is now included, even if
2751 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2752 address was recognized as an IP address, but was then not correctly decoded
2753 into binary, causing unexpected and incorrect effects when compared with
2760 1. Very minor rewording of debugging text in manualroute to say "list of
2761 hosts" instead of "hostlist".
2763 2. If verify=header_syntax was set, and a header line with an unqualified
2764 address (no domain) and a large number of spaces between the end of the
2765 name and the colon was received, the reception process suffered a buffer
2766 overflow, and (when I tested it) crashed. This was caused by some obsolete
2767 code that should have been removed. The fix is to remove it!
2769 3. When running in the test harness, delay a bit after writing a bounce
2770 message to get a bit more predictability in the log output.
2772 4. Added a call to search_tidyup() just before forking a reception process. In
2773 theory, someone could use a lookup in the expansion of smtp_accept_max_
2774 per_host which, without the tidyup, could leave open a database connection.
2776 5. Added the variables $recipient_data and $sender_data which get set from a
2777 lookup success in an ACL "recipients" or "senders" condition, or a router
2778 "senders" option, similar to $domain_data and $local_part_data.
2780 6. Moved the writing of debug_print from before to after the "senders" test
2783 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2784 problems for message scanning, either using a data ACL, or using
2785 local_scan() because the Received: header was not generated till after they
2786 were called (in order to set the time as the time of reception completion).
2787 I have revised the way this works. The header is now generated after the
2788 body is received, but before the ACL or local_scan() are called. After they
2789 are run, the timestamp in the header is updated.
2795 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2796 before starting a queue runner without re-exec. This happened only when
2797 deliver_drop_privilege was set or when the Exim user was set to root. The
2798 effect of the bug was that timeouts during subsequent deliveries caused
2799 crashes instead of being properly handled. The handler is now left at its
2800 default (and expected) setting.
2802 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2803 message, again when deliver_drop_privilege is set or Exim is run as root.
2804 The bug described in (1) was not present in this case, but the tidying up
2805 of the other signals was missing. I have made the two cases consistent.
2807 3. The ignore_target_hosts setting on a manualroute router was being ignored
2808 for hosts that were looked up using the /MX notation.
2810 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2813 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2814 operated on the sender address. After changing the $sender_address to <>
2815 for the sender address verify, Exim was re-instated it as the original
2816 (before rewriting) address, but remembering that it had rewritten it, so it
2817 wasn't rewriting it again. This bug also had the effect of breaking the
2818 sender address verification caching when the sender address was rewritten.
2820 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2821 This has been changed so that if the ip literal address matches
2822 ignore_target_hosts, the router declines.
2824 7. Added expansion conditions match_domain, match_address, and match_local_
2825 part (NOT match_host).
2827 8. The placeholder for the Received: header didn't have a length field set.
2829 9. Added code to Exim itself and to exim_lock to test for a specific race
2830 condition that could lead to file corruption when using MBX delivery. The
2831 issue is with the lockfile that is created in /tmp. If this file is removed
2832 after a process has opened it but before that process has acquired a lock,
2833 there is the potential for a second process to recreate the file and also
2834 acquire a lock. This could lead to two Exim processes writing to the file
2835 at the same time. The added code performs the same test as UW imapd; it
2836 checks after acquiring the lock that its file descriptor still refers to
2837 the same named file.
2839 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2840 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2843 11. Added the smtp_active_hostname option. If used, this will typically be made
2844 to depend on the incoming interface address. Because $interface_address is
2845 not set up until the daemon has forked a reception process, error responses
2846 that can happen earlier (such as "too many connections") no longer contain
2849 12. If an expansion in a condition on a "warn" statement fails because a lookup
2850 defers, the "warn" statement is abandoned, and the next ACL statement is
2851 processed. Previously this caused the whole ACL to be aborted.
2853 13. Added the iplsearch lookup type.
2855 14. Added ident_timeout as a log selector.
2857 15. Added tls_certificate_verified as a log selector.
2859 16. Added a global option tls_require_ciphers (compare the smtp transport
2860 option of the same name). This controls incoming TLS connections.
2862 17. I finally figured out how to make tls_require_ciphers do a similar thing
2863 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2864 before starting the TLS session.
2866 18. Tabs are now shown as \t in -bP output.
2868 19. If the log selector return_path_on_delivery was set, Exim crashed when
2869 bouncing a message because it had too many Received: header lines.
2871 20. If two routers both had headers_remove settings, and the first one included
2872 a superfluous trailing colon, the final name in the first list and the
2873 first name in the second list were incorrectly joined into one item (with a
2874 colon in the middle).
2880 1. Added -C and -D options to the exinext utility, mainly to make it easier
2881 to include in the automated testing, but these could be helpful when
2882 multiple configurations are in use.
2884 2. The exinext utility was not formatting the output nicely when there was
2885 an alternate port involved in the retry record key, nor when there was a
2886 message id as well (for retries that were specific to a specific message
2887 and a specific host). It was also confused by IPv6 addresses, because of
2888 the additional colons they contain. I have fixed the IPv4 problem, and
2889 patched it up to do a reasonable job for IPv6.
2891 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2892 delivery, the log line now contains "pipelined" if PIPELINING was used.
2894 4. An SMTP transport process used to panic and die if the bind() call to set
2895 an explicit outgoing interface failed. This has been changed; it is now
2896 treated in the same way as a connect() failure.
2898 5. A reference to $sender_host_name in the part of a conditional expansion
2899 that was being skipped was still causing a DNS lookup. This no longer
2902 6. The def: expansion condition was not recognizing references to header lines
2903 that used bh_ and bheader_.
2905 7. Added the _cache feature to named lists.
2907 8. The code for checking quota_filecount in the appendfile transport was
2908 allowing one more file than it should have been.
2910 9. For compatibility with Sendmail, the command line option
2918 and sets the incoming protocol and host name (for trusted callers). The
2919 host name and its colon can be omitted when only the protocol is to be set.
2920 Note the Exim already has two private options, -pd and -ps, that refer to
2921 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2922 "s", but I don't think that's a major issue.
2924 10. A number of refactoring changes to the code, none of which should affect
2927 (a) The number of logging options was getting close to filling up the
2928 32-bit word that was used as a bit map. I have split them into two classes:
2929 those that are passed in the argument to log_write(), and those that are
2930 only ever tested independently outside of that function. These are now in
2931 separate 32-bit words, so there is plenty of room for expansion again.
2932 There is no change in the user interface or the logging behaviour.
2934 (b) When building, for example, log lines, the code previously used a
2935 macro that called string_cat() twice, in order to add two strings. This is
2936 not really sufficiently general. Furthermore, there was one instance where
2937 it was actually wrong because one of the argument was used twice, and in
2938 one call a function was used. (As it happened, calling the function twice
2939 did not affect the overall behaviour.) The macro has been replaced by a
2940 function that can join an arbitrary number of extra strings onto a growing
2943 (c) The code for expansion conditions now uses a table and a binary chop
2944 instead of a serial search (which was left over from when there were very
2945 few conditions). Also, it now recognizes conditions like "pam" even when
2946 the relevant support is not compiled in: a suitably worded error message is
2947 given if an attempt is made to use such a condition.
2949 11. Added ${time_interval:xxxxx}.
2951 12. A bug was causing one of the ddress fields not to be passed back correctly
2952 from remote delivery subprocesses. The field in question was not being
2953 subsequently used, so this caused to problems in practice.
2955 13. Added new log selectors queue_time and deliver_time.
2957 14. Might have fixed a bug in maildirsizefile handling that threw up
2958 "unexpected character" debug warnings, and recalculated the data
2959 unnecessarily. In any case, I expanded the warning message to give more
2962 15. Added the message "Restricted characters in address" to the statements in
2963 the default ACL that block characters like @ and % in local parts.
2965 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2966 Three changes have been made:
2968 (a) There was a serious bug; a negative response to MAIL caused the whole
2969 recipient domain to be cached as invalid, thereby blocking all messages
2970 to all local parts at the same domain, from all senders. This bug has
2971 been fixed. The domain is no longer cached after a negative response to
2972 MAIL if the sender used is not empty.
2974 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2977 (c) A new callout option, "use_sender" has been added for people who want
2978 the modified behaviour.
2984 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2987 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2988 indeed breaks things for older releases.
2990 3. Added additional logging to the case where there is a problem reading data
2991 from a filter that is running in a subprocess using a pipe, in order to
2992 try to track down a specific problem.
2994 4. Testing facility fudge: when running in the test harness and attempting
2995 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2996 getting "No route to host". Convert this to a timeout.
2998 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3001 6. Some OS don't have socklen_t but use size_t instead. This affects the
3002 fifth argument of getsockopt() amongst other things. This is now
3003 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3004 can be set for individual OS. I have set it for SunOS5, OSF1, and
3005 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3006 some earlier ones do not.
3008 7. Change 4.30/15 was not doing the test caselessly.
3010 8. The standard form for an IPv6 address literal was being rejected by address
3011 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3012 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3013 this, as well as the form without the "IPv6" on the front (but only when
3014 address literals are enabled, of course).
3016 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3018 10. Exim crashed if a message with an empty sender address specified by -f
3019 encountered a router with an errors_to setting. This could be provoked only
3020 by a command such as
3024 where an empty string was supplied; "<>" did not hit this bug.
3026 11. Installed PCRE release 4.5.
3028 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3029 remained set. It is now erased.
3031 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3032 times from message ids (which are base 36 rather than the normal 62).
3034 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3035 were being counted as actual protocol errors, and logged if the log
3036 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3037 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3038 and DATA following a set of rejected RCPTs do not count as protocol errors.
3039 In other words, Exim assumes they were pipelined, though this may not
3040 actually be the case. Of course, in all cases the client gets an
3041 appropriate error code.
3043 15. If a lookup fails in an ACL condition, a message about the failure may
3044 be available; it is used if testing the ACL cannot continue, because most
3045 such messages specify what the cause of the deferral is. However, some
3046 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3047 that caused an old message to be retained and used if a later statement
3048 caused a defer, replacing the real cause of the deferral.
3050 16. If an IP address had so many PTR records that the DNS lookup buffer
3051 was not large enough to hold them, Exim could crash while trying to process
3052 the truncated data. It now detects and logs this case.
3054 17. Further to 4.21/58, another change has been made: if (and only if) the
3055 first line of a message (the first header line) ends with CRLF, a bare LF
3056 in a subsequent header line has a space inserted after it, so as not to
3057 terminate the header.
3059 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3060 unnecessarily, used atoi() instead of strtol(), and didn't check the
3061 termination when getting file sizes from file names by regex.
3063 19. Completely re-implemented the support for maildirsize files, in the light
3064 of a number of problems with the previous contributed implementation
3065 (4.30/29). In particular:
3067 . If the quota is zero, the maildirsize file is maintained, but no quota is
3070 . If the maildir directory does not exist, it is created before any attempt
3071 to write a maildirsize file.
3073 . The quota value in the file is just a cache; if the quota is changed in
3074 the transport, the new value overrides.
3076 . A regular expression is available for excluding directories from the
3079 20. The autoreply transport checks the characters in options that define the
3080 message's headers; it allows continued headers, but it was checking with
3081 isspace() after an embedded newline instead of explicitly looking for a
3084 21. If all the "regular" hosts to which an address was routed had passed their
3085 expiry times, and had not reached their retry times, the address was
3086 bounced, even if fallback hosts were defined. Now Exim should go on to try
3089 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3090 equivalent code in the SMTP transport. Some hosts send humungous responses
3091 to HELO/EHLO, more than 1024 it seems.
3093 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3094 gives compiler warnings in some environments. I've now done it "properly",
3097 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3098 (because of problems with the built-in one) was declared to return uschar *
3099 instead of char *, causing compiler failure.
3101 25. Fixed a file descriptor leak when processing alias/forward files.
3103 26. Fixed a minor format string issue in dbfn.c.
3105 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3107 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3108 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3111 29. When a sender address is verified, it is cached, to save repeating the test
3112 when there is more than one recipient in a message. However, when the
3113 verification involves a callout, it is possible for different callout
3114 options to be set for different recipients. It is too complicated to keep
3115 track of this in the cache, so now Exim always runs a verification when a
3116 callout is required, relying on the callout cache for the optimization.
3117 The overhead is duplication of the address routing, but this should not be
3120 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3121 to be verified with callout=postmaster, and the main callout worked but the
3122 postmaster check failed, the verification correctly failed. However, if a
3123 subsequent RCPT command asked for sender verification *without* the
3124 postmaster check, incorrect caching caused this verification also to fail,
3127 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3128 it was not caching the DNS options (qualify_single, search_parents) that
3129 were used when the lookup failed. A subsequent lookup with different
3130 options therefore always gave the same answer, though there were cases
3131 where it should not have. (Example: a "domains = !$mx_any" option on a
3132 dnslookup router: the "domains" option is always processed without any
3133 widening, but the router might have qualify_single set.) Now Exim uses the
3134 cached value only when the same options are set.
3136 32. Added John Jetmore's "exipick" utility to the distribution.
3138 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3139 than a timeout (e.g. a certificate is required, and is not provided), an
3140 Exim server now closes the connection immediately. Previously it waited for
3141 the client to close - but if the client is SSL, it seems that they each
3142 wait for each other, leading to a delay before one of them times out.
3144 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3145 maintained 0.8.x compatibility because I don't think many are using it, and
3146 it is clearly obsolete.
3148 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3151 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3152 client certificate was expired. A simple patch fixes this, though I don't
3153 understand the full logic of why the verify callback is called multiple
3156 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3157 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3158 which causes problems with some clients (such as the Certicom SSL Plus
3159 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3160 disables the coutermeasure allowing Eudora to connect."
3162 38. Exim was not checking that a write() to a log file succeeded. This could
3163 lead to Bad Things if a log got too big, in particular if it hit a file
3164 size limit. Exim now panics and dies if it cannot write to a log file, just
3165 as it does if it cannot open a log file.
3167 39. Modified OS/Makefile-Linux so that it now contains
3169 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3171 The two -D definitions ensure that Exim is compiled with large file
3172 support, which makes it possible to handle log files that are bigger than
3175 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3176 instance) a domain was checked against a named list that involved a lookup,
3177 causing $domain_data to be set, then another domain was checked against the
3178 same list, then the first domain was re-checked, the value of $domain_data
3179 after the final check could be wrong. In particular, if the second check
3180 failed, it could be set empty. This bug probably also applied to
3183 41. The strip_trailing_dot option was not being applied to the address given
3184 with the -f command-line option.
3186 42. The code for reading a message's header from the spool was incrementing
3187 $received_count, but never initializing it. This meant that the value was
3188 incorrect (doubled) while delivering a message in the same process in which
3189 it was received. In the most common configuration of Exim, this never
3190 happens - a fresh exec is done - but it can happen when
3191 deliver_drop_privilege is set.
3193 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3194 it now includes up to 150 characters of the unexpected data in the log
3197 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3198 and building data strings. The size of both of these buffers was 10 000
3199 bytes - far larger than anybody would *ever* want, thought I. Needless to
3200 say, somebody hit the limit. I have increased the maximum line length to
3201 20 000 and the maximum data length of concatenated lines to 100 000. I have
3202 also fixed two bugs, because there was no checking on these buffers. Tsk,
3203 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3204 buffer is too small.
3206 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3207 lsearch lookups. Now it does.
3209 46. When parsing a route_list item in a manualroute router, a fixed-length
3210 buffer was used for the list of hosts. I made this 1024 bytes long,
3211 thinking that nobody would ever have a list of hosts that long. Wrong.
3212 Somebody had a whole pile of complicated expansion conditions, and the
3213 string was silently truncated, leading to an expansion error. It turns out
3214 that it is easier to change to an unlimited length (owing to other changes
3215 that have happened since this code was originally written) than to build
3216 structure for giving a limitation error. The length of the item that
3217 expands into the list of hosts is now unlimited.
3219 47. The lsearch lookup could not handle data where the length of text line was
3220 more than 4095 characters. Such lines were truncated, leading to shortened
3221 data being returned. It should now handle lines of any length.
3223 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3224 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3227 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3228 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3229 no longer happens in comment lines. A list of replacements is now placed
3230 at the head of all of the source files, except those whose only change is
3231 to replace PERL_COMMAND in the very first #! line.
3233 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3234 messages on the queue, with a bottom-up merge sort, using code contributed
3235 by Michael Haardt. This should make operations like -bp somewhat faster on
3236 large queues. It won't affect queue runners, except when queue_run_in_order
3239 51. Installed eximstats 1.31 in the distribution.
3241 52. Added support for SRV lookups to the dnslookup router.
3243 53. If an ACL referred to $message_body or $message_body_end, the value was not
3244 reset for any messages that followed in the same SMTP session.
3246 54. The store-handling optimization for building very long strings was not
3247 differentiating between the different store pools. I don't think this
3248 actually made any difference in practice, but I've tidied it.
3250 55. While running the routers to verify a sender address, $sender_address
3251 was still set to the sender address. This is wrong, because when routing to
3252 send a bounce to the sender, it would be empty. Therefore, I have changed
3253 it so that, while verifying a sender address, $sender_address is set to <>.
3254 (There is no change to what happens when verifying a recipient address.)
3256 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3257 target A or AAAA records (if not already returned) without resetting the
3258 qualify_single or search_parents options of the DNS resolver. These are
3259 inappropriate in this case because the targets of MX and SRV records must
3260 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3261 target that, when qualified, matched something in the local domain. These
3262 two options are now turned off when doing these lookups.
3264 57. It seems that at least some releases of Reiserfs (which does not have the
3265 concept of a fixed number of inodes) returns zero and not -1 for the
3266 number of available inodes. This interacted badly with check_spool_inodes,
3267 which assumed that -1 was the "no such thing" setting. What I have done is
3268 to check that the total number of inodes is greater than zero before doing
3269 the test of how many are available.
3271 58. When a "warn" ACL statement has a log_message modifier, the message is
3272 remembered, and not repeated. This is to avoid a lot of repetition when a
3273 message has many recipients that cause the same warning to be written.
3274 Howewer, Exim was preserving the list of already written lines for an
3275 entire SMTP session, which doesn't seem right. The memory is now reset if a
3276 new message is started.
3278 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3279 debugging output unless log_rewrite was also set.
3281 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3282 of (int)(handle) when we know that handle contains (void *)(-1).
3284 61. The Exim daemon panic-logs an error return when it closes the incoming
3285 connection. However "connection reset by peer" seems to be common, and
3286 isn't really an error worthy of noting specially, so that particular error
3289 62. When Exim is trying to find all the local interfaces, it used to panic and
3290 die if the ioctl to get the interface flags failed. However, it seems that
3291 on at least one OS (Solaris 9) it is possible to have an interface that is
3292 included in the list of interfaces, but for which you get a failure error
3293 for this call. This happens when the interface is not "plumbed" into a
3294 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3295 failure of the "get flags" call assumes that the interface is down.
3297 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3298 makes life easier for people who are doing arithmetic on fields extracted
3299 from dates, where you often get leading zeros that should not be
3300 interpreted as octal.
3302 64. Added qualify_domain to the redirect router, to override the global
3305 65. If a pathologically long header line contained very many addresses (the
3306 report of this problem mentioned 10 000) and each of them was rewritten,
3307 Exim could use up a very large amount of memory. (It kept on making new
3308 copies of the header line as it rewrote, and never released the old ones.)
3309 At the expense of a bit more processing, the header rewriting function has
3310 been changed so that it no longer eats memory in this way.
3312 66. The generation of the Received: header has been moved from the time that a
3313 message starts to be received, to the time that it finishes. The timestamp
3314 in the Received: header should now be very close to that of the <= log
3315 line. There are two side-effects of this change:
3317 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3318 logged header lines no longer include the local Received: line, because
3319 it has not yet been created. The same applies to a copy of the message
3320 that is returned to a non-SMTP sender when a message is rejected.
3322 (b) When a filter file is tested using -bf, no additional Received: header
3323 is added to the test message. After some thought, I decided that this
3326 This change does not affect the value of $received_for. It is still set
3327 after address rewriting, but before local_scan() is called.
3329 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3331 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3332 gave an unhelpful panic error message, and a defer error. I have managed to
3333 change this behaviour so that it now rejects any supplied certificate,
3334 which seems right, as the list of acceptable certificates is empty.
3336 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3337 gave an unhelpful defer error. I have not managed to make this reject any
3338 supplied certificates, but the error message it gives is "no certificate
3339 supplied", which is not helpful.
3341 70. exigrep's output now also includes lines that are not associated with any
3342 message, but which match the given pattern. Implemented by a patch from
3343 Martin Sluka, which also tidied up the Perl a bit.
3345 71. Recipient callout verification, like sender verification, was using <> in
3346 the MAIL FROM command. This isn't really the right thing, since the actual
3347 sender may affect whether the remote host accepts the recipient or not. I
3348 have changed it to use the actual sender in the callout; this means that
3349 the cache record is now keyed on a recipient/sender pair, not just the
3350 recipient address. There doesn't seem to be a real danger of callout loops,
3351 since a callout by the remote host to check the sender would use <>.
3352 [SEE ABOVE: changed after hitting problems.]
3354 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3355 temporary errors. However, in the case of such a code being given after
3356 the end of a data transmission (i.e. after ".") Exim was failing to write
3357 a retry record for the message. (Yes, there was some broken host that was
3358 actually sending 8xx at this point.)
3360 73. An unknown lookup type in a host list could cause Exim to panic-die when
3361 the list was checked. (An example that provoked this was putting <; in the
3362 middle of a list instead of at the start.) If this happened during a DATA
3363 ACL check, a -D file could be left lying around. This kind of configuration
3364 error no longer causes Exim to die; instead it causes a defer errror. The
3365 incident is still logged to the main and panic logs.
3367 74. Buglet left over from Exim 3 conversion. The message "too many messages
3368 in one connection" was written to the rejectlog but not the mainlog, except
3369 when address rewriting (yes!) was being logged.
3371 75. Added write_rejectlog option.
3373 76. When a system filter was run not as root (that is, when system_filter_user
3374 was set), the values of the $n variables were not being returned to the
3375 main process; thus, they were not subsequently available in the $sn
3378 77. Added +return_path_on_delivery log selector.
3380 78. A connection timeout was being treated differently from recipients deferred
3381 when testing hosts_max_try with a message that was older than the host's
3382 retry timeout. (The host should not be counted, thus allowing all hosts to
3383 be tried at least once before bouncing.) This may have been the cause of an
3384 occasionally reported bug whereby a message would remain on the queue
3385 longer than the retry timeout, but would be bounced if a delivery was
3386 forced. I say "may" because I never totally pinned down the problem;
3387 setting up timeout/retry tests is difficult. See also the next item.
3389 79. The ultimate address timeout was not being applied to errors that involved
3390 a combination of host plus message (for example, a timeout on a MAIL
3391 command). When an address resolved to a number of possible hosts, and they
3392 were not all tried for each delivery (e.g. because of hosts_max_try), a
3393 message could remain on the queue longer than the retry timeout.
3395 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3398 81. Fixed an obscure SMTP outgoing bug which required at least the following
3399 conditions: (a) there was another message waiting for the same server;
3400 (b) the server returned 5xx to all RCPT commands in the first message so
3401 that the message was not completed; (c) the server dropped the connection
3402 or gave a negative response to the RSET that Exim sends to abort the
3403 transaction. The observed case was a dropped connection after DATA that had
3404 been sent in pipelining mode. That is, the server had advertised PIPELINING
3405 but was not implementing it correctly. The effect of the bug was incorrect
3406 behaviour, such as trying another host, and this could lead to a crash.
3412 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3413 and daemon.c were passed as pointers to ints; they should have been
3414 pointers to socklen_t variables (which are typically unsigned ints).
3416 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3419 3. Fixed a really odd bug that affected only the testing scheme; patching a
3420 certain fixed string in the binary changed the value of another string that
3421 happened to be identical to the end of the original first string.
3423 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3424 name", it returns that address as the IP address. On some operating
3425 systems (e.g. Solaris), it also passes back the IP address string as the
3426 "host name". However, on others (e.g. Linux), it passes back an empty
3427 string. Exim wasn't checking for this, and was changing the host name to an
3428 empty string, assuming it had been canonicized.
3430 5. Although rare, it is permitted to have more than one PTR record for a given
3431 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3432 all the names associated with an address, because they do in Solaris.
3433 However, it seems that they do not in Linux for data that comes from the
3434 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3435 I found this out when I moved to a new Linux workstation and tried to run
3436 the Exim test suite.
3438 To get round this problem I have changed the code so that it now does its
3439 own call to the DNS to look up PTR records when searching for a host name.
3440 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3441 addresses that are only in /etc/hosts are still found.
3443 This behaviour is, however, controlled by an option called host_lookup_
3444 order, which defaults to "bydns:byaddr". If people want to use the other
3445 order, or indeed, just use one or the other means of lookup, they can
3446 specify it in this variable.
3448 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3449 some operating systems, this comes back from gethostbyaddr() as an empty
3450 string, and this is what Exim used to test for. However, it seems that in
3451 other systems, "." is yielded. Exim now tests for this case too.
3453 7. The values of check_spool_space and check_log_space are now held internally
3454 as a number of kilobytes instead of an absolute number of bytes. If a
3455 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3456 kilobyte. This means that much larger values can be stored.
3458 8. Exim monitor: an attempt to get the action menu when not actually pointing
3459 at a message produces an empty menu entitled "No message selected". This
3460 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3461 no entries in it ("Shell widget menu has zero width and/or height"). So I
3462 have added a single, blank menu entry in this case.
3464 9. Added ${quote_local_part.
3466 10. MIME decoding is now applied to the contents of Subject: header lines when
3469 11. Now that a reference to $sender_host_address automatically causes a reverse
3470 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3471 host lookup before query-style lookups in lists that might use this
3472 variable. This has therefore been abolished, and the "net-" prefix is no
3473 longer necessary for query-style lookups.
3475 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3476 have been a typo for LFLAGS, so it has been changed.
3478 13. The install script calls Exim with "-C /dev/null" in order to find the
3479 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3480 to be output. Howeve, since Exim outputs its version number before the
3481 error, it didn't break the script. It just looked ugly. I fixed this by
3482 always allowing "-C /dev/null" if the caller is root.
3484 14. Ignore overlarge ACL variable number when reading spool file - insurance
3485 against a later release with more variables having written the file.
3487 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3488 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3489 this, as well as the form without the "IPv6" on the front.
3491 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3492 OS/Makefile-Darwin file.
3494 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3495 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3497 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3498 variable. However, some OS use ints for inodes. I've added cast to long int
3499 to get rid of the compiler warning.
3501 19. I had forgotten to lock out "/../" in configuration file names when
3502 ALT_CONFIG_PREFIX was set.
3504 20. Routers used for verification do not need to specify transports. However,
3505 if such a router generated a host list, and callout was configured, Exim
3506 crashed, because it could not find a port number from the (non-existent)
3507 transport. It now assumes port 25 in this circumstance.
3509 21. Added the -t option to exigrep.
3511 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3512 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3513 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3514 them is set, LOOKUP_LSEARCH is forced.
3516 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3518 24. Added sender and host information to the "rejected by local_scan()" log
3519 line; previously there was no indication of these.
3521 25. Added .include_if_exists.
3523 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3524 Linux. It turns out that not all file systems support this. Apparently some
3525 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3526 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3527 not supported on the file descriptor, is now ignored when Exim is trying to
3528 sync a directory. This applies only to Linux.
3530 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3532 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3533 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3536 29. Added support for maildirsize files from supplied patch (modified a bit).
3538 30. The use of :fail: followed by an empty string could lead Exim to respond to
3539 sender verification failures with (e.g.):
3541 550 Verification failed for <xxx>
3542 550 Sender verify failed
3544 where the first response line was missing the '-' that indicates it is not
3545 the final line of the response.
3547 31. The loop for finding the name of the user that called Exim had a hardwired
3548 limit of 10; it now uses the value of finduser_retries, which is used for
3549 all other user lookups.
3551 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3554 33. Exim was neglecting to zero errno before one call of strtol() when
3555 expanding a string and expecting an integer value. On some systems this
3556 resulted in spurious "integer overflow" errors. Also, it was casting the
3557 result into an int without checking.
3559 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3560 did not work. The code looks as if it has *never* worked, though it appears
3561 to have been documented since at least releast 1.62. I have made it work.
3563 35. The "timeout_DNS" error in retry rules, also documented since at least
3564 1.62, also never worked. As it isn't clear exactly what this means, and
3565 clearly it isn't a major issue, I have abolished the feature by treating it
3566 as "timeout", and writing a warning to the main and panic logs.
3568 36. The display of retry rules for -brt wasn't always showing the error code
3571 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3572 timeout_connect_A, timeout_connect_MX.
3574 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3575 to the empty sender.
3577 39. The daemon was not analysing the content of -oX till after it had closed
3578 stderr and disconnected from the controlling terminal. This meant that any
3579 syntax errors were only noted on the panic log, and the return code from
3580 the command was 0. By re-arranging the code a little, I've made the
3581 decoding happen first, so such errors now appear on stderr, and the return
3582 code is 1. However, the actual setting up of the sockets still happens in
3583 the disconnected process, so errors there are still only recorded on the
3586 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3587 connections (as happens on some IP stacks) was logged at start up time as
3588 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3589 it from "IPv6 and IPv4", which means that two separate sockets are being
3592 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3593 says whether AF_INET or AF_INET6 was passed as an argument.
3595 42. Exiwhat output was messed up when time zones were included in log
3598 43. Exiwhat now gives more information about the daemon's listening ports,
3599 and whether -tls-on-connect was used.
3601 44. The "port" option of the smtp transport is now expanded.
3603 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3604 silently ignored. Now an error message is written to the main and panic
3607 46. There's a new ACL modifier called "logwrite" which writes to a log file
3608 as soon as it is encountered.
3610 47. Added $local_user_uid and $local_user_gid at routing time.
3612 48. Exim crashed when trying to verify a sender address that was being
3615 49. Exim was recognizing only a space character after ".include". It now also
3616 recognizes a tab character.
3618 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3619 extracting the relevant information from the specification. The man page no
3620 longer contains scrambled data for the -d option, and I've added a section
3621 at the front about calling Exim under different names.
3623 51. Added "extra_headers" argument to the "mail" command in filter files.
3625 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3628 53. Installed eximstats 1.29.
3630 54. Added transport_filter_timeout as a generic transport option.
3632 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3633 Cc: header lines. This was required by RFC 822, but it not required by RFC
3636 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3637 incoming messages that did not have them. Now it does so only if the
3638 message originates locally, that is, if there is no associated remote host
3639 address. When Resent- header lines are present, this applies to the Resent-
3640 lines rather than the non-Resent- lines.
3642 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3643 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3645 58. Messages for configuration errors now include the name of the main
3646 configuration file - useful now that there may be more than one file in a
3647 list (.included file names were always shown).
3649 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3650 for those rare installations that do not start the daemon as root or run it
3651 setuid root. I've cut out the call to initgroups() if the daemon is not
3654 60. The Exim user and group can now be bound into the binary as text strings
3655 that are looked up at the start of Exim's processing.
3657 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3659 62. Added $mailstore_basename variable.
3661 63. Installed patch to sieve.c from Michael Haardt.
3663 64. When Exim failed to open the panic log after failing to open the main log,
3664 the original message it was trying to log was written to stderr and debug
3665 output, but if they were not available (the usual case in production), it
3666 was lost. Now it is written to syslog before the two lines that record the
3667 failures to open the logs.
3669 65. Users' Exim filters run in subprocesses under the user's uid. It is
3670 possible for a "deliver" command or an alias in a "personal" command to
3671 provoke an address rewrite. If logging of address rewriting is configured,
3672 this fails because the process is not running as root or exim. There may be
3673 a better way of dealing with this, but for the moment (because 4.30 needs
3674 to be released), I have disabled address rewrite logging when running a
3675 filter in a non-root, non-exim process.
3681 1. The buildconfig auxiliary program wasn't quoting the value set for
3682 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3683 not defined. This bug was masked in 4.22 by the effect that was fixed in
3686 2. Some messages that were rejected after a message id was allocated were
3687 shown as "incomplete" by exigrep. It no longer does this for messages that
3688 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3690 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3691 have allow_domain_literals set, the ID did not get logged in the <= line.
3692 Domain literals are now always recognized in Message-ID: header lines.
3694 4. The first argument for a ${extract expansion item is the key name or field
3695 number. Leading and trailing spaces in this item were not being ignored,
3696 causing some misleading effects.
3698 5. When deliver_drop_privilege was set, single queue runner processes started
3699 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3700 same command in the process it spins off) were not dropping privilege.
3702 6. When the daemon running as "exim" started a queue runner, it always
3703 re-executed Exim in the spun-off process. This is a waste of effort when
3704 deliver_drop_privilege is set. The new process now just calls the
3705 queue-runner function directly.
3711 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3714 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3715 ignored. Though the use of -oP was forcing the writing of a pid file, it
3716 was always written to the default place.
3718 3. If the message "no IP address found for host xxxx" is generated during
3719 incoming verification, it is now followed by identification of the incoming
3720 connection (so you can more easily find what provoked it).
3722 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3724 5. Added some features to "harden" Exim a bit more against certain attacks:
3726 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3727 be put in Local/Makefile. This is like the never_users runtime option,
3728 but it cannot be overridden. The default setting is "root".
3730 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3731 prefix string with which any file named in a -C command line option
3734 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3735 is retained for -C and -D only if the caller of Exim is root. Without
3736 it, the exim user may also use -C and -D and retain privilege.
3738 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3739 command line option is disabled.
3741 6. Macro names set by the -D option must start with an upper case letter, just
3742 like macro names defined in the configuration file.
3744 7. Added "dereference=" facility to LDAP.
3746 8. Two instances of the typo "uknown" in the source files are fixed.
3748 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3749 the Configure-Makefile script screwed up while processing it.
3751 10. Incorporated PCRE 4.4.
3753 11. The SMTP synchronization check was not operating right at the start of an
3754 SMTP session. For example, it could not catch a HELO sent before the client
3755 waited for the greeting. There is now a check for outstanding input at the
3756 point when the greeting is written. Because of the duplex, asynchronous
3757 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3758 way, but not yet received, when the check is performed.
3760 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3761 on TCP/IP sockets, because this apparently causes some broken clients to
3764 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3765 unchanged) from the Cygwin maintainer.
3767 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3768 was supported instead of testing for mbx. Effectively a typo.
3770 15. The spa authenticator server code was not checking that the input it
3771 received was valid base64.
3773 16. The debug output line for the "set" modifier in ACLs was not showing the
3774 name of the variable that was being set.
3776 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3778 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3779 lookup on its own. Something else was needed to trigger the lookup. For
3780 example, a match in host_lookup or the need for a host name in a host list.
3781 Now, if $sender_host_name is referenced and the host name has not yet been
3782 looked up, a lookup is performed. If the lookup fails, the variable remains
3783 empty, and $host_lookup_failed is set to "1".
3785 19. Added "eqi" as a case-independent comparison operator.
3787 20. The saslauthd authentication condition could segfault if neither service
3788 nor realm was specified.
3790 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3791 error message that was logged was misleading, and incoming SMTP
3792 connections were dropped. The message is now more accurate, and temporary
3793 errors are given to SMTP connections.
3795 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3796 (except RSET) with a 421 error, until QUIT is received. However, it was
3797 failing to send a response to QUIT.
3799 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3800 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3801 ACL is now run after the helo_try_verify_hosts code.
3803 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3804 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3805 case-independent, so other case variants are also recognized). Apparently
3806 some systems use these upper case variants.
3808 25. If more than two messages were waiting for the same host, and a transport
3809 filter was specified for the transport, Exim sent two messages over the
3810 same TCP/IP connection, and then failed with "socket operation on non-
3811 socket" when it tried to send the third.
3813 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3815 27. The extern definition of crypt16() in expand.c was not being excluded when
3816 the OS had its own crypt16() function.
3818 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3819 as a preferred synonym for return_size_limit, both as an option and as an
3822 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3824 30. Changed the default configuration ACL to relax the local part checking rule
3825 for addresses that are not in any local domains. For these addresses,
3826 slashes and pipe symbols are allowed within local parts, but the sequence
3827 /../ is explicitly forbidden.
3829 31. SPA server authentication was not clearing the challenge buffer before
3832 32. log_message in a "warn" ACL statement was writing to the reject log as
3833 well as to the main log, which contradicts the documentation and doesn't
3834 seem right (because no rejection is happening). So I have stopped it.
3836 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3837 However, I am unable to do any testing of this.
3839 34. Fixed an infelicity in the appendfile transport. When checking directories
3840 for a mailbox, to see if any needed to be created, it was accidentally
3841 using path names with one or more superfluous leading slashes; tracing
3842 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3844 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3845 contents are added to the log line that is written for every discarded
3846 recipient. (Previously a log_message setting was ignored.)
3848 36. The ${quote: operator now quotes the string if it is empty.
3850 37. The install script runs exim in order to find its version number. If for
3851 some reason other than non-existence or emptiness, which it checks, it
3852 could not run './exim', it was installing it with an empty version number,
3853 i.e. as "exim-". This error state is now caught, and the installation is
3856 38. An argument was missing from the function that creates an error message
3857 when Exim fails to connect to the socket for saslauthd authentication.
3858 This could cause Exim to crash, or give a corrupted message.
3860 39. Added isip, isip4, and isip6 to ${if conditions.
3862 40. The ACL variables $acl_xx are now saved with the message, and can be
3863 accessed later in routers, transports, and filters.
3865 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3866 strings in the file are not string-expanded.
3868 42. If a MAIL command specified a SIZE value that was too large to fit into an
3869 int variable, the check against message_size_limit failed. Such values are
3870 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3871 day this will have to be increased, but I don't think I want to be around
3872 when emails are that large.
3879 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3880 iconv() is not standard in FreeBSD.
3882 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3883 IPv6 enabled. The observed symptom was a segmentation fault on return from
3884 the function os_common_find_running_interfaces() in src/os.c.
3886 3. In the check_special_case() function in daemon.c I had used "errno" as an
3887 argument name, which causes warnings on some systems. This was basically a
3888 typo, since it was named "eno" in the comments!
3890 4. The code that waits for the clock to tick (at a resolution of some fraction
3891 of a second) so as to ensure message-id uniqueness was always waiting for
3892 at least one whole tick, when it could have waited for less. [This is
3893 almost certainly not relevant at current processor speeds, where it is
3894 unlikely to ever wait at all. But we try to future-proof.]
3896 5. The function that sleeps for a time interval that includes fractions of a
3897 second contained a race. It did not block SIGALRM between setting the
3898 timer, and suspending (a couple of lines later). If the interval was short
3899 and the sigsuspend() was delayed until after it had expired, the suspension
3900 never ended. On busy systems this could lead to processes getting stuck for
3903 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3904 process, before it forks any delivery processes. The open lookup caching
3905 mechanism meant that the open file or database connection was passed into
3906 the delivery process. The problem was that delivery processes always tidy
3907 up cached lookup data. This could cause a problem for the next delivery
3908 process started by the queue runner, because the external queue runner
3909 process does not know about the closure. So the next delivery process
3910 still has data in the lookup cache. In the case of a file lookup, there was
3911 no problem because closing a file descriptor in a subprocess doesn't affect
3912 the parent. However, if the lookup was caching a connection to a database,
3913 the connection was closed, and the second delivery process was likely to
3914 see errors such as "PGSQL: query failed: server closed the connection
3915 unexpectedly". The problem has been fixed by closing all cached lookups
3916 in a queue runner before running a delivery process.
3918 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3919 seem to have the "const" qualifier which it has on other OS. I've
3922 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3923 *of the same type* (client or server) with the same public name that an
3924 error should be diagnosed.
3926 9. When Exim looked up a host name for an IP address, but failed to find the
3927 original IP address when looking up the host name (a safety check), it
3928 output the message "<ip address> does not match any IP for NULL", which was
3929 confusing, to say the least. The bug was that the host name should have
3930 appeared instead of "NULL".
3932 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3933 user that is built into the binary, and the -C or -D options is used, root
3934 privilege is dropped before the configuration file is read. In addition,
3935 logging is switched to stderr instead of the normal log files. If the
3936 configuration then re-defines the Exim user, the unprivileged environment
3937 is probably not what is expected, so Exim logs a panic warning message (but
3940 However, if deliver_drop_privilege is set, the unprivileged state may well
3941 be exactly what is intended, so the warning has been cut out in that case,
3942 and Exim is allowed to try to write to its normal log files.
3948 1. smtp_return_error_details was not giving details for temporary sender
3949 or receiver verification errors.
3951 2. Diagnose a configuration error if two authenticators have the same public
3954 3. Exim used not to create the message log file for a message until the first
3955 delivery attempt. This could be confusing when incoming messages were held
3956 for policy or load reasons. The message log file is now created at the time
3957 the message is received, and an initial "Received" line is written to it.
3959 4. The automatically generated man page for command line options had a minor
3960 bug that caused no ill effects; however, a more serious problem was that
3961 the procedure for building the man page automatically didn't always
3962 operate. Consequently, release 4.20 contains an out-of-date version. This
3963 shouldn't happen again.
3965 5. When building Exim with embedded Perl support, the script that builds the
3966 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3967 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3969 6. The freeze_tell option was not being used for messages that were frozen on
3970 arrival, either by an ACL or by local_scan().
3972 7. Added the smtp_incomplete_transaction log selector.
3974 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3975 was accepting AUTH without a new EHLO.
3977 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3978 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3981 10. Logging of TCP/IP connections (when configured) now happens in the main
3982 daemon process instead of the child process, so that the TCP/IP connection
3983 count is more accurate (but it can never be perfect).
3985 11. The use of "drop" in a nested ACL was not being handled correctly in the
3986 outer ACL. Now, if condition failure induced by the nested "drop" causes
3987 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3988 or "require"), the connection is dropped.
3990 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3991 that yield "discard" can now be used with an "accept" or a "discard" verb,
3992 but an error is generated for any others (because I can't see a useful way
3993 to define what should happen).
3995 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3996 are now processed in the same way as lines in the Exim configuration file.
3997 In particular, continuation lines are supported.
3999 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4001 15. Added -ti meaning -t -i.
4003 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4004 domains, and warn by logging if others are found.
4006 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4007 returned from GIFCONF when Exim is trying to find the list of interfaces on
4008 a host. The code in os.c has been modified to copy each ifreq to an aligned
4009 structure in all cases.
4011 Also, in some cases, the returned ifreq's were being copied to a 'struct
4012 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4013 means the last couple of bytes of an IPv6 address could be chopped if the
4014 ifreq contained only a normal sockaddr (14 bytes storage).
4016 18. Named domain lists were not supported in the hosts_treat_as_local option.
4017 An entry such as +xxxx was not recognized, and was treated as a literal
4020 19. Ensure that header lines added by a DATA ACL are included in the reject log
4021 if the ACL subsequently rejects the message.
4023 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4024 MD5 (which is deprecated).
4026 21. When testing a filter file using -bf, Exim was writing a message when it
4027 took the sender from a "From " line in the message, but it was not doing so
4028 when it took $return_path from a Return-Path: header line. It now does.
4030 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4031 with a valid header line field name (a series of printing characters
4032 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4034 23. Changed "disc" in the source to "disk" to conform to the documentation and
4035 the book and for uniformity.
4037 24. Ignore Sendmail's -Ooption=value command line item.
4039 25. When execve() failed while trying to run a command in a pipe transport,
4040 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4041 could be confused with a return value of 69 from the command itself. This
4042 has been changed to 127, the value the shell returns if it is asked to run
4043 a non-existent command. The wording for the related log line suggests a
4044 non-existent command as the problem.
4046 26. If received_header_text expands to an empty string, do not add a Received:
4047 header line to the message. (Well, it adds a token one on the spool, but
4048 marks it "old" so that it doesn't get used or transmitted.)
4050 27. Installed eximstats 1.28 (addition of -nt option).
4052 28. There was no check for failure on the call to getsockname() in the daemon
4053 code. This can fail if there is a shortage of resources on the system, with
4054 ENOMEM, for example. A temporary error is now given on failure.
4056 29. Contrary to the C standard, it seems that in some environments, the
4057 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4058 program. Exim now does this explicitly; it affects the formatting of
4059 timestamps using strftime().
4061 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4062 complaints. I've now initialized all the variables, to avoid this.
4064 32. Header lines added by a system filter were not being "seen" during
4065 transport-time rewrites.
4067 33. The info_callback() function passed to OpenSSL is set up with type void
4068 (*)(SSL *, int, int), as described somewhere. However, when calling the
4069 function (actually a macro) that sets it up, the type void(*)() is
4070 expected. I've put in a cast to prevent warnings from picky compilers.
4072 34. If a DNS black list lookup found a CNAME record, but there were no A
4073 records associated with the domain it pointed at, Exim crashed.
4075 35. If a DNS black list lookup returned more than one A record, Exim ignored
4076 all but the first. It now scans all returned addresses if a particular IP
4077 value is being sought. In this situation, the contents of the
4078 $dnslist_value variable are a list of all the addresses, separated by a
4081 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4082 to accept a host name and all its aliases if the forward lookup for any of
4083 them yielded the IP address of the incoming connection. Now it accepts only
4084 those names whose forward lookup yields the correct IP address. Any other
4085 names are discarded. This closes a loophole whereby a rogue DNS
4086 administrator could create reverse DNS records to break through a
4087 wildcarded host restriction in an ACL.
4089 37. If a user filter or a system filter that ran in a subprocess used any of
4090 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4091 the wrong values were passed to the pipe command ($thisaddress had the
4092 value of $0, $0 had the value of $1, etc). This bug was introduced by
4093 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4095 38. Improved the line breaking for long SMTP error messages from ACLs.
4096 Previously, if there was no break point between 40 and 75 characters, Exim
4097 left the rest of the message alone. Two changes have been made: (a) I've
4098 reduced the minimum length to 35 characters; (b) if it can't find a break
4099 point between 35 and 75 characters, it looks ahead and uses the first one
4100 that it finds. This may give the occasional overlong line, but at least the
4101 remaining text gets split now.
4103 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4104 file descriptors might be low, and that setting 1000 would always raise it.
4105 It turns out that in some environments, the limit is already over 1000 and
4106 that lowering it causes trouble. So now Exim takes care not to decrease it.
4108 40. When delivering a message, the value of $return_path is set to $sender_
4109 address at the start of routing (routers may change the value). By an
4110 oversight, this default was not being set up when an address was tested by
4111 -bt or -bv, which affected the outcome if any router or filter referred to
4114 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4115 log only when "message" or "log_message" are set. However, if one of the
4116 conditions was an address verification, or a call to a nested ACL, the
4117 messages generated by the underlying test were being passed through. This
4118 no longer happens. The underlying message is available in $acl_verify_
4119 message for both "message" and "log_message" expansions, so it can be
4120 passed through if needed.
4122 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4123 new expansion $bh_ to give the encoded byte string without charset
4124 translation. Translation happens only if iconv() is available; HAVE_ICONV
4125 indicates this at build time. HEADERS_CHARSET gives the charset to
4126 translate to; headers_charset can change it in the configuration, and
4127 "headers charset" can change it in an individual filter file.
4129 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4130 that creates RFC 2047 encoded "words" labels them as that charset instead
4131 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4132 expansion operator; (ii) when Exim creates a From: line for a local
4133 message; (iii) when a header line is rewritten to include a "phrase" part.
4135 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4136 buggy, causing it to skip the first lines of messages whose message ID
4137 ended in 'D'. This would not have bitten before Exim release 4.14, because
4138 message IDs were unlikely to end in 'D' before then. The effect was to have
4139 incorrect size information for certain domains.
4141 45. #include "config.h" was missing at the start of the crypt16.c module. This
4142 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4145 46. If there was a timeout during a "random" callout check, Exim treated it as
4146 a failure of the random address, and carried on sending RSET and the real
4147 address. If the delay was just some slowness somewhere, the response to the
4148 original RCPT would be taken as a response to RSET and so on, causing
4149 mayhem of various kinds.
4151 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4152 when I implemented it. It didn't allow for the fact that some option values
4153 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4154 the right test for positive values.
4156 48. Domain names in DNS records are case-independent. Exim always looks them up
4157 in lower case. Some resolvers return domain names in exactly the case they
4158 appear in the zone file, that is, they may contain uppercase letters. Not
4159 all resolvers do this - some return always lower case. Exim was treating a
4160 change of case by a resolver as a change of domain, similar to a widening
4161 of a domain abbreviation. This triggered its re-routing code and so it was
4162 trying to route what was effectively the same domain again. This normally
4163 caused routing to fail (because the router wouldn't handle the domain
4164 twice). Now Exim checks for this case specially, and just changes the
4165 casing of the domain that it ultimately uses when it transmits the message
4168 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4171 50. If a filter generated a file delivery with a non-absolute name (possible if
4172 no home directory exists for the router), the forbid_file option was not
4175 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4176 the existing equality matching.
4178 52. Exim was using ints instead of ino_t variables in some places where it was
4179 dealing with inode numbers.
4181 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4182 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4183 called TMPDIR, and if it finds it is different, it changes its value.
4185 54. The smtp_printf() function is now made available to local_scan() so
4186 additional output lines can be written before returning. There is also an
4187 smtp_fflush() function to enable the detection of a dropped connection.
4188 The variables smtp_input and smtp_batched_input are exported to
4191 55. Changed the default runtime configuration: the message "Unknown user"
4192 has been removed from the ACL, and instead placed on the localuser router,
4193 using the cannot_route_message feature. This means that any verification
4194 failures that generate their own messages won't get overridden. Similarly,
4195 the "Unrouteable address" message that was in the ACL for unverifiable
4196 relay addresses has also been removed.
4198 56. Added hosts_avoid_esmtp to the smtp transport.
4200 57. The exicyclog script was not checking for the esoteric option
4201 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4202 will work only if exicyclog is run under the appropriate euid.
4204 58. Following a discussion on the list, the rules by which Exim recognises line
4205 endings on incoming messages have been changed. The -dropcr and drop_cr
4206 options are now no-ops, retained only for backwards compatibility. The
4207 following line terminators are recognized: LF CRLF CR. However, special
4208 processing applies to CR:
4210 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4211 nor a local message in the state where . is a terminator.
4213 (ii) If a bare CR is encountered in a header line, an extra space is added
4214 after the line terminator so as not to end the header. The reasoning
4215 behind this is that bare CRs in header lines are most likely either
4216 to be mistakes, or people trying to play silly games.
4218 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4219 was being incorrectly given as 18 bytes larger than it should have been.
4220 This is a VOB (very old bug).
4222 60. This may never have affected anything current, but just in case it has:
4223 When the local host is found other than at the start of a list of hosts,
4224 the local host, those with the same MX, and any that follow, are discarded.
4225 When the list in question was part of a longer list of hosts, the following
4226 hosts (not currently being processed) were also being discarded. This no
4227 longer happens. I'm not sure if this situation could ever has previously
4230 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4233 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4234 header. This is something that is recommended in a new Internet Draft, and
4235 is something that is documented as being done by Sendmail. There are two
4236 possible values. For messages generated by the autoreply transport, Exim
4239 Auto-Submitted: auto-replied
4241 whereas for all other generated messages (e.g. bounces) it adds
4243 Auto-Submitted: auto-generated
4245 63. The "personal" condition in filters now includes a test for the
4246 Auto-Submitted: header. If it contains the string "auto-" the message it
4247 not considered personal.
4249 64. Added rcpt_include_affixes as a generic transport option.
4251 65. Added queue_only_override (default true).
4253 66. Added the syslog_duplication option.
4255 67. If what should have been the first header line of a message consisted of
4256 a space followed by a colon, Exim was mis-interpreting it as a header line.
4257 It isn't of course - it is syntactically invalid and should therefore be
4258 treated as the start of the message body. The misbehaviour could have
4259 caused a number of strange effects, including loss of data in subsequent
4260 header lines, and spool format errors.
4262 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4263 client host had authenticated. This control can now be exercised by an ACL
4264 for more flexibility.
4266 69. By default, callouts do not happen when testing with -bh. There is now a
4267 variant, -bhc, which does actually run the callout code, including
4268 consulting and updating the callout cache.
4270 70. Added support for saslauthd authentication, courtesy of Alexander
4273 71. If statvfs() failed on the spool or log directories while checking their
4274 size for availability, Exim confusingly gave the error "space shortage".
4275 Furthermore, in debugging mode it crashed with a floating point exception.
4276 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4277 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4278 serious problem, Exim now writes to the main and panic logs when this
4279 happens, with details of the failure. It then refuses to accept the
4280 incoming message, giving the message "spool directory problem" or "log
4281 directory problem" with a 421 code for SMTP messages.
4283 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4284 0, 1, and 2 exist, because some OS complain for execs without them (see
4285 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4286 descriptors. However, the code omitted to check that the open succeeded,
4287 causing mysterious errors if for some reason the permissions on /dev/null
4288 got screwed. Now Exim writes a message to the main and panic logs, and
4289 bombs out if it can't open /dev/null.
4291 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4292 interact so that it is all more flexible. It is supposed to remain
4293 backwards compatible. Also added extra_local_interfaces.
4295 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4296 to bomb out with an assertion failure - to the client this appears as a
4297 connection drop. This problem occurs in the part of the code that was taken
4298 from the Samba project. Fortunately, the assertion is in a very simple
4299 function, so I have fixed this by reproducing the function inline in the
4300 one place where it is called, and arranging for authentication to fail
4301 instead of killing the process with assert().
4303 75. The SPA client code was not working when the server requested OEM rather
4304 than Unicode encoding.
4306 76. Added code to make require_files with a specific uid setting more usable in
4307 the case where statting the file as root fails - usually a non-root-mounted
4308 NFS file system. When this happens and the failure is EACCES, Exim now
4309 forks a subprocess and does the per-uid checking as the relevant uid.
4311 77. Added process_log_path.
4313 78. If log_file_path was not explicitly set, a setting of check_log_space or
4314 check_log_inodes was ignored.
4316 79. If a space check for the spool or log partitions fails, the incident is now
4317 logged. Of course, in the latter case the data may get lost...
4319 80. Added the %p formatting code to string_format() so that it can be used to
4320 print addresses in debug_print(). Adjusted all the address printing in the
4321 debugging in store.c to use %p rather than %d.
4323 81. There was a concern that a line of code in smtp_in.c could overflow a
4324 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4325 initially expressed, the concern was not well-founded, because trailing
4326 spaces are removed early. However, if the trailing spaces were followed by
4327 a NULL, they did not get removed, so the overflow was possible. Two fixes
4330 (a) I re-wrote the offending code in a cleaner fashion.
4331 (b) If an incoming SMTP command contains a NULL character, it is rejected
4334 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4335 runs initgroups(), so that if the Exim user is in any additional groups,
4336 they will be used during message reception.
4342 The change log for 4.20 and earlier releases has been archived.