1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.480 2007/02/14 15:33:40 ph10 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
10 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
13 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
16 PH/02 In an ACL statement such as
18 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
20 if a client was not listed at all, or was listed with a value other than
21 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
22 the condition was not true (as it should be), so access was not denied.
23 The bug was that the ! inversion was incorrectly passed on to the second
24 item. This has been fixed.
26 PH/03 Added additional dnslists conditions == and =& which are different from
27 = and & when the dns lookup returns more than one IP address.
29 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
30 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
32 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
33 FSYNC, which compiles an option called disable_fsync that allows for
34 bypassing fsync(). The documentation is heavily laced with warnings.
36 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
38 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
39 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
40 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
41 including adding "make clean"; (3) Added -fPIC when compiling the test
42 dynamically loaded module, to get rid of a warning.
44 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
45 message fails, move_frozen_messages = true and ignore_bounce_errors_after
46 = 0s. The bug is otherwise harmless.
48 PH/07 There was a bug in the dovecot authenticator such that the value of
49 $auth1 could be overwritten, and so not correctly preserved, after a
50 successful authentication. This usually meant that the value preserved by
51 the server_setid option was incorrect.
53 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
55 PH/09 Installed PCRE release 7.0.
57 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
58 run for batched SMTP input. It is now run at the start of every message
59 in the batch. While fixing this I discovered that the process information
60 (output by running exiwhat) was not always getting set for -bs and -bS
61 input. This is fixed, and it now also says "batched" for BSMTP.
63 PH/11 Added control=no_pipelining.
65 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
66 patch, slightly modified), and move the expansion of helo_data till after
67 the connection is made in the smtp transport (so it can use these
70 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
72 PH/14 Added log_selector = +pid.
74 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
76 PH/16 Add ${if forany and ${if forall.
78 PH/17 Added dsn_from option to vary the From: line in DSNs.
80 PH/18 Flush SMTP output before performing a callout, unless control =
81 no_callout_flush is set.
83 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
84 was true (the default) a successful delivery failed to delete the retry
85 item, thus causing premature timeout of the address. The bug is now
88 PH/20 Added hosts_avoid_pipelining to the smtp transport.
90 PH/21 Long custom messages for fakedefer and fakereject are now split up
91 into multiline reponses in the same way that messages for "deny" and
92 other ACL rejections are.
94 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
95 with slight modification.
97 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
98 draft, changing the syntax and factoring some duplicate code".
100 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
101 for deliveries of the second and subsequent messages over the same SMTP
104 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
105 ${reduce, with only minor "tidies".
107 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
109 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
110 expansion side effects.
116 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
117 fixed by 4.65/MH/01 (is this a record?) are fixed:
119 (i) An empty string was always treated as zero by the numeric comparison
120 operators. This behaviour has been restored.
122 (ii) It is documented that the numeric comparison operators always treat
123 their arguments as decimal numbers. This was broken in that numbers
124 starting with 0 were being interpreted as octal.
126 While fixing these problems I realized that there was another issue that
127 hadn't been noticed. Values of message_size_limit (both the global option
128 and the transport option) were treated as octal if they started with 0.
129 The documentation was vague. These values are now always treated as
130 decimal, and I will make that clear in the documentation.
136 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
137 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
140 MH/01 Don't check that the operands of numeric comparison operators are
141 integers when their expansion is in "skipping" mode (fixes bug
142 introduced by 4.64-PH/07).
144 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
145 child addresses, Exim now panics and dies. Previously, because the count
146 is held in a short int, deliveries were likely to be lost. As such a
147 large number of recipients for a single message is ridiculous
148 (performance will be very, very poor), I have chosen to impose a limit
149 rather than extend the field.
155 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
156 leftover -K file (the existence of which was triggered by #402).
157 While we were at it, introduced process PID as part of the -K
158 filename. This should rule out race conditions when creating
161 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
162 processing considerably. Previous code took too long for large mails,
163 triggering a timeout which in turn triggers #401.
165 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
166 in the DK code in transports.c. sendfile() is not really portable,
167 hence the _LINUX specificness.
169 TF/01 In the add_headers option to the mail command in an Exim filter,
170 there was a bug that Exim would claim a syntax error in any
171 header after the first one which had an odd number of characters
174 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
175 callout verification, Exim cached a "reject" for the entire domain. This
176 is correct for most verifications, but it is not correct for a recipient
177 verification with use_sender or use_postmaster set, because in that case
178 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
179 case of MAIL FROM:<> rejection from other early rejections (e.g.
180 rejection of HELO). When verifying a recipient using a non-null MAIL
181 address, the cache is ignored if it shows MAIL FROM:<> rejection.
182 Whatever the result of the callout, the value of the domain cache is
183 left unchanged (for any other kind of callout, getting as far as trying
184 RCPT means that the domain itself is ok).
186 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
189 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
190 manifest itself as EPIPE rather than ECONNECT. When tidying away a
191 session, the daemon ignores ECONNECT errors and logs others; it now
192 ignores EPIPE as well.
194 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
195 (quoted-printable decoding).
197 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
198 later the small subsequent patch to fix an introduced bug.
200 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
202 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
204 PH/08 An error is now given if message_size_limit is specified negative.
206 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
207 to be given (somewhat) arbitrary names.
209 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
212 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
213 miscellaneous code fixes
215 PH/10 Added the log_reject_target ACL modifier to specify where to log
218 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
219 hostname. This is wrong, because it relates to the incoming message (and
220 probably the interface on which it is arriving) and not to the outgoing
221 callout (which could be using a different interface). This has been
222 changed to use the value of the helo_data option from the smtp transport
223 instead - this is what is used when a message is actually being sent. If
224 there is no remote transport (possible with a router that sets up host
225 addresses), $smtp_active_hostname is used.
227 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
228 tweaks were necessary in order to get it to work (see also 21 below):
229 (a) The code assumed that strncpy() returns a negative number on buffer
230 overflow, which isn't the case. Replaced with Exim's string_format()
232 (b) There were several signed/unsigned issues. I just did the minimum
233 hacking in of casts. There is scope for a larger refactoring.
234 (c) The code used strcasecmp() which is not a standard C function.
235 Replaced with Exim's strcmpic() function.
236 (d) The code set only $1; it now sets $auth1 as well.
237 (e) A simple test gave the error "authentication client didn't specify
238 service in request". It would seem that Dovecot has changed its
239 interface. Fortunately there's a specification; I followed it and
240 changed what the client sends and it appears to be working now.
242 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
245 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
246 address A is aliased to B and C, where B exists and C does not. Without
247 -v the output is "A verified" because verification stops after a
248 successful redirection if more than one address is generated. However,
249 with -v the child addresses are also verified. Exim was outputting "A
250 failed to verify" and then showing the successful verification for C,
251 with its parentage. It now outputs "B failed to verify", showing B's
252 parentage before showing the successful verification of C.
254 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
255 look up a TXT record in a specific list after matching in a combined
258 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
259 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
260 they consult the DNS. I had assumed they would set it the way they
261 wanted; and indeed my experiments on Linux seem to show that in some
262 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
263 To be on the safe side, however, I have now made the interface to
264 host_find_byname() similar to host_find_bydns(), with an argument
265 containing the DNS resolver options. The host_find_byname() function now
266 sets these options at its start, just as host_find_bydns() does. The smtp
267 transport options dns_qualify_single and dns_search_parents are passed to
268 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
269 of host_find_byname() use the default settings of RES_DEFNAMES
270 (qualify_single) but not RES_DNSRCH (search_parents).
272 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
273 spool_read_header() do less string testing, by means of a preliminary
274 switch on the second character of optional "-foo" lines. (This is
275 overdue, caused by the large number of possibilities that now exist.
276 Originally there were few.) While I was there, I also converted the
277 str(n)cmp tests so they don't re-test the leading "-" and the first
278 character, in the hope this might squeeze out yet more improvement.
280 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
281 flag allowing group syntax was set by the header_syntax check but not
282 turned off, possible causing trouble later; (2) The flag was not being
283 set at all for the header_verify test, causing "group"-style headers to
284 be rejected. I have now set it in this case, and also caused header_
285 verify to ignore an empty address taken from a group. While doing this, I
286 came across some other cases where the code for allowing group syntax
287 while scanning a header line wasn't quite right (mostly, not resetting
288 the flag correctly in the right place). These bugs could have caused
289 trouble for malformed header lines. I hope it is now all correct.
291 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
292 with the "reply" argument non-NULL. The code, however (which originally
293 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
294 but it didn't always do it. This confused somebody who was copying the
295 code for some other use. I have removed all the tests.
297 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
298 feature that was used to support insecure browsers during the U.S. crypto
299 embargo. It requires special client support, and Exim is probably the
300 only MTA that supported it -- and would never use it because real RSA is
301 always available. This code has been removed, because it had the bad
302 effect of slowing Exim down by computing (never used) parameters for the
303 RSA_EXPORT functionality.
305 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
306 authenticator to fail if there's a tab character in the incoming data
307 (there should never be unless someone is messing about, as it's supposed
308 to be base64-encoded). Also added, on Timo's advice, the "secured" option
309 if the connection is using TLS or if the remote IP is the same as the
310 local IP, and the "valid-client-cert option" if a client certificate has
313 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
314 authenticators. This can be used for authorization after authentication
315 succeeds. (In the case of plaintext, it servers for both authentication
318 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
319 if any retry times were supplied.
321 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
322 connection, where there is no client IP address to check. In this
323 situation, the verify now always succeeds.
325 PH/25 Applied John Jetmore's -Mset patch.
327 PH/26 Added -bem to be like -Mset, but loading a message from a file.
329 PH/27 In a string expansion for a processed (not raw) header when multiple
330 headers of the same name were present, leading whitespace was being
331 removed from all of them, but trailing whitespace was being removed only
332 from the last one. Now trailing whitespace is removed from each header
333 before concatenation. Completely empty headers in a concatenation (as
336 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
337 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
339 PH/29 [Removed. This was a change that I later backed out, and forgot to
340 correct the ChangeLog entry (that I had efficiently created) before
341 committing the later change.]
343 PH/30 Exim was sometimes attempting to deliver messages that had suffered
344 address errors (4xx response to RCPT) over the same connection as other
345 messages routed to the same hosts. Such deliveries are always "forced",
346 so retry times are not inspected. This resulted in far too many retries
347 for the affected addresses. The effect occurred only when there were more
348 hosts than the hosts_max_try setting in the smtp transport when it had
349 the 4xx errors. Those hosts that it had tried were not added to the list
350 of hosts for which the message was waiting, so if all were tried, there
351 was no problem. Two fixes have been applied:
353 (i) If there are any address or message errors in an SMTP delivery, none
354 of the hosts (tried or untried) are now added to the list of hosts
355 for which the message is waiting, so the message should not be a
356 candidate for sending over the same connection that was used for a
357 successful delivery of some other message. This seems entirely
358 reasonable: after all the message is NOT "waiting for some host".
359 This is so "obvious" that I'm not sure why it wasn't done
360 previously. Hope I haven't missed anything, but it can't do any
361 harm, as the worst effect is to miss an optimization.
363 (ii) If, despite (i), such a delivery is accidentally attempted, the
364 routing retry time is respected, so at least it doesn't keep
365 hammering the server.
367 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
368 in ${readsocket because some servers need this prod.
370 PH/32 Added some extra debug output when updating a wait-xxx database.
372 PH/33 The hint "could be header name not terminated by colon", which has been
373 given for certain expansion errors for a long time, was not being given
374 for the ${if def:h_colon_omitted{... case.
376 PH/34 The spec says: "With one important exception, whenever a domain list is
377 being scanned, $domain contains the subject domain." There was at least
378 one case where this was not true.
380 PH/35 The error "getsockname() failed: connection reset by peer" was being
381 written to the panic log as well as the main log, but it isn't really
382 panic-worthy as it just means the connection died rather early on. I have
383 removed the panic log writing for the ECONNRESET error when getsockname()
386 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
387 runs only) independently of the message's sender address. This meant
388 that, if the 4xx error was in fact related to the sender, a different
389 message to the same recipient with a different sender could confuse
390 things. In particualar, this can happen when sending to a greylisting
391 server, but other circumstances could also provoke similar problems.
392 I have changed the default so that the retry time for these errors is now
393 based a combination of the sender and recipient addresses. This change
394 can be overridden by setting address_retry_include_sender=false in the
397 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
398 remote server are returned as part of bounce messages. This was not
399 happening for LMTP over a pipe (the lmtp transport), but now it is the
400 same for both kinds of LMTP.
402 PH/38 Despite being documented as not happening, Exim was rewriting addresses
403 in header lines that were in fact CNAMEs. This is no longer the case.
405 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
406 and queue runs started by the daemon processed all messages. This has
407 been fixed so that -R and -S can now usefully be given with -q<time>.
409 PH/40 Import PCRE release 6.7 (fixes some bugs).
411 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
413 PH/42 Give an error if -q is specified more than once.
415 PH/43 Renamed the variables $interface_address and $interface_port as
416 $received_ip_address and $received_port, to make it clear that these
417 values apply to message reception, and not to the outgoing interface when
418 a message is delivered. (The old names remain recognized, of course.)
420 PH/44 There was no timeout on the connect() call when using a Unix domain
421 socket in the ${readsocket expansion. There now is.
423 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
424 be meaningful with "accept".
426 SC/01 Eximstats V1.43
427 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
429 SC/02 Eximstats V1.44
430 Use a glob alias rather than an array ref in the generated
431 parser. This improves both readability and performance.
433 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
434 Collect SpamAssassin and rejection statistics.
435 Don't display local sender or destination tables unless
436 there is data to show.
437 Added average volumes into the top table text output.
439 SC/04 Eximstats V1.46
440 Collect data on the number of addresses (recipients)
441 as well as the number of messages.
443 SC/05 Eximstats V1.47
444 Added 'Message too big' to the list of mail rejection
445 reasons (thanks to Marco Gaiarin).
447 SC/06 Eximstats V1.48
448 Mainlog lines which have GMT offsets and are too short to
449 have a flag are now skipped.
451 SC/07 Eximstats V1.49 (Alain Williams)
452 Added the -emptyok flag.
454 SC/08 Eximstats V1.50
455 Fixes for obtaining the IP address from reject messages.
457 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
458 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
459 whitesspace changes from 4.64-PH/27
461 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
464 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
465 are found, allow negative numbers in numeric criteria)
467 JJ/06 exipick.20061117.2, added new $message_body_missing variable
469 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
470 to match changes made in 4.64-PH/43
472 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
474 PH/47 Put in an explicit test for a DNS lookup of an address record where the
475 "domain" is actually an IP address, and force a failure. This locks out
476 those revolvers/nameservers that support "A-for-A" lookups, in
477 contravention of the specifications.
479 PH/48 When a host name was looked up from an IP address, and the subsequent
480 forward lookup of the name timed out, the host name was left in
481 $sender_host_name, contrary to the specification.
483 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
484 restricted to single-key lookups, Exim was not diagnosing an error if
485 * or *@ was used with a query-style lookup.
487 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
489 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
490 long ago, but noone interested enough thought of it. Let's just say that
491 the "1.1" means that there are some new functions that weren't there at
492 some point in the past.
494 PH/51 Error processing for expansion failure of helo_data from an smtp
495 transport during callout processing was broken.
497 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
498 tested/used via the -bh/-bhc/-bs options.
500 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
501 bug, fixed in subsequent PCRE releases).
503 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
504 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
506 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
512 SC/01 Use a glob alias rather than an array ref in eximstats generated
513 parser. This improves both readability and performance.
515 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
516 Don't display local sender or destination tables in eximstats unless
517 there is data to show.
518 Added average volumes into the eximstats top table text output.
520 SC/03 Collect data on the number of addresses (recipients) as well
521 as the number of messages in eximstats.
523 TF/01 Correct an error in the documentation for the redirect router. Exim
524 does (usually) call initgroups() when daemonizing.
526 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
527 with consistent privilege compared to when running as a daemon.
529 TF/03 Note in the spec that $authenticated_id is not set for local
530 submissions from trusted users.
532 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
533 Thanks to Dean Brooks <dean@iglou.com> for the patch.
535 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
536 by adding some example configuration directives to the default
537 configuration file. A little bit of work is required to uncomment the
538 directives and define how usernames and passwords are checked, but
539 there is now a framework to start from.
541 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
542 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
543 without this. I don't know how relevant this is to other LDAP libraries.
545 PH/02 Add the verb name to the "unknown ACL verb" error.
547 PH/03 Magnus Holmgren's patch for filter_prepend_home.
549 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
551 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
552 directory not expanded when it should be if an expanded home directory
553 was set for the address (which is overridden by the transport).
555 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
558 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
559 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
560 because it is too late at that time, and has no effect.
562 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
563 security issue with \' (bugzilla #107). I could not use the
564 PQescapeStringConn() function, because it needs a PGconn value as one of
567 PH/08 When testing addresses using -bt, indicate those final addresses that
568 are duplicates that would not cause an additional delivery. At least one
569 person was confused, thinking that -bt output corresponded to deliveries.
570 (Suppressing duplicates isn't a good idea as you lose the information
571 about possibly different redirections that led to the duplicates.)
573 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
574 systems where poll() doesn't work, in particular OS X.
576 PH/10 Added more information to debugging output for retry time not reached.
578 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
579 operations in malware.c.
581 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
584 PH/13 If write_rejectlog was set false when logging was sent to syslog with
585 syslog_duplication set false, log lines that would normally be written
586 both the the main log and to the reject log were not written to syslog at
589 PH/14 In the default configuration, change the use of "message" in ACL warn
590 statements to "add_header".
592 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
593 not followed by a command (e.g. "seen endif").
595 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
596 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
599 PH/17 Added extra conditions to the default value of delay_warning_condition
603 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
604 { match{$h_precedence:}{(?i)bulk|list|junk} } \
605 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
608 The Auto-Submitted: and various List- headers are standardised, whereas I
609 don't think Precedence: ever was.
611 PH/18 Refactored debugging code in route_finduser() to show more information,
612 in particular, the error code if getpwnam() issues one.
614 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
615 This is apparently needed in addition to the PH/07 change above to avoid
616 any possible encoding problems.
618 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
619 but not after initializing Perl.
621 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
622 output them only if debugging. By default they are written stderr,
623 apparently, which is not desirable.
625 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
628 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
631 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
633 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
634 authenticated or an ident call has been made. Suppress the default
635 values for $authenticated_id and $authenticated_sender (but permit -oMai
636 and -oMas) when testing with -bh.
638 PH/24 Re-jigged the order of the tests in the default configuration so that the
639 tests for valid domains and recipients precede the DNS black list and CSA
640 tests, on the grounds that those ones are more expensive.
642 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
643 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
644 This bug exists in every version of Exim that I still have, right back to
647 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
648 However, an attempt to turn on case-sensitivity in a regex key by
649 including (?-i) didn't work because the subject string was already
650 lowercased, and the effects were non-intuitive. It turns out that a
651 one-line patch can be used to allow (?-i) to work as expected.
657 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
658 other effects) broke the use of negated acl sub-conditions.
660 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
663 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
664 "Deny" causes Exim to reject the incoming connection with a 554 error.
665 Unfortunately, if there is a major crisis, such as a disk failure,
666 tcp-wrappers gives "deny", whereas what one would like would be some
667 kind of temporary error. A kludge has been added to help with this.
668 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
669 554 error is used if errno is still zero or contains ENOENT (which occurs
670 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
673 PH/03 Add -lutil to the default FreeBSD LIBS setting.
675 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
676 errors. Otherwise a message that provokes a temporary error (when other
677 messages do not) can cause a whole host to time out.
679 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
680 addresses were routed directly to files or pipes from a redirect router.
681 File deliveries just didn't batch; pipe deliveries might have suffered
684 PH/06 A failure to get a lock for a hints database would erroneously always say
685 "Failed to get write lock", even when it was really a read lock.
687 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
688 of 0600. This has been changed to use the value of the lockfile_mode
689 option (which defaults to 0600).
691 PH/08 Applied small patch from the Sieve maintainer.
693 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
694 folder from quota calculations, a direct delivery into this folder messed
695 up the contents of the maildirsize file. This was because the regex was
696 used only to exclude .Trash (or whatever) when the size of the mailbox
697 was calculated. There was no check that a delivery was happening into an
698 excluded directory. This bug has been fixed by ignoring all quota
699 processing for deliveries into excluded directories.
701 PH/10 Added the maildirfolder_create_regex option to appendfile.
707 PH/01 The code for finding all the local interface addresses on a FreeBSD
708 system running IPv6 was broken. This may well have applied to all BSD
709 systems, as well as to others that have similar system calls. The broken
710 code found IPv4 interfaces correctly, but gave incorrect values for the
711 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
712 that it would not match correctly against @[] and not recognize the IPv6
715 PH/02 The ipliteral router was not recognizing addresses of the form user@
716 [ipv6:....] because it didn't know about the "ipv6:" prefix.
718 PH/03 Added disable_ipv6.
720 PH/04 Changed $reply_address to use the raw form of the headers instead of the
721 decoded form, because it is most often used to construct To: headers
722 lines in autoreplies, and the decoded form may well be syntactically
723 invalid. However, $reply_address has leading white space removed, and all
724 newlines turned into spaces so that the autoreply transport does not
727 PH/05 If group was specified without a user on a router, and no group or user
728 was specified on a transport, the group from the router was ignored.
730 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
731 for visible compile-time settings that can be used to change these
732 numbers, for those that want even more. Backwards compatibility with old
733 spool files has been maintained. However, going back to a previous Exim
734 release will lost any variables that are in spool files.
736 PH/07 Two small changes when running in the test harness: increase delay when
737 passing a TCP/IP connection to a new process, in case the original
738 process has to generate a bounce, and remove special handling of
739 127.0.0.2 (sic), which is no longer necessary.
741 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
742 be the same on different OS.
744 PH/09 Moved a debug statement in filter processing to avoid a race problem when
747 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
748 whether --show-vars was specified or not
750 JJ/02 exipick: Added support for new ACL variable spool format introduced
753 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
754 syntactically invalid From: or Reply-to: line, and a filter used this to
755 generate an autoreply, and therefore failed to obtain an address for the
756 autoreply, Exim could try to deliver to a non-existent relative file
757 name, causing unrelated and misleading errors. What now happens is that
758 it logs this as a hard delivery error, but does not attempt to create a
761 PH/11 The exinext utility has a -C option for testing purposes, but although
762 the given file was scanned by exinext itself; it wasn't being passed on
763 when Exim was called.
765 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
766 an end-of-file indication when reading a command response.
768 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
769 compiled. In many other places in Exim, IPv6 addresses are always
770 recognized, so I have changed this. It also means that IPv4 domain
771 literals of the form [IPV4:n.n.n.n] are now always recognized.
773 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
774 used if the router is not running as root, for example, when verifying at
775 ACL time, or when using -bh. The debugging output from this situation was
776 non-existent - all you got was a failure to exec. I have made two
779 (a) Failures to set uid/gid, the current directory, or a process leader
780 in a subprocess such as that created by queryprogram now generate
781 suitable debugging ouput when -d is set.
783 (b) The queryprogram router detects when it is not running as root,
784 outputs suitable debugging information if -d is set, and then runs
785 the subprocess without attempting to change uid/gid.
787 PH/15 Minor change to Makefile for building test_host (undocumented testing
790 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
791 additional section of a DNS packet that returns MX or SRV records.
792 Instead, it always explicitly searches for A/AAAA records. This avoids
793 major problems that occur when a DNS server includes only records of one
794 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
795 fixed another bug: if SRV records were looked up and the corresponding
796 address records were *not* found in the additional section, the port
797 values from the SRV records were lost.
799 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
800 using the correct key (the original address) when searching the retry
801 rules in order to find which one to use for generating the retry hint.
803 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
804 adding the default one. Similarly, if it contains a Reply-To: header, the
805 errors_reply_to option, if set, is not used.
807 PH/19 When calculating a retry time, Exim used to measure the "time since
808 failure" by looking at the "first failed" field in the retry record. Now
809 it does not use this if it is later than than the arrival time of the
810 message. Instead it uses the arrival time. This makes for better
811 behaviour in cases where some deliveries succeed, thus re-setting the
812 "first failed" field. An example is a quota failure for a huge message
813 when small messages continue to be delivered. Without this change, the
814 "time since failure" will always be short, possible causing more frequent
815 delivery attempts for the huge message than are intended.
816 [Note: This change was subsequently modified - see PH/04 for 4.62.]
818 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
819 $1, $2, $3) because the numerical variables can be reset during some
820 expansion items (e.g. "match"), thereby losing the authentication data.
822 PH/21 Make -bV show the size of off_t variables so that the test suite can
823 decide whether to run tests for quotas > 2G.
825 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
826 mailbox_size, and mailbox_filecount in the appendfile transport. If a
827 filecount value is greater than 2G or if a quota value is greater than 2G
828 on a system where the size of off_t is not greater than 4, a panic error
831 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
832 never match. The debug and -bh output now contains an explicit error
833 message indicating a malformed IPv4 address or mask.
835 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
836 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
839 PH/25 Do not write to syslog when running in the test harness. The only
840 occasion when this arises is a failure to open the main or panic logs
841 (for which there is an explicit test).
843 PH/26 Added the /no_tell option to "control=freeze".
845 PH/27 If a host name lookup failed very early in a connection, for example, if
846 the IP address matched host_lookup and the reverse lookup yielded a name
847 that did not have a forward lookup, an error message of the form "no IP
848 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
849 could be logged. Now it outputs the IP address instead of "NULL".
851 PH/28 An enabling patch from MH: add new function child_open_exim2() which
852 allows the sender and the authenticated sender to be set when
853 submitting a message from within Exim. Since child_open_exim() is
854 documented for local_scan(), the new function should be too.
856 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
857 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
858 results in an empty string is now treated as unset.
860 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
862 PH/31 Added sender_verify_fail logging option.
864 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
865 needed by RFC 822 but not by RFC 2822 was commented out. I have now
866 tidied the source and removed it altogether.
868 PH/33 When a queue run was abandoned because the load average was too high, a
869 log line was always written; now it is written only if the queue_run log
870 selector is set. In addition, the log line for abandonment now contains
871 information about the queue run such as the pid. This is always present
872 in "start" and "stop" lines but was omitted from the "abandon" line.
874 PH/34 Omit spaces between a header name and the colon in the error message that
875 is given when verify = headers_syntax fails (if there are lots of them,
876 the message gets confusing).
878 PH/35 Change the default for dns_check_names_pattern to allow slashes within
879 names, as there are now some PTR records that contain slashes. This check
880 is only to protect against broken name servers that fall over on strange
881 characters, so the fact that it applies to all lookups doesn't matter.
883 PH/36 Now that the new test suite is complete, we can remove some of the
884 special code in Exim that was needed for the old test suite. For example,
885 sorting DNS records because real resolvers return them in an arbitrary
886 order. The new test suite's fake resolver always returns records in the
889 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
890 bounces) except when queue_only is set, to avoid logging races between
891 the different processes.
893 PH/38 Panic-die if .include specifies a non-absolute path.
895 PH/39 A tweak to the "H" retry rule from its user.
897 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
898 a label. They prevented compilation on older perls.
900 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
901 a warning to be raised on newish perls.
903 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
904 on queue. Changes to match documented behaviour of showing count of
905 messages matching specified criteria.
907 PH/40 Changed the default ident timeout from 30s to 5s.
909 PH/41 Added support for the use of login_cap features, on those BSD systems
910 that have them, for controlling the resources used by pipe deliveries.
912 PH/42 The content-scanning code uses fopen() to create files in which to put
913 message data. Previously it was not paying any attention to the mode of
914 the files. Exim runs with umask(0) because the rest of the code creates
915 files with open(), and sets the required mode explicitly. Thus, these
916 files were ending up world-writeable. This was not a big issue, because,
917 being within the spool directory, they were not world-accessible. I have
918 created a function called modefopen, which takes an additional mode
919 argument. It sets umask(777), creates the file, chmods it to the required
920 mode, then resets the umask. All the relevant calls to fopen() in the
921 content scanning code have been changed to use this function.
923 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
924 to 24 hours. This avoids potential overflow problems when processing G
925 and H retry rules. I suspect nobody ever tinkers with this value.
927 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
929 PH/45 When the plaintext authenticator is running as a client, the server's
930 challenges are checked to ensure they are valid base64 strings. By
931 default, the authentication attempt is cancelled if an invalid string is
932 received. Setting client_ignore_invalid_base64 true ignores these errors.
933 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
934 they are received. Thus, the responses can be made to depend on the
935 challenges. If an invalid string is ignored, an empty string is placed in
938 PH/46 Messages that are created by the autoreply transport now contains a
939 References: header, in accordance with RFCs 2822 and 3834.
941 PH/47 Added authenticated_sender_force to the smtp transport.
943 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
945 PH/49 Installed latest patch from the Sieve maintainer.
947 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
948 was also set, the appendfile transport was unnecessarily scanning a
949 directory of message files (e.g. for maildir delivery) to find the count
950 of files (along with the size), even though it did not need this
951 information. It now does the scan only if it needs to find either the
952 size of the count of files.
954 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
956 PH/52 Two bugs concerned with error handling when the smtp transport is
959 (i) Exim was not creating retry information for temporary errors given
960 for individual recipients after the DATA command when the smtp transport
961 was used in LMTP mode. This meant that they could be retried too
962 frequently, and not timed out correctly.
964 (ii) Exim was setting the flag that allows error details to be returned
965 for LMTP errors on RCPT commands, but not for LMTP errors for individual
966 recipients that were returned after the DATA command.
968 PH/53 This is related to PH/52, but is more general: for any failing address,
969 when detailed error information was permitted to be returned to the
970 sender, but the error was temporary, then after the final timeout, only
971 "retry timeout exceeded" was returned. Now it returns the full error as
972 well as "retry timeout exceeded".
974 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
975 do this, and (what is worse) MTAs that accept it.
977 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
978 will now be deprecated.
980 PH/56 New os.c-cygwin from the Cygwin maintainer.
982 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
983 formats (previously only available in exim formats via -bpr, -bpru,
984 and -bpra. Now also available in native and exiqgrep formats)
986 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
987 with very large, slow to parse queues
989 JJ/08 exipick: added ! as generic prefix to negate any criteria format
991 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
993 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
994 responses to authentication challenges, though it was showing the
995 challenges; (ii) I've removed the CR characters from the debug output for
998 PH/58 Allow for the insertion of a newline as well as a space when a string
999 is turned into more than one encoded-word during RFC 2047 encoding. The
1000 Sieve code now uses this.
1002 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1003 data_4xx, lost_connection, tls_required.
1005 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1006 message was being sent as an SMTP response.
1008 PH/61 Add -l and -k options to exicyclog.
1010 PH/62 When verifying, if an address was redirected to one new address, so that
1011 verification continued, and the new address failed or deferred after
1012 having set something in $address_data, the value of $address_data was not
1013 passed back to the ACL. This was different to the case when no
1014 redirection occurred. The value is now passed back in both cases.
1016 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1017 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1018 use login_cap.h, so on its own it isn't the distinguishing feature. The
1019 new name refers directly to the setclassresources() function.
1021 PH/65 Added configuration files for NetBSD3.
1023 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1025 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1026 is preferred over IPv4.
1028 PH/68 The bounce_return_message and bounce_return_body options were not being
1029 honoured for bounces generated during the reception of non-SMTP messages.
1030 In particular, this applied to messages rejected by the ACL. This bug has
1031 been fixed. However, if bounce_return_message is true and bounce_return_
1032 body is false, the headers that are returned for a non-SMTP message
1033 include only those that have been read before the error was detected.
1034 (In the case of an ACL rejection, they have all been read.)
1036 PH/69 The HTML version of the specification is now built in a directory called
1037 spec_html instead of spec.html, because the latter looks like a path with
1038 a MIME-type, and this confuses some software.
1040 PH/70 Catch two compiler warnings in sieve.c.
1042 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1043 function verify_get_ident() calls ip_connect() to connect a socket, but
1044 if the "connect()" function timed out, ip_connect() used to close the
1045 socket. However, verify_get_ident() also closes the socket later, and in
1046 between Exim writes to the log, which may get opened at this point. When
1047 the socket was closed in ip_connect(), the log could get the same file
1048 descriptor number as the socket. This naturally causes chaos. The fix is
1049 not to close the socket in ip_connect(); the socket should be closed by
1050 the function that creates it. There was only one place in the code where
1051 this was missing, in the iplookup router, which I don't think anybody now
1052 uses, but I've fixed it anyway.
1054 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1055 well as to direct DNS lookups. Otherwise the handling of names in host
1056 lists is inconsistent and therefore confusing.
1062 PH/01 Two changes to the default runtime configuration:
1064 (1) Move the checks for relay_from_hosts and authenticated clients from
1065 after to before the (commented out) DNS black list checks.
1067 (2) Add control=submission to the relay_from_hosts and authenticated
1068 clients checks, on the grounds that messages accepted by these
1069 statements are most likely to be submissions.
1071 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1073 (1) Generate an error if the third argument for the ${prvs expansion is
1076 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1079 (3) Reset the variables that are obtained from the first argument of
1080 ${prvscheck and used in the second argument before leaving the code,
1081 because their memory is reclaimed, so using them afterwards may do
1084 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1085 one (it's much easier than Tom thought :-).
1087 (5) Because of (4), we can now allow for the use of $prvscheck_result
1088 inside the third argument.
1090 PH/03 For some reason, the default setting of PATH when running a command from
1091 a pipe transport was just "/usr/bin". I have changed it to
1094 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1095 anything to be listed in the output from -bV.
1097 PH/05 When a filter generated an autoreply, the entire To: header line was
1098 quoted in the delivery log line, like this:
1100 => >A.N.Other <ano@some.domain> <original@ddress> ...
1102 This has been changed so that it extracts the operative address. There
1103 may be more than one such address. If so, they are comma-separated, like
1106 => >ano@some.domain,ona@other.domain <original@ddress> ...
1108 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1109 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1110 not being looked up in the rDNS to get a host name, Exim was showing the
1111 IP address twice in Received: lines, even though the IP addresses were
1112 identical. For example:
1114 Received: from [1.2.3.4] (helo=[1.2.3.4])
1116 However, if the real host name was known, it was omitting the HELO data
1117 if it matched the actual IP address. This has been tidied up so that it
1118 doesn't show the same IP address twice.
1120 PH/07 When both +timestamp and +memory debugging was on, the value given by
1121 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1122 called by the debug printing, thereby overwriting the timestamp buffer.
1123 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1125 PH/08 When the original message was included in an autoreply transport, it
1126 always said "this is a copy of the message, including all the headers",
1127 even if body_only or headers_only was set. It now gives an appropriate
1130 PH/09 Applied a patch from the Sieve maintainer which:
1132 o fixes some comments
1133 o adds the (disabled) notify extension core
1134 o adds some debug output for the result of if/elsif tests
1135 o points to the current vacation draft in the documentation
1136 and documents the missing references header update
1140 o fixes a bug in processing the envelope test (when testing
1141 multiple envelope elements, the last element determinted the
1144 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1145 Electronic Mail") by including:
1147 Auto-submitted: auto-generated
1149 in the messages that it generates (bounce messages and others, such as
1150 warnings). In the case of bounce messages for non-SMTP mesages, there was
1151 also a typo: it was using "Auto_submitted" (underscore instead of
1152 hyphen). Since every message generated by Exim is necessarily in response
1153 to another message, thes have all been changed to:
1155 Auto-Submitted: auto-replied
1157 in accordance with these statements in the RFC:
1159 The auto-replied keyword:
1161 - SHOULD be used on messages sent in direct response to another
1162 message by an automatic process,
1164 - MUST NOT be used on manually-generated messages,
1166 - MAY be used on Delivery Status Notifications (DSNs) and Message
1167 Disposition Notifications (MDNs),
1169 - MUST NOT be used on messages generated by automatic or periodic
1170 processes, except for messages which are automatic responses to
1173 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1174 to the default Received: header definition.
1176 PH/12 Added log selector acl_warn_skipped (default on).
1178 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1179 variables because (a) they are in the wrong storage pool and (b) even if
1180 they were copied, it wouldn't work properly because of the caching.
1182 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1183 checking when decoding. Apparently there are clients that generate
1184 overlong encoded strings. Why am I not surprised?
1186 PH/15 If the first argument of "${if match_address" was not empty, but did not
1187 contain an "@" character, Exim crashed. Now it writes a panic log message
1188 and treats the condition as false.
1190 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1192 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1193 extension "envelope-auth". The code is finished and in agreement with
1194 other implementations, but there is no documentation so far and in fact,
1195 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1196 not changing the active code.
1198 Print executed "if" and "elsif" statements when debugging is used. This
1199 helps a great deal to understand what a filter does.
1201 Document more things not specified clearly in RFC3028. I had all this
1202 sorted out, when out of a sudden new issues came to my mind. Oops."
1204 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1207 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1208 canonical form (as documented). However, after a host name lookup from
1209 the IP address, check_host() was doing a simple string comparison with
1210 addresses acquired from the DNS when checking that the found name did
1211 have the original IP as one of its addresses. Since any found IPv6
1212 addresses are likely to be in abbreviated form, the comparison could
1213 fail. Luckily, there already exists a function for doing the comparison
1214 by converting both addresses to binary, so now that is used instead of
1215 the text comparison.
1217 PH/20 There was another similar case to PH/19, when a complete host name was
1218 given in a host list; looking up its IP address could give an abbreviated
1219 form, whereas the current host's name might or might not be abbreviated.
1220 The same fix has been applied.
1226 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1227 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1230 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1231 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1233 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1235 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1236 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1237 Also, for "Auto-Submitted", treat anything other than "no" as
1238 non-personal, in accordance with RFC 3834. (Previously it treated
1239 anything starting "auto-" as non-personal.)
1241 TF/01 The control=submission/name=... option had a problem with syntax
1242 errors if the name included a slash character. The /name= option
1243 now slurps the rest of the string, so it can include any characters
1244 but it must come last in the list of options (after /sender_retain
1247 PH/05 Some modifications to the interface to the fake nameserver for the new
1255 TK/01 Added the "success_on_redirect" address verification option. See
1256 NewStuff for rationale and an example.
1258 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1260 PH/02 Patch to exigrep to allow it to work on syslog lines.
1262 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1263 fread() to skip over the body file's header line, because in Cygwin the
1264 header line is locked and is inaccessible.
1266 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1267 co-exist for some time) to make it clear that it is the Exim ID that is
1268 referenced, not the Message-ID: header line.
1270 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1271 string_format() function, because snprintf() does not exist on all
1274 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1275 ${stat: expansion item.
1277 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1278 protocol synchronization error", to keep the pedants happy.
1280 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1281 well as for IRIX systems, when gcc is being used. See the host.c source
1284 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1286 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1289 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1290 transport and to the smtp transport in LMTP mode.
1292 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1294 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1295 run a filter in a subprocess. This could lead to confusion in subsequent
1296 lookups in the parent process. There should also be a search_tidyup() at
1297 the end of the subprocess.
1299 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1300 only if the host matched helo_try_verify_hosts, which caused the
1301 verification to occur when the EHLO/HELO command was issued. The ACL just
1302 tested the remembered result. Now, if a previous verification attempt has
1303 not happened, "verify = helo" does it there and then.
1305 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1307 TK/03 Fix log output including CR from clamd.
1309 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1310 exist provoked a memory error which could cause a segfault.
1312 PH/15 Installed PCRE 6.2
1314 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1316 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1317 of the problem. Specifically, suggested +O2 rather than +O1 for the
1320 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1322 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1323 record, the port was not correctly logged when the outgoing_port log
1324 selector was set (it logged the transort's default port).
1326 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1327 fallback_hosts, and "hosts" in the smtp transport.
1329 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1330 host errors such as "Connection refused".
1332 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1333 authentication with radiusclient 0.4.9:
1335 - Error returned from rc_read_config was caught wrongly
1336 - Username/password not passed on to radius server due to wrong length.
1338 The presumption is that some radiusclient API changes for 4.51/PH/17
1339 were not taken care of correctly. The code is still untested by me (my
1340 Linux distribution still has 0.3.2 of radiusclient), but it was
1341 contributed by a Radius user.
1343 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1344 expanding the "port" option of the smtp transport.
1346 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1347 while reading a MIME header. Thanks to Tom Hughes for a patch.
1349 PH/24 Include config.h inside local_scan.h so that configuration settings are
1352 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1353 that in an ACL for RCPT (for example), you can examine exactly what was
1356 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1357 commands, but it was not correctly comparing the address with the actual
1358 client host address. Thus, it would show the EHLO address in Received:
1359 header lines when this was not necessary.
1361 PH/27 Added the % operator to ${eval:}.
1363 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1364 it should be ignoring failures (because with -C, for example, it has lost
1365 privilege). It wasn't ignoring creation failures other than "already
1368 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1371 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1372 because an input line was too long, either on its own, or by virtue of
1373 too many continuations, the temporary file was not being removed, and the
1374 return code was incorrect.
1376 PH/31 Missing "BOOL" in function definition in filtertest.c.
1378 PH/32 Applied Sieve patches from the maintainer.
1380 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1382 PH/33 Added "verify = not_blind".
1384 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1385 Local/Makefile (with some defaults set). These are used in built scripts
1386 such as exicyclog, but they have never been used in the exim_install
1387 script (though there are many overriding facilities there). I have
1388 arranged that the exim_install script now takes note of these two
1391 PH/35 Installed configuration files for Dragonfly.
1393 PH/36 When a locally submitted message by a trusted user did not contain a
1394 From: header, and the sender address was obtained from -f or from an SMTP
1395 MAIL command, and the trusted user did not use -F to supply a sender
1396 name, $originator_name was incorrectly used when constructing a From:
1397 header. Furthermore, $originator_name was used for submission mode
1398 messages from external hosts without From: headers in a similar way,
1399 which is clearly wrong.
1401 PH/37 Added control=suppress_local_fixups.
1403 PH/38 When log_selector = +received_sender was set, and the addition of the
1404 sender made the log line's construction buffer exactly full, or one byte
1405 less than full, an overflow happened when the terminating "\n" was
1408 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1409 when the result of a list match is failure because a DNS lookup failed.
1411 PH/40 RM_COMMAND is now used in the building process.
1413 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1414 the "build-* directories that it finds.
1416 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1417 address in a domain literal was a prefix of an interface address.
1419 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1420 when verifying a sender address, unless rewrite_headers is false.
1422 PH/44 Wrote a long comment about why errors_to addresses are verified as
1423 recipients, not senders.
1425 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1426 the ratelimit ACL was added.
1428 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1430 PH/46 Added extra information about PostgreSQL errors to the error string.
1432 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1433 avoiding the need to install special zones in a real server. This is
1434 backwards compatible; if it can't find the fake resolver, it drops back.
1435 Thus, both old and new test suites can be run.
1437 TF/02 Added util/ratelimit.pl
1439 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1440 clock is set back in time.
1442 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1443 Candler <B.Candler@pobox.com>.
1445 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1446 OK for addresses that are the result of redirections.
1448 PH/48 A number of further additions for the benefit of the new test suite,
1449 including a fake gethostbyname() that interfaces to the fake DNS resolver
1452 TF/06 The fix for widen_domains has also been applied to qualify_single and
1453 search_parents which are the other dnslookup options that can cause
1456 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1459 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1461 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1462 service" if it encountered an MX record with an empty target hostname.
1463 The message is now "an MX or SRV record indicated no SMTP service".
1465 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1466 if the DNS of the sending site is misconfigured. This is quite a
1467 common situation. This change restores the behaviour of treating a
1468 helo verification defer as a failure.
1470 PH/51 If self=fail was set on a router, the bounce message did not include the
1471 actual error message.
1477 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1479 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1480 command itself ended in error, the underlying message about the transport
1481 filter timeout was being overwritten with the pipe command error. Now the
1482 underlying error message should be appended to the second error message.
1484 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1486 PH/02 Reduce the amount of output that "make" produces by default. Full output
1487 can still be requested.
1489 PH/03 The warning log line about a condition test deferring for a "warn" verb
1490 was being output only once per connection, rather than after each
1491 occurrence (because it was using the same function as for successful
1492 "warn" verbs). This seems wrong, so I have changed it.
1494 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1495 it should not have, which might have caused a crash in the right
1496 circumstances, but probably never did.
1498 PH/04 Installed a modified version of Tony Finch's patch to make submission
1499 mode fix the return path as well as the Sender: header line, and to
1500 add a /name= option so that you can make the user's friendly name appear
1503 TF/03 Added the control = fakedefer ACL modifier.
1505 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1506 Mark Lowes for thorough testing.
1508 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1510 TK/03 Merged latest SRS patch from Miles Wilton.
1512 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1513 with the definition in sysexits.h (which is #included earlier).
1514 Fortunately, Exim does not actually use EX_OK. The code used to try to
1515 preserve the sysexits.h value, by assumimg that macro definitions were
1516 scanned for macro replacements. I have been disabused of this notion,
1517 so now the code just undefines EX_OK before #including unistd.h.
1519 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1520 in the smtp transport. When a block could not be written in a single
1521 write() function, the timeout was being re-applied to each part-write.
1522 This seems wrong - if the receiver was accepting one byte at a time it
1523 would take for ever. The timeout is now adjusted when this happens. It
1524 doesn't have to be particularly precise.
1526 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1527 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1529 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1530 without a domain if the check to <postmaster@domain> fails.
1532 SC/01 Eximstats: added -xls and the ability to specify output files
1533 (patch written by Frank Heydlauf).
1535 SC/02 Eximstats: use FileHandles for outputing results.
1537 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1539 SC/04 Eximstats: fixed display of large numbers with -nvr option
1541 SC/05 Eximstats: fixed merging of reports with empty tables.
1543 SC/06 Eximstats: added the -include_original_destination flag
1545 SC/07 Eximstats: removed tabs and trailing whitespace.
1547 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1549 TK/06 MBOX spool code: Add real "From " MBOX separator line
1550 so the .eml file is really in mbox format (even though
1551 most programs do not really care). Patch from Alex Miller.
1553 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1554 The latter is generated from $received_to and is only set if the
1555 message has one envelope recipient. SA can use these headers,
1556 obviously out-of-the-box. Patch from Alex Miller.
1558 PH/08 The ${def test on a variable was returning false if the variable's
1559 value was "0", contrary to what the specification has always said!
1560 The result should be true unless the variable is empty.
1562 PH/09 The syntax error of a character other than { following "${if
1563 def:variable_name" (after optional whitespace) was not being diagnosed.
1564 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1565 accidental colon was present, for example, could give incorrect results.
1567 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1568 result is used (not including appendfile, where other changes are about
1571 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1572 This involved changing a lot of size variables from int to off_t. It
1573 should work with maildirs and everything.
1575 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1576 spamd dying while we are connected to it.
1578 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1581 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1582 function for BDB 4.3.
1584 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1586 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1587 the sources for PCRE have been re-arranged and more files are now
1590 PH/15 The code I had for printing potentially long long variables in PH/11
1591 above was not the best (it lost precision). The length of off_t variables
1592 is now inspected at build time, and an appropriate printing format (%ld
1593 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1594 to be "long long int" or "long int". This is needed for the internal
1595 formatting function string_vformat().
1597 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1598 the configuration file to be ":syslog", then the script "guesses" where
1599 the logs files are, rather than using the compiled in default. In our
1600 case the guess is not the same as the compiled default, so the script
1601 suddenly stopped working when I started to use syslog. The patch checks
1602 to see if log_file_path is "". If so, it attempts to read it from exim
1603 with no configuration file to get the compiled in version, before it
1604 falls back to the previous guessing code."
1606 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1607 implementing BATV in an Exim configuration. See NewStuff for the gory
1610 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1611 Makefile that are specific to HP-UX.
1613 PH/18 If the "use_postmaster" option was set for a recipient callout together
1614 with the "random" option, the postmaster address was used as the MAIL
1615 FROM address for the random test, but not for the subsequent recipient
1616 test. It is now used for both.
1618 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1619 patch removes a few documentation additions to RFC 3028, because the
1620 latest draft now contains them. It adds the new en;ascii-case comparator
1621 and a new error check for 8bit text in MIME parts. Comparator and
1622 require names are now matched exactly. I enabled the subaddress
1623 extension, but it is not well tested yet (read: it works for me)."
1625 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1626 rework some of the code of TK/09 above to avoid the hardwired use of
1627 "%lld" and "long long". Replaced the call to snprintf() with a call to
1630 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1631 records point to non-existent hosts", "retry timeout exceeded", and
1632 "retry time not reached for any host after a long failure period".
1634 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1635 experimental DomainKeys support:
1637 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1638 (2) On an error such as an illegally used "control", the wrong name for
1639 the control was given.
1641 These problems did NOT occur unless DomainKeys support was compiled.
1643 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1645 PH/24 Added ${if match_ip condition.
1647 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1648 databases so that it will be absolutely obvious if a crash occurs in the
1649 DB library. This is a regular occurrence (often caused by mis-matched
1652 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1653 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1654 picked up on a user's system that detects such things. There doesn't seem
1655 to be a gcc warning option for this - only an attribute that has to be
1656 put on the function's prototype. It seems that in Fedora Core 4 they have
1657 set this on a number of new functions. No doubt there will be more in due
1660 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1661 specify a transport. However, if an address that was verified by such a
1662 router was the subject of a callout, Exim crashed because it tried to
1663 read the rcpt_include_affixes from the non-existent transport. Now it
1664 just assumes that the setting of that option is false. This bug was
1665 introduced by 4.51/PH/31.
1667 PH/28 Changed -d+all to exclude +memory, because that information is very
1668 rarely of interest, but it makes the output a lot bigger. People tend to
1669 do -d+all out of habit.
1671 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1672 code in os-type was giving problems when libc.so lives in lib64, like on
1675 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1676 aren't the modern standard, and it seems that some systems' include files
1677 don't always have them. Exim was already checking for some of the newer
1678 ones like T_AAAA, and defining it itself. I've added checks for all the
1679 record types that Exim uses.
1681 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1682 not automatically generating a new one, as it is supposed to. This
1683 prevented TLS from working. If the file did exist, but contained invalid
1684 data, a new version was generated, as expected. It was only the case of a
1685 non-existent file that was broken.
1687 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1688 with a change in libdomainkeys > 0.64.
1690 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1691 from DNS. If the selector record carries the flag, it now has
1692 precedence over the domain-wide flag.
1694 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1696 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1697 the use of a port name that isn't defined in /etc/services) occurred, the
1698 message was deferred as in a normal delivery, and thus remained on the
1699 spool, instead of being failed because of the mua_wrapper setting. This
1700 is now fixed, and I tidied up some of the mua_wrapper messages at the
1703 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1704 the messages in a hash of arrays rather than using individual hashes.
1705 This is a bit cleaner and results in dramatic memory savings, albeit
1706 at a slight CPU cost.
1708 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1709 as requested by Marc Sherman.
1711 SC/10 Eximstats: added histograms for user specified patterns as requested
1714 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1716 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1717 fopen() in the content-scanning modules that did not already have it.
1723 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1724 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1726 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1728 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1730 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1731 to test Sieve filters that use "vacation".
1733 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1734 that changes the way the GnuTLS parameters are stored in the cache file.
1735 The new format can be generated externally. For backward compatibility,
1736 if the data in the cache doesn't make sense, Exim assumes it has read an
1737 old-format file, and it generates new data and writes a new file. This
1738 means that you can't go back to an older release without removing the
1741 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1742 work if there are any delivery delays because "one_time" forces the
1743 parent to be marked "delivered", so its unseen clone is never tried
1744 again. For this reason, Exim now forbids the simultaneous setting of
1747 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1748 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1749 entry if you want to know the details. The fix, however, neglected to
1750 consider the case when local delivery batching is involved. The test for
1751 "previously delivered" was not happening when checking to see if an
1752 address could be batched with a previous (undelivered) one; under
1753 certain circumstances this could lead to multiple deliveries to the same
1756 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1757 in its include files, and this causes problems building Exim.
1759 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1760 header_syntax) but Exim was just ignoring anything given after a slash.
1761 In particular, this caused confusion with an attempt to use "verify =
1762 reverse_host_lookup/defer_ok". An error is now given when options are
1763 supplied for verify items that do not have them. (Maybe reverse_host_
1764 lookup should have a defer_ok option, but that's a different point.)
1766 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1767 defined by RFC 821) to 2048, because there were problems with some AUTH
1768 commands, and RFC 1869 says the size should be increased for extended
1769 SMTP commands that take arguments.
1771 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1774 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1775 "unknown" error; now it says that the functionality isn't in the binary.
1777 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1778 an address' error message when a string expansion fails (syntax or
1779 whatever). Otherwise the password may appear in the log. Following change
1780 PH/42 below, there is no longer a chance of it appearing in a bounce
1783 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1785 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1786 of its addresses was ever tried. (Bugzilla bug #2.)
1788 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1789 the result incorrectly in the debug output. (It correctly added a newline
1790 to what was transported.)
1792 TF/01 Added $received_time.
1794 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1795 commented out examples of how to interface to a virus scanner and to
1796 SpamAssassin. Also added commented examples of av_scanner and
1797 spamd_address settings.
1799 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1800 and controls are allowed in which ACLs. There were a couple of minor
1801 errors. Some of the entries in the conditions table (which is a table of
1802 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1803 negation of where the condition IS allowed.
1805 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1807 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1808 header file does not have a version number, so I've had to invent a new
1809 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1810 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1811 radiusclient), but it was contributed by a Radius user.
1813 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1814 files or directories, for OpenSSL.
1816 PH/19 When an Exim process that is running as root has to create an Exim log
1817 file, it does so in a subprocess that runs as exim:exim so as to get the
1818 ownership right at creation (otherwise, other Exim processes might see
1819 the file with the wrong ownership). There was no test for failure of this
1820 fork() call, which would lead to the process getting stuck as it waited
1821 for a non-existent subprocess. Forks do occasionally fail when resources
1822 run out. I reviewed all the other calls to fork(); they all seem to check
1825 PH/20 When checking for unexpected SMTP input at connect time (before writing
1826 the banner), Exim was not dealing correctly with a non-positive return
1827 from the read() function. If the client had disconnected by this time,
1828 the result was a log entry for a synchronization error with an empty
1829 string after "input=" when read() returned zero. If read() returned -1
1830 (an event I could not check), uninitialized data bytes were printed.
1831 There were reports of junk text (parts of files, etc) appearing after
1834 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1836 PH/22 Added support for macro redefinition, and (re)definition in between
1837 driver and ACL definitions.
1839 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1840 forgetting to use the resulting value; it was using the unexpanded value.
1842 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1843 hadn't been configured. The fix is from Juergen Kreileder, who
1844 understands it better than I do:
1846 "Here's what I see happening with three configured cyrus_sasl
1847 authenticators configured (plain, login, cram-md5):
1849 On startup auth_cyrus_sasl_init() gets called for each of these.
1850 This means three calls to sasl_listmech() without a specified mech_list.
1851 => SASL tests which mechs of all available mechs actually work
1852 => three warnings about OTP not working
1853 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1855 With the patch, sasl_listmech() also gets called three times. But now
1856 SASL's mech_list option is set to the server_mech specified in the the
1857 authenticator. Or in other words, the answer from sasl_listmech()
1858 gets limited to just the mech you're testing for (which is different
1860 => the return list contains just 'plain' or 'login', 'cram-md5' or
1861 nothing depending on the value of ob->server_mech.
1863 I've just tested the patch: Authentication still works fine,
1864 unavailable mechs specified in the exim configuration are still
1865 caught, and the auth.log warnings about OTP are gone."
1867 PH/25 When debugging is enabled, the contents of the command line are added
1868 to the debugging output, even when log_selector=+arguments is not
1871 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1872 answer is "GNU", and only if the return is "GNU/something" is the answer
1875 PH/27 $acl_verify_message is now set immediately after the failure of a
1876 verification in an ACL, and so is available in subsequent modifiers. In
1877 particular, the message can be preserved by coding like this:
1879 warn !verify = sender
1880 set acl_m0 = $acl_verify_message
1882 Previously, $acl_verify_message was set only while expanding "message"
1883 and "log_message" when a very denied access.
1885 PH/28 Modified OS/os.c-Linux with
1887 -#ifndef OS_LOAD_AVERAGE
1888 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1890 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1891 nomenclature these days.)
1893 PH/29 Installed patch from the Sieve maintainer that adds the options
1894 sieve_useraddress and sieve_subaddress to the redirect router.
1896 PH/30 In these circumstances:
1897 . Two addresses routed to the same list of hosts;
1898 . First host does not offer TLS;
1899 . First host accepts first address;
1900 . First host gives temporary error to second address;
1901 . Second host offers TLS and a TLS session is established;
1902 . Second host accepts second address.
1903 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1904 and peerdn, if requested) that were in fact used only for the second
1907 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1908 attention to any local part prefix or suffix that was matched by the
1909 router that accepted the address. It now behaves in the same way as it
1910 does for delivery: the affixes are removed from the local part unless
1911 rcpt_include_affixes is set on the transport.
1913 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1914 timeout during the DATA phase of an incoming message.
1916 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1917 applied a patch sanctioned by the Sieve maintainer.
1919 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1920 the uid or gid is negative. A case of a negative gid caused this to be
1921 noticed. The fix allows for either to be negative.
1923 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1924 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1927 PH/36 The change PH/12 above was broken. Fixed it.
1929 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1930 the grounds that routing the same address twice would always produce the
1931 same answer. This might have been true once, but it is certainly no
1932 longer true now. Routing a child address may depend on the previous
1933 routing that produced that child. Some complicated redirection strategies
1934 went wrong when messages had multiple recipients, and made Exim's
1935 behaviour dependent on the order in which the addresses were given.
1937 I have moved the duplicate checking until after the routing is complete.
1938 Exim scans the addresses that are assigned to local and remote
1939 transports, and removes any duplicates. This means that more work will be
1940 done, as duplicates will always all be routed, but duplicates are
1941 presumably rare, so I don't expect this is of any significance.
1943 For deliveries to pipes, files, and autoreplies, the duplicate checking
1944 still happens during the routing process, since they are not going to be
1947 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1948 It corrects a timeout issue with spamd. This is Ian's comment: "The
1949 background is that sometimes spamd either never reads data from a
1950 connection it has accepted, or it never writes response data. The exiscan
1951 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1952 blindly assumes that writes won't block so it may never time out."
1954 PH/39 Allow G after quota size as well as K and M.
1956 PH/40 The value set for $authenticated_id in an authenticator may not contain
1957 binary zeroes or newlines because the value is written to log lines and
1958 to spool files. There was no check on this. Now the value is run through
1959 the string_printing() function so that such characters are converted to
1960 printable escape sequences.
1962 PH/41 $message_linecount is a new variable that contains the total number of
1963 lines in the message. Compare $body_linecount, which is the count for the
1966 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1967 bounce and delay warning messages, except in certain special cases, which
1970 (a) An SMTP error message from a remote host;
1971 (b) A message specified in a :fail: redirection;
1972 (c) A message specified in a "fail" command in a system filter;
1973 (d) A message specified in a FAIL return from the queryprogram router;
1974 (e) A message specified by the cannot_route_message router option.
1976 In these cases only, Exim does include the error details in bounce and
1977 warning messages. There are also a few cases where bland messages such
1978 as "unrouteable address" or "local delivery error" are given.
1980 PH/43 $value is now also set for the "else" part of a ${run expansion.
1982 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1983 being worked on, but at least Exim now implements the latest version to
1986 PH/45 In a pipe transport, although a timeout while waiting for the pipe
1987 process to complete was treated as a delivery failure, a timeout while
1988 writing the message to the pipe was logged, but erroneously treated as a
1989 successful delivery. Such timeouts include transport filter timeouts. For
1990 consistency with the overall process timeout, these timeouts are now
1991 treated as errors, giving rise to delivery failures by default. However,
1992 there is now a new Boolean option for the pipe transport called
1993 timeout_defer, which, if set TRUE, converts the failures into defers for
1994 both kinds of timeout. A transport filter timeout is now identified in
1997 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1998 systems where "make" and "gmake" are different, calling "gmake" at top
1999 level broke things. I've arranged for the value of $(MAKE) to be passed
2000 from the Makefile to this script so that it can call the same version of
2004 A note about Exim versions 4.44 and 4.50
2005 ----------------------------------------
2007 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2008 changes of various kinds. As a consequence, a big documentation update was
2009 needed. This delayed the release for rather longer than seemed good, especially
2010 in the light of a couple of (minor) security issues. Therefore, the changes
2011 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2012 release. So 4.44 and 4.50 are in effect two different branches that both start
2015 I have left the 4.50 change log unchanged; it contains all the changes since
2016 4.43. The change log for 4.44 is below; many of its items are identical to
2017 those for 4.50. This seems to be the most sensible way to preserve the
2018 historical information.
2024 1. Minor wording change to the doc/README.SIEVE file.
2026 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2027 computation of the current number of files was incorrect.
2029 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2030 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2033 4. Give more explanation in the error message when the command for a transport
2034 filter fails to execute.
2036 5. There are several places where Exim runs a non-Exim command in a
2037 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2038 was being done only for the command run by the queryprogram router. It is
2039 now done for all such subprocesses. The other cases are: ${run, transport
2040 filters, and the commands run by the lmtp and pipe transports.
2042 6. Added CONFIGURE_GROUP build-time option.
2044 7. Some older OS have a limit of 256 on the maximum number of file
2045 descriptors. Exim was using setrlimit() to set 1000 as a large value
2046 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2047 systems. I've change it so that if it can't get 1000, it tries for 256.
2049 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2050 was an oversight, and furthermore, ever since the addition of extra
2051 controls (e.g. 4.43/32), the checks on when to allow different forms of
2052 "control" were broken. There should now be diagnostics for all cases when a
2053 control that does not make sense is encountered.
2055 9. Added the /retain_sender option to "control=submission".
2057 10. $recipients is now available in the predata ACL (oversight).
2059 11. Tidy the search cache before the fork to do a delivery from a message
2060 received from the command line. Otherwise the child will trigger a lookup
2061 failure and thereby defer the delivery if it tries to use (for example) a
2062 cached ldap connection that the parent has called unbind on.
2064 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2065 of $address_data from the recipient verification was clobbered by the
2066 sender verification.
2068 13. The value of address_data from a sender verification is now available in
2069 $sender_address_data in subsequent conditions in the ACL statement.
2071 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2073 15. Added a new option "connect=<time>" to callout options, to set a different
2076 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2077 was its contents. (It was OK if the option was not defined at all.)
2079 17. A "Completed" log line is now written for messages that are removed from
2080 the spool by the -Mrm option.
2082 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2083 information about exactly what failed.
2085 19. Added -dd to debug only the daemon process.
2087 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2088 handles timeouts, both on the server side and network timeouts. Renamed the
2089 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2091 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2092 It is now set to "smtps".
2094 22. $host_address is now set to the target address during the checking of
2095 ignore_target_hosts.
2097 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2098 being passed; this would have caused $sender_host_name to have been used if
2099 matching the list had actually called for a host name (not very likely,
2100 since this list is usually IP addresses). A host name is now passed as
2103 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2104 code that specifies a non-restarting handler (typically sigaction() in
2105 modern systems) in an attempt to fix a rare and obscure crash bug.
2107 25. Narrowed the window for a race in the daemon that could cause it to ignore
2108 SIGCHLD signals. This is not a major problem, because they are used only to
2109 wake it up if nothing else does.
2111 26. A malformed maildirsize file could cause Exim to calculate negative values
2112 for the mailbox size or file count. Odd effects could occur as a result.
2113 The maildirsize information is now recalculated if the size or filecount
2116 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2117 support for a long time. Removed HAVE_SYS_VFS_H.
2119 28. Installed the latest version of exipick from John Jetmore.
2121 29. In an address list, if the pattern was not a regular expression, an empty
2122 subject address (from a bounce message) matched only if the pattern was an
2123 empty string. Non-empty patterns were not even tested. This was the wrong
2124 because it is perfectly reasonable to use an empty address as part of a
2125 database query. An empty address is now tested by patterns that are
2126 lookups. However, all the other forms of pattern expect the subject to
2127 contain a local part and a domain, and therefore, for them, an empty
2128 address still always fails if the pattern is not itself empty.
2130 30. Exim went into a mad DNS loop when attempting to do a callout where the
2131 host was specified on an smtp transport, and looking it up yielded more
2132 than one IP address.
2134 31. Re-factored the code for checking spool and log partition space into a
2135 function that finds that data and another that does the check. The former
2136 is then used to implement four new variables: $spool_space, $log_space,
2137 $spool_inodes, and $log_inodes.
2139 32. The RFC2047 encoding function was originally intended for short strings
2140 such as real names; it was not keeping to the 75-character limit for
2141 encoded words that the RFC imposes. It now respects the limit, and
2142 generates multiple encoded words if necessary. To be on the safe side, I
2143 have increased the buffer size for the ${rfc2047: expansion operator from
2146 33. It is now permitted to omit both strings after an "if" condition; if the
2147 condition is true, the result is "true". As before, when the second string
2148 is omitted, a false condition yields an empty string. This makes it less
2149 cumbersome to write custom ACL and router conditions.
2151 34. Failure to deliver a bounce message always caused it to be frozen, even if
2152 there was an errors_to setting on the router. The errors_to setting is now
2155 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2156 canonical form (fully expanded) before being placed in
2157 $sender_host_address.
2159 36. The table in the code that translates DNS record types into text (T_A to
2160 "A" for instance) was missing entries for NS and CNAME. It is just possible
2161 that this could have caused confusion if both these types were looked up
2162 for the same domain, because the text type is used as part of Exim's
2163 per-process caching. But the chance of anyone hitting this buglet seems
2166 37. The dnsdb lookup has been extended in a number of ways.
2168 (1) There is a new type, "zns", which walks up the domain tree until it
2169 finds some nameserver records. It should be used with care.
2171 (2) There is a new type, "mxh", which is like "mx" except that it returns
2172 just the host names, not the priorities.
2174 (3) It is now possible to give a list of domains (or IP addresses) to be
2175 looked up. The behaviour when one of the lookups defers can be
2176 controlled by a keyword.
2178 (4) It is now possible to specify the separator character for use when
2179 multiple records are returned.
2181 38. The dnslists ACL condition has been extended: it is now possible to supply
2182 a list of IP addresses and/or domains to be looked up in a particular DNS
2185 39. Added log_selector=+queue_time_overall.
2187 40. When running the queue in the test harness, wait just a tad after forking a
2188 delivery process, to get repeatability of debugging output.
2190 41. Include certificate and key file names in error message when GnuTLS fails
2191 to set them up, because the GnuTLS error message doesn't include the name
2192 of the failing file when there is a problem reading it.
2194 42. Allow both -bf and -bF in the same test run.
2196 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2198 44. The "Exiscan patch" is now merged into the mainline Exim source.
2200 45. Sometimes the final signoff response after QUIT could fail to get
2201 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2202 < 0 before doing a fflush(). This bug looks as though it goes back to the
2203 introduction of TLS in release 3.20, but "sometimes" must have been rare
2204 because the tests only now provoked it.
2206 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2207 (this can affect the format of dates).
2209 47. exim_tidydb, when checking for the continued existence of a message for
2210 which it has found a message-specific retry record, was not finding
2211 messages that were in split spool directories. Consequently, it was
2212 deleting retry records that should have stayed in existence.
2214 48. Steve fixed some bugs in eximstats.
2216 49. The SPA authentication driver was not abandoning authentication and moving
2217 on to the next authenticator when an expansion was forced to fail,
2218 contradicting the general specification for all authenticators. Instead it
2219 was generating a temporary error. It now behaves as specified.
2221 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2222 (the order specifies the preference for clients). The order is now AES256,
2223 AES128, 3DES, ARCFOUR128.
2225 51. Small patch to Sieve code - explicitly set From: when generating an
2228 52. Exim crashed if a remote delivery caused a very long error message to be
2229 recorded - for instance if somebody sent an entire SpamAssassin report back
2230 as a large number of 550 error lines. This bug was coincidentally fixed by
2231 increasing the size of one of Exim's internal buffers (big_buffer) that
2232 happened as part of the Exiscan merge. However, to be on the safe side, I
2233 have made the code more robust (and fixed the comments that describe what
2236 53. Now that there can be additional text after "Completed" in log lines (if
2237 the queue_time_overall log selector is set), a one-byte patch to exigrep
2238 was needed to allow it to recognize "Completed" as not the last thing in
2241 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2242 patch that reportedly fixes this has been added. I am not expert enough to
2243 create a test for it. This is what the patch creator wrote:
2245 "I found a little strange behaviour of ldap code when working with
2246 Windows 2003 AD Domain, where users was placed in more than one
2247 Organization Units. When I tried to give exim partial DN, the exit code
2248 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2249 But simultaneously result of request was absolutely normal ldap result,
2250 so I produce this patch..."
2252 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2253 REFERENCE, so I have modified the code to exclude the patch when that macro
2256 55. Some experimental protocols are using DNS PTR records for new purposes. The
2257 keys for these records are domain names, not reversed IP addresses. The
2258 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2259 leaves it alone. Component reversal etc. now happens only for IP addresses.
2262 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2264 57. Double the size of the debug message buffer (to 2048) so that more of very
2265 long debug lines gets shown.
2267 58. The exicyclog utility now does better if the number of log files to keep
2268 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2270 59. Two changes related to the smtp_active_hostname option:
2272 (1) $smtp_active_hostname is now available as a variable.
2273 (2) The default for smtp_banner uses $smtp_active_hostname instead
2274 of $primary_hostname.
2276 60. The host_aton() function is supposed to be passed a string that is known
2277 to be a valid IP address. However, in the case of IPv6 addresses, it was
2278 not checking this. This is a hostage to fortune. Exim now panics and dies
2279 if the condition is not met. A case was found where this could be provoked
2280 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2281 components; fortuitously, this particular loophole had already been fixed
2282 by change 4.50/55 above.
2284 If there are any other similar loopholes, the new check in host_aton()
2285 itself should stop them being exploited. The report I received stated that
2286 data on the command line could provoke the exploit when Exim was running as
2287 exim, but did not say which command line option was involved. All I could
2288 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2289 running as the user.
2292 61. There was a buffer overflow vulnerability in the SPA authentication code
2293 (which came originally from the Samba project). I have added a test to the
2294 spa_base64_to_bits() function which I hope fixes it.
2297 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2298 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2300 63. The daemon start-up calls getloadavg() while still root for those OS that
2301 need the first call to be done as root, but it missed one case: when
2302 deliver_queue_load_max is set with deliver_drop_privilege. This is
2303 necessary for the benefit of the queue runner, because there is no re-exec
2304 when deliver_drop_privilege is set.
2306 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2307 This has been fixed.
2309 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2310 was in use, was not putting the data itself into the right store pool;
2311 consequently, it could be overwritten for a subsequent message in the same
2312 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2315 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2317 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2318 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2319 the function were treating the return as a boolean value, which happened to
2320 work because 0=false and not-0=true, but is not correct code.
2322 68. The host_aton() function was not handling scoped IPv6 addresses (those
2323 with, for example, "%eth0" on the end) correctly.
2325 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2326 negated items (that is, ~something) in unsigned ints. Some compilers
2327 apparently mutter when there is no cast.
2329 70. If an address verification called from an ACL failed, and did not produce a
2330 user-specific message (i.e. there was only a "system" message), nothing was
2331 put in $acl_verify_message. In this situation, it now puts the system
2334 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2335 session; change 4.31/43 added the unwanted input to the log line - except
2336 that it did not do this in the start of session case. It now does.
2338 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2339 This is wrong and can cause the other end to generate a synchronization
2340 error if it is another Exim or anything else that does the synchronization
2341 check. A QUIT command is no longer sent after a timeout.
2343 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2344 during host lookups.
2346 74. The defer_ok option of callout verification was not working if it was used
2347 when verifying addresses in header lines, that is, for this case:
2349 verify = header_sender/callout=defer_ok
2351 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2352 those file descriptors could be used for SMTP connections. If anything
2353 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2354 could be sent to the SMTP client, causing chaos. The daemon now opens
2355 stdin, stdout, and stderr to /dev/null when it puts itself into the
2358 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2359 log by default. The user can override this with suitable Perl magic.
2361 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2362 the log message when discard triggers, was not working for the DATA ACL or
2363 for the non-SMTP ACL.
2365 78. Error message wording change in sieve.c.
2367 79. If smtp_accept_max_per_host was set, the number of connections could be
2368 restricted to fewer than expected, because the daemon was trying to set up
2369 a new connection before checking whether the processes handling previous
2370 connections had finished. The check for completed processes is now done
2371 earlier. On busy systems, this bug wouldn't be noticed because something
2372 else would have woken the daemon, and it would have reaped the completed
2375 80. If a message was submitted locally by a user whose login name contained one
2376 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2377 It caused a spool format error. I have fixed the spool reading code. A
2378 related problem was that the "from" clause in the Received: line became
2379 illegal because of the space(s). It is now covered by ${quote_local_part.
2381 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2384 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2386 83. Patch from Sieve maintainer to fix unterminated string problem in
2387 "vacation" handling.
2389 84. Some minor changes to the Linux configuration files to help with other
2390 OS variants using glibc.
2392 85. One more patch for Sieve to update vacation handling to latest spec.
2395 ----------------------------------------------------
2396 See the note above about the 4.44 and 4.50 releases.
2397 ----------------------------------------------------
2403 1. Change 4.43/35 introduced a bug that caused file counts to be
2404 incorrectly computed when quota_filecount was set in an appendfile
2407 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2408 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2411 3. Give more explanation in the error message when the command for a transport
2412 filter fails to execute.
2414 4. There are several places where Exim runs a non-Exim command in a
2415 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2416 was being done only for the command run by the queryprogram router. It is
2417 now done for all such subprocesses. The other cases are: ${run, transport
2418 filters, and the commands run by the lmtp and pipe transports.
2420 5. Some older OS have a limit of 256 on the maximum number of file
2421 descriptors. Exim was using setrlimit() to set 1000 as a large value
2422 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2423 systems. I've change it so that if it can't get 1000, it tries for 256.
2425 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2426 was an oversight, and furthermore, ever since the addition of extra
2427 controls (e.g. 4.43/32), the checks on when to allow different forms of
2428 "control" were broken. There should now be diagnostics for all cases when a
2429 control that does not make sense is encountered.
2431 7. $recipients is now available in the predata ACL (oversight).
2433 8. Tidy the search cache before the fork to do a delivery from a message
2434 received from the command line. Otherwise the child will trigger a lookup
2435 failure and thereby defer the delivery if it tries to use (for example) a
2436 cached ldap connection that the parent has called unbind on.
2438 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2439 of $address_data from the recipient verification was clobbered by the
2440 sender verification.
2442 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2443 was its contents. (It was OK if the option was not defined at all.)
2445 11. A "Completed" log line is now written for messages that are removed from
2446 the spool by the -Mrm option.
2448 12. $host_address is now set to the target address during the checking of
2449 ignore_target_hosts.
2451 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2452 being passed; this would have caused $sender_host_name to have been used if
2453 matching the list had actually called for a host name (not very likely,
2454 since this list is usually IP addresses). A host name is now passed as
2457 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2458 code that specifies a non-restarting handler (typically sigaction() in
2459 modern systems) in an attempt to fix a rare and obscure crash bug.
2461 15. Narrowed the window for a race in the daemon that could cause it to ignore
2462 SIGCHLD signals. This is not a major problem, because they are used only to
2463 wake it up if nothing else does.
2465 16. A malformed maildirsize file could cause Exim to calculate negative values
2466 for the mailbox size or file count. Odd effects could occur as a result.
2467 The maildirsize information is now recalculated if the size or filecount
2470 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2471 support for a long time. Removed HAVE_SYS_VFS_H.
2473 18. Updated exipick to current release from John Jetmore.
2475 19. Allow an empty sender to be matched against a lookup in an address list.
2476 Previously the only cases considered were a regular expression, or an
2479 20. Exim went into a mad DNS lookup loop when doing a callout where the
2480 host was specified on the transport, if the DNS lookup yielded more than
2483 21. The RFC2047 encoding function was originally intended for short strings
2484 such as real names; it was not keeping to the 75-character limit for
2485 encoded words that the RFC imposes. It now respects the limit, and
2486 generates multiple encoded words if necessary. To be on the safe side, I
2487 have increased the buffer size for the ${rfc2047: expansion operator from
2490 22. Failure to deliver a bounce message always caused it to be frozen, even if
2491 there was an errors_to setting on the router. The errors_to setting is now
2494 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2495 canonical form (fully expanded) before being placed in
2496 $sender_host_address.
2498 24. Updated eximstats to version 1.33
2500 25. Include certificate and key file names in error message when GnuTLS fails
2501 to set them up, because the GnuTLS error message doesn't include the name
2502 of the failing file when there is a problem reading it.
2504 26. Expand error message when OpenSSL has problems setting up cert/key files.
2507 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2508 (this can affect the format of dates).
2510 28. exim_tidydb, when checking for the continued existence of a message for
2511 which it has found a message-specific retry record, was not finding
2512 messages that were in split spool directories. Consequently, it was
2513 deleting retry records that should have stayed in existence.
2515 29. eximstats updated to version 1.35
2516 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2517 1.35 - bugfix such that pie charts by volume are generated correctly
2519 30. The SPA authentication driver was not abandoning authentication and moving
2520 on to the next authenticator when an expansion was forced to fail,
2521 contradicting the general specification for all authenticators. Instead it
2522 was generating a temporary error. It now behaves as specified.
2524 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2525 (the order specifies the preference for clients). The order is now AES256,
2526 AES128, 3DES, ARCFOUR128.
2528 31. Small patch to Sieve code - explicitly set From: when generating an
2531 32. Exim crashed if a remote delivery caused a very long error message to be
2532 recorded - for instance if somebody sent an entire SpamAssassin report back
2533 as a large number of 550 error lines. This bug was coincidentally fixed by
2534 increasing the size of one of Exim's internal buffers (big_buffer) that
2535 happened as part of the Exiscan merge. However, to be on the safe side, I
2536 have made the code more robust (and fixed the comments that describe what
2539 33. Some experimental protocols are using DNS PTR records for new purposes. The
2540 keys for these records are domain names, not reversed IP addresses. The
2541 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2542 leaves it alone. Component reversal etc. now happens only for IP addresses.
2545 34. The host_aton() function is supposed to be passed a string that is known
2546 to be a valid IP address. However, in the case of IPv6 addresses, it was
2547 not checking this. This is a hostage to fortune. Exim now panics and dies
2548 if the condition is not met. A case was found where this could be provoked
2549 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2550 components; fortuitously, this particular loophole had already been fixed
2551 by change 4.50/55 or 4.44/33 above.
2553 If there are any other similar loopholes, the new check in host_aton()
2554 itself should stop them being exploited. The report I received stated that
2555 data on the command line could provoke the exploit when Exim was running as
2556 exim, but did not say which command line option was involved. All I could
2557 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2558 running as the user.
2561 35. There was a buffer overflow vulnerability in the SPA authentication code
2562 (which came originally from the Samba project). I have added a test to the
2563 spa_base64_to_bits() function which I hope fixes it.
2566 36. The daemon start-up calls getloadavg() while still root for those OS that
2567 need the first call to be done as root, but it missed one case: when
2568 deliver_queue_load_max is set with deliver_drop_privilege. This is
2569 necessary for the benefit of the queue runner, because there is no re-exec
2570 when deliver_drop_privilege is set.
2572 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2573 was in use, was not putting the data itself into the right store pool;
2574 consequently, it could be overwritten for a subsequent message in the same
2575 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2578 38. Sometimes the final signoff response after QUIT could fail to get
2579 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2580 < 0 before doing a fflush(). This bug looks as though it goes back to the
2581 introduction of TLS in release 3.20, but "sometimes" must have been rare
2582 because the tests only now provoked it.
2588 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2589 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2590 Now it returns an integer. A number of calls were still expecting a T/F
2591 return. Fortuitously, in all cases, the tests worked in OK situations,
2592 which is the norm. However, things would have gone wrong on any write
2593 failures on the smtp file descriptor. This function is used when sending
2594 messages over SMTP and also when doing verify callouts.
2596 2. When Exim is called to do synchronous delivery of a locally submitted
2597 message (the -odf or -odi options), it no longer closes stderr before doing
2600 3. Implemented the mua_wrapper option.
2602 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2604 5. Implemented the functions header_remove(), header_testname(),
2605 header_add_at_position(), and receive_remove_recipient(), and exported them
2608 6. If an ACL "warn" statement specified the addition of headers, Exim already
2609 inserted X-ACL-Warn: at the start if there was no header name. However, it
2610 was not making this test for the second and subsequent header lines if
2611 there were newlines in the string. This meant that an invalid header could
2612 be inserted if Exim was badly configured.
2614 7. Allow an ACL "warn" statement to add header lines at the start or after all
2615 the Received: headers, as well as at the end.
2617 8. Added the rcpt_4xx retry error code.
2619 9. Added postmaster_mailfrom=xxx to callout verification option.
2621 10. Added mailfrom=xxxx to the callout verification option, for verify=
2624 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2625 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2627 12. Inserted some casts to stop certain compilers warning when using pointer
2628 differences as field lengths or precisions in printf-type calls (mostly
2629 affecting debugging statements).
2631 13. Added optional readline() support for -be (dynamically loaded).
2633 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2634 same clock tick as a message's arrival, so that its received time was the
2635 same as the "first fail" time on the retry record, and that message
2636 remained on the queue past the ultimate address timeout, every queue runner
2637 would try a delivery (because it was past the ultimate address timeout) but
2638 after another failure, the ultimate address timeout, which should have then
2639 bounced the address, did not kick in. This was a "< instead of <=" error;
2640 in most cases the first failure would have been in the next clock tick
2641 after the received time, and all would be well.
2643 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2644 being recognized when the domain list was tested by the match_domain
2645 condition in an expansion string.
2647 16. Added the ${str2b64: operator.
2649 17. Exim was always calling setrlimit() to set a large limit for the number of
2650 processes, without checking whether the existing limit was already
2651 adequate. (It did check for the limit on file descriptors.) Furthermore,
2652 errors from getrlimit() and setrlimit() were being ignored. Now they are
2653 logged to the main and panic logs, but Exim does carry on, to try to do its
2654 job under whatever limits there are.
2656 18. Imported PCRE 5.0.
2658 19. Trivial typo in log message " temporarily refused connection" (the leading
2661 20. If the log selector return_path_on_delivery was set and an address was
2662 redirected to /dev/null, the delivery process crashed because it assumed
2663 that a return path would always be set for a "successful" delivery. In this
2664 case, the whole delivery is bypassed as an optimization, and therefore no
2667 21. Internal re-arrangement: the function for sending a challenge and reading
2668 a response while authentication was assuming a zero-terminated challenge
2669 string. It's now changed to take a pointer and a length, to allow for
2670 binary data in such strings.
2672 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2674 23. Exim was not respecting finduser_retries when seeking the login of the
2675 uid under which it was called; it was always trying 10 times. (The default
2676 setting of finduser_retries is zero.) Also, it was sleeping after the final
2677 failure, which is pointless.
2679 24. Implemented tls_on_connect_ports.
2681 25. Implemented acl_smtp_predata.
2683 26. If the domain in control=submission is set empty, Exim assumes that the
2684 authenticated id is a complete email address when it generates From: or
2685 Sender: header lines.
2687 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2688 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2689 chown and chgrp in /bin and hostname in /usr/bin.
2691 28. Exim was keeping the "process log" file open after each use, just as it
2692 does for the main log. This opens the possibility of it remaining open for
2693 long periods when the USR1 signal hits a daemon. Occasional processlog
2694 errors were reported, that could have been caused by this. Anyway, it seems
2695 much more sensible not to leave this file open at all, so that is what now
2698 29. The long-running daemon process does not normally write to the log once it
2699 has entered its main loop, and it closes the log before doing so. This is
2700 so that log files can straightforwardly be renamed and moved. However,
2701 there are a couple of unusual error situations where the daemon does write
2702 log entries, and I had neglected to close the log afterwards.
2704 30. The text of an SMTP error response that was received during a remote
2705 delivery was being truncated at 512 bytes. This is too short for some of
2706 the long messages that one sometimes sees. I've increased the limit to
2709 31. It is now possible to make retry rules that apply only when a message has a
2710 specific sender, in particular, an empty sender.
2712 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2713 it possible to be selective about when SMTP synchronization is enforced.
2715 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2717 32. Implemented hosts_connection_nolog.
2719 33. Added an ACL for QUIT.
2721 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2724 35. Added mailbox_size and mailbox_filecount to appendfile.
2726 36. Added control = no_multiline_responses to ACLs.
2728 37. There was a bug in the logic of the code that waits for the clock to tick
2729 in the case where the clock went backwards by a substantial amount such
2730 that the microsecond fraction of "now" was more than the microsecond
2731 fraction of "then" (but the whole seconds number was less).
2733 38. Added support for the libradius Radius client library this is found on
2734 FreeBSD (previously only the radiusclient library was supported).
2740 1. When certain lookups returned multiple values in the form name=value, the
2741 quoting of the values was not always being done properly. Specifically:
2742 (a) If the value started with a double quote, but contained no whitespace,
2744 (b) If the value contained whitespace other than a space character (i.e.
2745 tabs or newlines or carriage returns) it was not quoted.
2746 This fix has been applied to the mysql and pgsql lookups by writing a
2747 separate quoting function and calling it from the lookup code. The fix
2748 should probably also be applied to nisplus, ibase and oracle lookups, but
2749 since I cannot test any of those, I have not disturbed their existing code.
2751 2. A hit in the callout cache for a specific address caused a log line with no
2752 reason for rejecting RCPT. Now it says "Previous (cached) callout
2753 verification failure".
2755 3. There was an off-by-one bug in the queryprogram router. An over-long
2756 return line was truncated at 256 instead of 255 characters, thereby
2757 overflowing its buffer with the terminating zero. As well as fixing this, I
2758 have increased the buffer size to 1024 (and made a note to document this).
2760 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2761 when Exim is waiting for an SMTP response from a remote server, Exim
2762 restarts its select() call on the socket, thereby resetting its timeout.
2763 This is not a problem when such interrupts are rare. Somebody set up a cron
2764 job to run exiwhat every 2 minutes, which is less than the normal select()
2765 timeout (5 or 10 minutes). This meant that the select() timeout never
2766 kicked in because it was always reset. I have fixed this by comparing the
2767 time when an interrupt arrives with the time at the start of the first call
2768 to select(). If more time than the timeout has elapsed, the interrupt is
2769 treated as a timeout.
2771 5. Some internal re-factoring in preparation for the addition of Sieve
2772 extensions (by MH). In particular, the "personal" test is moved to a
2773 separate function, and given an option for scanning Cc: and Bcc: (which is
2774 not set for Exim filters).
2776 6. When Exim created an email address using the login of the caller as the
2777 local part (e.g. when creating a From: or Sender: header line), it was not
2778 quoting the local part when it contained special characters such as @.
2780 7. Installed new OpenBSD configuration files.
2782 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2783 try to make them clearer.
2785 9. Callout options, other than the timeout value, were being ignored when
2786 verifying sender addresses in header lines. For example, when using
2788 verify = header_sender/callout=no_cache
2790 the cache was (incorrectly) being used.
2792 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2793 only the Cygwin environment.
2795 11. When return_path_on_delivery was set as a log selector, if different remote
2796 addresses in the same message used different return paths and parallel
2797 remote delivery occurred, the wrong values would sometimes be logged.
2798 (Whenever a remote delivery process finished, the return path value from
2799 the most recently started remote delivery process was logged.)
2801 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2802 lines when AUTH and/or TLS are in use. This is the "received protocol"
2803 field. Exim used to use "asmtp" for authenticated SMTP, without any
2804 indication (in the protocol name) for TLS use. Now it follows the RFC and
2805 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2806 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2807 names appear in log lines as well as in Received: header lines.
2809 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2810 extensions, and comparison tests, and to fix some bugs.
2812 14. Changes to the "personal" filter test:
2814 (1) The test was buggy in that it was just doing the equivalent of
2815 "contains" tests on header lines. For example, if a user's address was
2816 anne@some.where, the "personal" test would incorrectly be true for
2818 To: susanne@some.where
2820 This test is now done by extracting each address from the header in turn,
2821 and checking the entire address. Other tests that are part of "personal"
2822 are now done using regular expressions (for example, to check local parts
2823 of addresses in From: header lines).
2825 (2) The list of non-personal local parts in From: addresses has been
2826 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2827 taken from the Sieve specification recommendations.
2829 (3) If the message contains any header line starting with "List-" it is
2830 treated as non-personal.
2832 (4) The test for "circular" in the Subject: header line has been removed
2833 because it now seems ill-conceived.
2835 15. Minor typos in src/EDITME comments corrected.
2837 16. Installed latest exipick from John Jetmore.
2839 17. If headers_add on a router specified a text string that was too long for
2840 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2841 of string_sprintf() is now avoided.
2843 18. $message_body_size was not set (it was always zero) when running the DATA
2844 ACL and the local_scan() function.
2846 19. For the "mail" command in an Exim filter, no default was being set for
2847 the once_repeat time, causing a random time value to be used if "once" was
2848 specified. (If the value happened to be <= 0, no repeat happened.) The
2849 default is now 0s, meaning "never repeat". The "vacation" command was OK
2850 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2851 (I found it when inspecting the code).
2853 20. There is now an overall timeout for performing a callout verification. It
2854 defaults to 4 times the callout timeout, which applies to individual SMTP
2855 commands during the callout. The overall timeout applies when there is more
2856 than one host that can be tried. The timeout is checked before trying the
2857 next host. This prevents very long delays if there are a large number of
2858 hosts and all are timing out (e.g. when the network connections are timing
2859 out). The value of the overall timeout can be changed by specifying an
2860 additional sub-option for "callout", called "maxwait". For example:
2862 verify = sender/callout=5s,maxwait=20s
2864 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2865 to the end before writing, but this should make it even safer).
2867 22. Exim was forgetting that it had advertised PIPELINING for the second and
2868 subsequent messages on an SMTP connection. It was also not resetting its
2869 memory on STARTTLS and an internal HELO.
2871 23. When Exim logs an SMTP synchronization error within a session, it now
2872 records whether PIPELINING has been advertised or not.
2874 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2875 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2876 rather than long int.
2878 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2880 26. Added the never_mail option to autoreply.
2886 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2887 crash if the getsockname() call failed; this can happen if a connection is
2888 closed very soon after it is established. The problem was simply in the
2889 order in which certain operations were done, causing Exim to try to write
2890 to the SMTP stream before it had set up the file descriptor. The bug has
2891 been fixed by making things happen in the correct order.
2897 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2898 before the connection was closed, thus losing the rejection response.
2900 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2901 some early Solaris releases, but causes trouble in current releases where
2902 socklen_t is defined.
2904 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2907 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2910 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2911 that releasing the top of it at the end releases what was used for sub-
2912 expansions (unless the block got too big). However, discard this block if
2913 the first thing is a variable or header, so that we can use its block when
2914 it is dynamic (useful for very large $message_headers, for example).
2916 6. Lookups now cache *every* query, not just the most recent. A new, separate
2917 store pool is used for this. It can be recovered when all lookup caches are
2918 flushed. Lookups now release memory at the end of their result strings.
2919 This has involved some general refactoring of the lookup sources.
2921 7. Some code has been added to the store_xxx() functions to reduce the amount
2922 of flapping under certain conditions.
2924 8. log_incoming_interface used to affect only the <= reception log lines. Now
2925 it causes the local interface and port to be added to several more SMTP log
2926 lines, for example "SMTP connection from", and rejection lines.
2928 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2930 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2932 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2933 could be overwritten at the end of the current message (or the start of a
2934 new message if it was set in a HELO ACL). The value is now preserved for
2935 the duration of the SMTP connection.
2937 12. If a transport had a headers_rewrite setting, and a matching header line
2938 contained an unqualified address, that address was qualified, even if it
2939 did not match any rewriting rules. The underlying bug was that the values
2940 of the flags that permit the existence of unqualified sender and recipient
2941 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2942 non-local messages, and by -bnq for local messages) were not being
2943 preserved with the message after it was received.
2945 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2946 "next input=" as part of the text comprising the host identity instead of
2947 the correct text. The code was using the same buffer for two different
2948 strings. However, depending on which order the printing function evaluated
2949 its arguments, the bug did not always show up. Under Linux, for example, my
2950 test suite worked just fine.
2952 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2953 doesn't work with some older versions of Perl. It has been changed to "my",
2954 which in any case is probably the better facility to use.
2956 15. A really picky compiler found some instances of statements for creating
2957 error messages that either had too many or two few arguments for the format
2960 16. The size of the buffer for calls to the DNS resolver has been increased
2961 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2962 for addresses that have a lot of PTR records. This alleviates a problem; it
2963 does not fully solve it.
2965 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2966 buffer now truncates the list and logs the incident, which is the same
2967 action as happens when Exim is looking up a host name and its aliases.
2968 Previously in this situation something unpredictable would happen;
2969 sometimes it was "internal error: store_reset failed".
2971 18. If a server dropped the connection unexpectedly when an Exim client was
2972 using GnuTLS and trying to read a response, the client delivery process
2973 crashed while trying to generate an error log message.
2975 19. If a "warn" verb in an ACL added multiple headers to a message in a single
2976 string, for example:
2978 warn message = H1: something\nH2: something
2980 the text was added as a single header line from Exim's point of view
2981 though it ended up OK in the delivered message. However, searching for the
2982 second and subsequent header lines using $h_h2: did not work. This has been
2983 fixed. Similarly, if a system filter added multiple headers in this way,
2984 the routers could not see them.
2986 20. Expanded the error message when iplsearch is called with an invalid key to
2987 suggest using net-iplsearch in a host list.
2989 21. When running tests using -bh, any delays imposed by "delay" modifiers in
2990 ACLs are no longer actually imposed (and a message to that effect is
2993 22. If a "gecos" field in a passwd entry contained escaped characters, in
2994 particular, if it contained a \" sequence, Exim got it wrong when building
2995 a From: or a Sender: header from that name. A second bug also caused
2996 incorrect handling when an unquoted " was present following a character
2997 that needed quoting.
2999 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3000 was not being matched caselessly.
3002 24. Arranged for all hyphens in the exim.8 source to be escaped with
3005 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3006 properly. Recipient callout cache records were still being keyed to include
3007 the sender, even when use_sender was set false. This led to far more
3008 callouts that were necessary. The sender is no longer included in the key
3009 when use_sender is false.
3011 26. Added "control = submission" modifier to ACLs.
3013 27. Added the ${base62d: operator to decode base 62 numbers.
3015 28. dnsdb lookups can now access SRV records.
3017 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3018 the configuration file.
3020 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3021 (-v) mode. This makes the output for a verbose queue run more intelligible.
3023 31. Added a use_postmaster feature to recipient callouts.
3025 32. Added the $body_zerocount variable, containing the number of binary zero
3026 bytes in the message body.
3028 33. The time of last modification of the "new" subdirectory is now used as the
3029 "mailbox time last read" when there is a quota error for a maildir
3032 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3034 35. Added +ignore_unknown as a special item in host lists.
3036 36. Code for decoding IPv6 addresses in host lists is now included, even if
3037 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3038 address was recognized as an IP address, but was then not correctly decoded
3039 into binary, causing unexpected and incorrect effects when compared with
3046 1. Very minor rewording of debugging text in manualroute to say "list of
3047 hosts" instead of "hostlist".
3049 2. If verify=header_syntax was set, and a header line with an unqualified
3050 address (no domain) and a large number of spaces between the end of the
3051 name and the colon was received, the reception process suffered a buffer
3052 overflow, and (when I tested it) crashed. This was caused by some obsolete
3053 code that should have been removed. The fix is to remove it!
3055 3. When running in the test harness, delay a bit after writing a bounce
3056 message to get a bit more predictability in the log output.
3058 4. Added a call to search_tidyup() just before forking a reception process. In
3059 theory, someone could use a lookup in the expansion of smtp_accept_max_
3060 per_host which, without the tidyup, could leave open a database connection.
3062 5. Added the variables $recipient_data and $sender_data which get set from a
3063 lookup success in an ACL "recipients" or "senders" condition, or a router
3064 "senders" option, similar to $domain_data and $local_part_data.
3066 6. Moved the writing of debug_print from before to after the "senders" test
3069 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3070 problems for message scanning, either using a data ACL, or using
3071 local_scan() because the Received: header was not generated till after they
3072 were called (in order to set the time as the time of reception completion).
3073 I have revised the way this works. The header is now generated after the
3074 body is received, but before the ACL or local_scan() are called. After they
3075 are run, the timestamp in the header is updated.
3081 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3082 before starting a queue runner without re-exec. This happened only when
3083 deliver_drop_privilege was set or when the Exim user was set to root. The
3084 effect of the bug was that timeouts during subsequent deliveries caused
3085 crashes instead of being properly handled. The handler is now left at its
3086 default (and expected) setting.
3088 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3089 message, again when deliver_drop_privilege is set or Exim is run as root.
3090 The bug described in (1) was not present in this case, but the tidying up
3091 of the other signals was missing. I have made the two cases consistent.
3093 3. The ignore_target_hosts setting on a manualroute router was being ignored
3094 for hosts that were looked up using the /MX notation.
3096 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3099 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3100 operated on the sender address. After changing the $sender_address to <>
3101 for the sender address verify, Exim was re-instated it as the original
3102 (before rewriting) address, but remembering that it had rewritten it, so it
3103 wasn't rewriting it again. This bug also had the effect of breaking the
3104 sender address verification caching when the sender address was rewritten.
3106 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3107 This has been changed so that if the ip literal address matches
3108 ignore_target_hosts, the router declines.
3110 7. Added expansion conditions match_domain, match_address, and match_local_
3111 part (NOT match_host).
3113 8. The placeholder for the Received: header didn't have a length field set.
3115 9. Added code to Exim itself and to exim_lock to test for a specific race
3116 condition that could lead to file corruption when using MBX delivery. The
3117 issue is with the lockfile that is created in /tmp. If this file is removed
3118 after a process has opened it but before that process has acquired a lock,
3119 there is the potential for a second process to recreate the file and also
3120 acquire a lock. This could lead to two Exim processes writing to the file
3121 at the same time. The added code performs the same test as UW imapd; it
3122 checks after acquiring the lock that its file descriptor still refers to
3123 the same named file.
3125 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3126 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3129 11. Added the smtp_active_hostname option. If used, this will typically be made
3130 to depend on the incoming interface address. Because $interface_address is
3131 not set up until the daemon has forked a reception process, error responses
3132 that can happen earlier (such as "too many connections") no longer contain
3135 12. If an expansion in a condition on a "warn" statement fails because a lookup
3136 defers, the "warn" statement is abandoned, and the next ACL statement is
3137 processed. Previously this caused the whole ACL to be aborted.
3139 13. Added the iplsearch lookup type.
3141 14. Added ident_timeout as a log selector.
3143 15. Added tls_certificate_verified as a log selector.
3145 16. Added a global option tls_require_ciphers (compare the smtp transport
3146 option of the same name). This controls incoming TLS connections.
3148 17. I finally figured out how to make tls_require_ciphers do a similar thing
3149 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3150 before starting the TLS session.
3152 18. Tabs are now shown as \t in -bP output.
3154 19. If the log selector return_path_on_delivery was set, Exim crashed when
3155 bouncing a message because it had too many Received: header lines.
3157 20. If two routers both had headers_remove settings, and the first one included
3158 a superfluous trailing colon, the final name in the first list and the
3159 first name in the second list were incorrectly joined into one item (with a
3160 colon in the middle).
3166 1. Added -C and -D options to the exinext utility, mainly to make it easier
3167 to include in the automated testing, but these could be helpful when
3168 multiple configurations are in use.
3170 2. The exinext utility was not formatting the output nicely when there was
3171 an alternate port involved in the retry record key, nor when there was a
3172 message id as well (for retries that were specific to a specific message
3173 and a specific host). It was also confused by IPv6 addresses, because of
3174 the additional colons they contain. I have fixed the IPv4 problem, and
3175 patched it up to do a reasonable job for IPv6.
3177 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3178 delivery, the log line now contains "pipelined" if PIPELINING was used.
3180 4. An SMTP transport process used to panic and die if the bind() call to set
3181 an explicit outgoing interface failed. This has been changed; it is now
3182 treated in the same way as a connect() failure.
3184 5. A reference to $sender_host_name in the part of a conditional expansion
3185 that was being skipped was still causing a DNS lookup. This no longer
3188 6. The def: expansion condition was not recognizing references to header lines
3189 that used bh_ and bheader_.
3191 7. Added the _cache feature to named lists.
3193 8. The code for checking quota_filecount in the appendfile transport was
3194 allowing one more file than it should have been.
3196 9. For compatibility with Sendmail, the command line option
3204 and sets the incoming protocol and host name (for trusted callers). The
3205 host name and its colon can be omitted when only the protocol is to be set.
3206 Note the Exim already has two private options, -pd and -ps, that refer to
3207 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3208 "s", but I don't think that's a major issue.
3210 10. A number of refactoring changes to the code, none of which should affect
3213 (a) The number of logging options was getting close to filling up the
3214 32-bit word that was used as a bit map. I have split them into two classes:
3215 those that are passed in the argument to log_write(), and those that are
3216 only ever tested independently outside of that function. These are now in
3217 separate 32-bit words, so there is plenty of room for expansion again.
3218 There is no change in the user interface or the logging behaviour.
3220 (b) When building, for example, log lines, the code previously used a
3221 macro that called string_cat() twice, in order to add two strings. This is
3222 not really sufficiently general. Furthermore, there was one instance where
3223 it was actually wrong because one of the argument was used twice, and in
3224 one call a function was used. (As it happened, calling the function twice
3225 did not affect the overall behaviour.) The macro has been replaced by a
3226 function that can join an arbitrary number of extra strings onto a growing
3229 (c) The code for expansion conditions now uses a table and a binary chop
3230 instead of a serial search (which was left over from when there were very
3231 few conditions). Also, it now recognizes conditions like "pam" even when
3232 the relevant support is not compiled in: a suitably worded error message is
3233 given if an attempt is made to use such a condition.
3235 11. Added ${time_interval:xxxxx}.
3237 12. A bug was causing one of the ddress fields not to be passed back correctly
3238 from remote delivery subprocesses. The field in question was not being
3239 subsequently used, so this caused to problems in practice.
3241 13. Added new log selectors queue_time and deliver_time.
3243 14. Might have fixed a bug in maildirsizefile handling that threw up
3244 "unexpected character" debug warnings, and recalculated the data
3245 unnecessarily. In any case, I expanded the warning message to give more
3248 15. Added the message "Restricted characters in address" to the statements in
3249 the default ACL that block characters like @ and % in local parts.
3251 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3252 Three changes have been made:
3254 (a) There was a serious bug; a negative response to MAIL caused the whole
3255 recipient domain to be cached as invalid, thereby blocking all messages
3256 to all local parts at the same domain, from all senders. This bug has
3257 been fixed. The domain is no longer cached after a negative response to
3258 MAIL if the sender used is not empty.
3260 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3263 (c) A new callout option, "use_sender" has been added for people who want
3264 the modified behaviour.
3270 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3273 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3274 indeed breaks things for older releases.
3276 3. Added additional logging to the case where there is a problem reading data
3277 from a filter that is running in a subprocess using a pipe, in order to
3278 try to track down a specific problem.
3280 4. Testing facility fudge: when running in the test harness and attempting
3281 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3282 getting "No route to host". Convert this to a timeout.
3284 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3287 6. Some OS don't have socklen_t but use size_t instead. This affects the
3288 fifth argument of getsockopt() amongst other things. This is now
3289 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3290 can be set for individual OS. I have set it for SunOS5, OSF1, and
3291 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3292 some earlier ones do not.
3294 7. Change 4.30/15 was not doing the test caselessly.
3296 8. The standard form for an IPv6 address literal was being rejected by address
3297 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3298 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3299 this, as well as the form without the "IPv6" on the front (but only when
3300 address literals are enabled, of course).
3302 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3304 10. Exim crashed if a message with an empty sender address specified by -f
3305 encountered a router with an errors_to setting. This could be provoked only
3306 by a command such as
3310 where an empty string was supplied; "<>" did not hit this bug.
3312 11. Installed PCRE release 4.5.
3314 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3315 remained set. It is now erased.
3317 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3318 times from message ids (which are base 36 rather than the normal 62).
3320 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3321 were being counted as actual protocol errors, and logged if the log
3322 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3323 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3324 and DATA following a set of rejected RCPTs do not count as protocol errors.
3325 In other words, Exim assumes they were pipelined, though this may not
3326 actually be the case. Of course, in all cases the client gets an
3327 appropriate error code.
3329 15. If a lookup fails in an ACL condition, a message about the failure may
3330 be available; it is used if testing the ACL cannot continue, because most
3331 such messages specify what the cause of the deferral is. However, some
3332 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3333 that caused an old message to be retained and used if a later statement
3334 caused a defer, replacing the real cause of the deferral.
3336 16. If an IP address had so many PTR records that the DNS lookup buffer
3337 was not large enough to hold them, Exim could crash while trying to process
3338 the truncated data. It now detects and logs this case.
3340 17. Further to 4.21/58, another change has been made: if (and only if) the
3341 first line of a message (the first header line) ends with CRLF, a bare LF
3342 in a subsequent header line has a space inserted after it, so as not to
3343 terminate the header.
3345 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3346 unnecessarily, used atoi() instead of strtol(), and didn't check the
3347 termination when getting file sizes from file names by regex.
3349 19. Completely re-implemented the support for maildirsize files, in the light
3350 of a number of problems with the previous contributed implementation
3351 (4.30/29). In particular:
3353 . If the quota is zero, the maildirsize file is maintained, but no quota is
3356 . If the maildir directory does not exist, it is created before any attempt
3357 to write a maildirsize file.
3359 . The quota value in the file is just a cache; if the quota is changed in
3360 the transport, the new value overrides.
3362 . A regular expression is available for excluding directories from the
3365 20. The autoreply transport checks the characters in options that define the
3366 message's headers; it allows continued headers, but it was checking with
3367 isspace() after an embedded newline instead of explicitly looking for a
3370 21. If all the "regular" hosts to which an address was routed had passed their
3371 expiry times, and had not reached their retry times, the address was
3372 bounced, even if fallback hosts were defined. Now Exim should go on to try
3375 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3376 equivalent code in the SMTP transport. Some hosts send humungous responses
3377 to HELO/EHLO, more than 1024 it seems.
3379 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3380 gives compiler warnings in some environments. I've now done it "properly",
3383 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3384 (because of problems with the built-in one) was declared to return uschar *
3385 instead of char *, causing compiler failure.
3387 25. Fixed a file descriptor leak when processing alias/forward files.
3389 26. Fixed a minor format string issue in dbfn.c.
3391 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3393 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3394 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3397 29. When a sender address is verified, it is cached, to save repeating the test
3398 when there is more than one recipient in a message. However, when the
3399 verification involves a callout, it is possible for different callout
3400 options to be set for different recipients. It is too complicated to keep
3401 track of this in the cache, so now Exim always runs a verification when a
3402 callout is required, relying on the callout cache for the optimization.
3403 The overhead is duplication of the address routing, but this should not be
3406 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3407 to be verified with callout=postmaster, and the main callout worked but the
3408 postmaster check failed, the verification correctly failed. However, if a
3409 subsequent RCPT command asked for sender verification *without* the
3410 postmaster check, incorrect caching caused this verification also to fail,
3413 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3414 it was not caching the DNS options (qualify_single, search_parents) that
3415 were used when the lookup failed. A subsequent lookup with different
3416 options therefore always gave the same answer, though there were cases
3417 where it should not have. (Example: a "domains = !$mx_any" option on a
3418 dnslookup router: the "domains" option is always processed without any
3419 widening, but the router might have qualify_single set.) Now Exim uses the
3420 cached value only when the same options are set.
3422 32. Added John Jetmore's "exipick" utility to the distribution.
3424 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3425 than a timeout (e.g. a certificate is required, and is not provided), an
3426 Exim server now closes the connection immediately. Previously it waited for
3427 the client to close - but if the client is SSL, it seems that they each
3428 wait for each other, leading to a delay before one of them times out.
3430 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3431 maintained 0.8.x compatibility because I don't think many are using it, and
3432 it is clearly obsolete.
3434 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3437 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3438 client certificate was expired. A simple patch fixes this, though I don't
3439 understand the full logic of why the verify callback is called multiple
3442 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3443 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3444 which causes problems with some clients (such as the Certicom SSL Plus
3445 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3446 disables the coutermeasure allowing Eudora to connect."
3448 38. Exim was not checking that a write() to a log file succeeded. This could
3449 lead to Bad Things if a log got too big, in particular if it hit a file
3450 size limit. Exim now panics and dies if it cannot write to a log file, just
3451 as it does if it cannot open a log file.
3453 39. Modified OS/Makefile-Linux so that it now contains
3455 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3457 The two -D definitions ensure that Exim is compiled with large file
3458 support, which makes it possible to handle log files that are bigger than
3461 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3462 instance) a domain was checked against a named list that involved a lookup,
3463 causing $domain_data to be set, then another domain was checked against the
3464 same list, then the first domain was re-checked, the value of $domain_data
3465 after the final check could be wrong. In particular, if the second check
3466 failed, it could be set empty. This bug probably also applied to
3469 41. The strip_trailing_dot option was not being applied to the address given
3470 with the -f command-line option.
3472 42. The code for reading a message's header from the spool was incrementing
3473 $received_count, but never initializing it. This meant that the value was
3474 incorrect (doubled) while delivering a message in the same process in which
3475 it was received. In the most common configuration of Exim, this never
3476 happens - a fresh exec is done - but it can happen when
3477 deliver_drop_privilege is set.
3479 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3480 it now includes up to 150 characters of the unexpected data in the log
3483 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3484 and building data strings. The size of both of these buffers was 10 000
3485 bytes - far larger than anybody would *ever* want, thought I. Needless to
3486 say, somebody hit the limit. I have increased the maximum line length to
3487 20 000 and the maximum data length of concatenated lines to 100 000. I have
3488 also fixed two bugs, because there was no checking on these buffers. Tsk,
3489 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3490 buffer is too small.
3492 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3493 lsearch lookups. Now it does.
3495 46. When parsing a route_list item in a manualroute router, a fixed-length
3496 buffer was used for the list of hosts. I made this 1024 bytes long,
3497 thinking that nobody would ever have a list of hosts that long. Wrong.
3498 Somebody had a whole pile of complicated expansion conditions, and the
3499 string was silently truncated, leading to an expansion error. It turns out
3500 that it is easier to change to an unlimited length (owing to other changes
3501 that have happened since this code was originally written) than to build
3502 structure for giving a limitation error. The length of the item that
3503 expands into the list of hosts is now unlimited.
3505 47. The lsearch lookup could not handle data where the length of text line was
3506 more than 4095 characters. Such lines were truncated, leading to shortened
3507 data being returned. It should now handle lines of any length.
3509 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3510 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3513 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3514 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3515 no longer happens in comment lines. A list of replacements is now placed
3516 at the head of all of the source files, except those whose only change is
3517 to replace PERL_COMMAND in the very first #! line.
3519 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3520 messages on the queue, with a bottom-up merge sort, using code contributed
3521 by Michael Haardt. This should make operations like -bp somewhat faster on
3522 large queues. It won't affect queue runners, except when queue_run_in_order
3525 51. Installed eximstats 1.31 in the distribution.
3527 52. Added support for SRV lookups to the dnslookup router.
3529 53. If an ACL referred to $message_body or $message_body_end, the value was not
3530 reset for any messages that followed in the same SMTP session.
3532 54. The store-handling optimization for building very long strings was not
3533 differentiating between the different store pools. I don't think this
3534 actually made any difference in practice, but I've tidied it.
3536 55. While running the routers to verify a sender address, $sender_address
3537 was still set to the sender address. This is wrong, because when routing to
3538 send a bounce to the sender, it would be empty. Therefore, I have changed
3539 it so that, while verifying a sender address, $sender_address is set to <>.
3540 (There is no change to what happens when verifying a recipient address.)
3542 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3543 target A or AAAA records (if not already returned) without resetting the
3544 qualify_single or search_parents options of the DNS resolver. These are
3545 inappropriate in this case because the targets of MX and SRV records must
3546 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3547 target that, when qualified, matched something in the local domain. These
3548 two options are now turned off when doing these lookups.
3550 57. It seems that at least some releases of Reiserfs (which does not have the
3551 concept of a fixed number of inodes) returns zero and not -1 for the
3552 number of available inodes. This interacted badly with check_spool_inodes,
3553 which assumed that -1 was the "no such thing" setting. What I have done is
3554 to check that the total number of inodes is greater than zero before doing
3555 the test of how many are available.
3557 58. When a "warn" ACL statement has a log_message modifier, the message is
3558 remembered, and not repeated. This is to avoid a lot of repetition when a
3559 message has many recipients that cause the same warning to be written.
3560 Howewer, Exim was preserving the list of already written lines for an
3561 entire SMTP session, which doesn't seem right. The memory is now reset if a
3562 new message is started.
3564 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3565 debugging output unless log_rewrite was also set.
3567 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3568 of (int)(handle) when we know that handle contains (void *)(-1).
3570 61. The Exim daemon panic-logs an error return when it closes the incoming
3571 connection. However "connection reset by peer" seems to be common, and
3572 isn't really an error worthy of noting specially, so that particular error
3575 62. When Exim is trying to find all the local interfaces, it used to panic and
3576 die if the ioctl to get the interface flags failed. However, it seems that
3577 on at least one OS (Solaris 9) it is possible to have an interface that is
3578 included in the list of interfaces, but for which you get a failure error
3579 for this call. This happens when the interface is not "plumbed" into a
3580 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3581 failure of the "get flags" call assumes that the interface is down.
3583 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3584 makes life easier for people who are doing arithmetic on fields extracted
3585 from dates, where you often get leading zeros that should not be
3586 interpreted as octal.
3588 64. Added qualify_domain to the redirect router, to override the global
3591 65. If a pathologically long header line contained very many addresses (the
3592 report of this problem mentioned 10 000) and each of them was rewritten,
3593 Exim could use up a very large amount of memory. (It kept on making new
3594 copies of the header line as it rewrote, and never released the old ones.)
3595 At the expense of a bit more processing, the header rewriting function has
3596 been changed so that it no longer eats memory in this way.
3598 66. The generation of the Received: header has been moved from the time that a
3599 message starts to be received, to the time that it finishes. The timestamp
3600 in the Received: header should now be very close to that of the <= log
3601 line. There are two side-effects of this change:
3603 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3604 logged header lines no longer include the local Received: line, because
3605 it has not yet been created. The same applies to a copy of the message
3606 that is returned to a non-SMTP sender when a message is rejected.
3608 (b) When a filter file is tested using -bf, no additional Received: header
3609 is added to the test message. After some thought, I decided that this
3612 This change does not affect the value of $received_for. It is still set
3613 after address rewriting, but before local_scan() is called.
3615 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3617 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3618 gave an unhelpful panic error message, and a defer error. I have managed to
3619 change this behaviour so that it now rejects any supplied certificate,
3620 which seems right, as the list of acceptable certificates is empty.
3622 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3623 gave an unhelpful defer error. I have not managed to make this reject any
3624 supplied certificates, but the error message it gives is "no certificate
3625 supplied", which is not helpful.
3627 70. exigrep's output now also includes lines that are not associated with any
3628 message, but which match the given pattern. Implemented by a patch from
3629 Martin Sluka, which also tidied up the Perl a bit.
3631 71. Recipient callout verification, like sender verification, was using <> in
3632 the MAIL FROM command. This isn't really the right thing, since the actual
3633 sender may affect whether the remote host accepts the recipient or not. I
3634 have changed it to use the actual sender in the callout; this means that
3635 the cache record is now keyed on a recipient/sender pair, not just the
3636 recipient address. There doesn't seem to be a real danger of callout loops,
3637 since a callout by the remote host to check the sender would use <>.
3638 [SEE ABOVE: changed after hitting problems.]
3640 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3641 temporary errors. However, in the case of such a code being given after
3642 the end of a data transmission (i.e. after ".") Exim was failing to write
3643 a retry record for the message. (Yes, there was some broken host that was
3644 actually sending 8xx at this point.)
3646 73. An unknown lookup type in a host list could cause Exim to panic-die when
3647 the list was checked. (An example that provoked this was putting <; in the
3648 middle of a list instead of at the start.) If this happened during a DATA
3649 ACL check, a -D file could be left lying around. This kind of configuration
3650 error no longer causes Exim to die; instead it causes a defer errror. The
3651 incident is still logged to the main and panic logs.
3653 74. Buglet left over from Exim 3 conversion. The message "too many messages
3654 in one connection" was written to the rejectlog but not the mainlog, except
3655 when address rewriting (yes!) was being logged.
3657 75. Added write_rejectlog option.
3659 76. When a system filter was run not as root (that is, when system_filter_user
3660 was set), the values of the $n variables were not being returned to the
3661 main process; thus, they were not subsequently available in the $sn
3664 77. Added +return_path_on_delivery log selector.
3666 78. A connection timeout was being treated differently from recipients deferred
3667 when testing hosts_max_try with a message that was older than the host's
3668 retry timeout. (The host should not be counted, thus allowing all hosts to
3669 be tried at least once before bouncing.) This may have been the cause of an
3670 occasionally reported bug whereby a message would remain on the queue
3671 longer than the retry timeout, but would be bounced if a delivery was
3672 forced. I say "may" because I never totally pinned down the problem;
3673 setting up timeout/retry tests is difficult. See also the next item.
3675 79. The ultimate address timeout was not being applied to errors that involved
3676 a combination of host plus message (for example, a timeout on a MAIL
3677 command). When an address resolved to a number of possible hosts, and they
3678 were not all tried for each delivery (e.g. because of hosts_max_try), a
3679 message could remain on the queue longer than the retry timeout.
3681 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3684 81. Fixed an obscure SMTP outgoing bug which required at least the following
3685 conditions: (a) there was another message waiting for the same server;
3686 (b) the server returned 5xx to all RCPT commands in the first message so
3687 that the message was not completed; (c) the server dropped the connection
3688 or gave a negative response to the RSET that Exim sends to abort the
3689 transaction. The observed case was a dropped connection after DATA that had
3690 been sent in pipelining mode. That is, the server had advertised PIPELINING
3691 but was not implementing it correctly. The effect of the bug was incorrect
3692 behaviour, such as trying another host, and this could lead to a crash.
3698 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3699 and daemon.c were passed as pointers to ints; they should have been
3700 pointers to socklen_t variables (which are typically unsigned ints).
3702 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3705 3. Fixed a really odd bug that affected only the testing scheme; patching a
3706 certain fixed string in the binary changed the value of another string that
3707 happened to be identical to the end of the original first string.
3709 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3710 name", it returns that address as the IP address. On some operating
3711 systems (e.g. Solaris), it also passes back the IP address string as the
3712 "host name". However, on others (e.g. Linux), it passes back an empty
3713 string. Exim wasn't checking for this, and was changing the host name to an
3714 empty string, assuming it had been canonicized.
3716 5. Although rare, it is permitted to have more than one PTR record for a given
3717 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3718 all the names associated with an address, because they do in Solaris.
3719 However, it seems that they do not in Linux for data that comes from the
3720 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3721 I found this out when I moved to a new Linux workstation and tried to run
3722 the Exim test suite.
3724 To get round this problem I have changed the code so that it now does its
3725 own call to the DNS to look up PTR records when searching for a host name.
3726 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3727 addresses that are only in /etc/hosts are still found.
3729 This behaviour is, however, controlled by an option called host_lookup_
3730 order, which defaults to "bydns:byaddr". If people want to use the other
3731 order, or indeed, just use one or the other means of lookup, they can
3732 specify it in this variable.
3734 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3735 some operating systems, this comes back from gethostbyaddr() as an empty
3736 string, and this is what Exim used to test for. However, it seems that in
3737 other systems, "." is yielded. Exim now tests for this case too.
3739 7. The values of check_spool_space and check_log_space are now held internally
3740 as a number of kilobytes instead of an absolute number of bytes. If a
3741 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3742 kilobyte. This means that much larger values can be stored.
3744 8. Exim monitor: an attempt to get the action menu when not actually pointing
3745 at a message produces an empty menu entitled "No message selected". This
3746 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3747 no entries in it ("Shell widget menu has zero width and/or height"). So I
3748 have added a single, blank menu entry in this case.
3750 9. Added ${quote_local_part.
3752 10. MIME decoding is now applied to the contents of Subject: header lines when
3755 11. Now that a reference to $sender_host_address automatically causes a reverse
3756 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3757 host lookup before query-style lookups in lists that might use this
3758 variable. This has therefore been abolished, and the "net-" prefix is no
3759 longer necessary for query-style lookups.
3761 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3762 have been a typo for LFLAGS, so it has been changed.
3764 13. The install script calls Exim with "-C /dev/null" in order to find the
3765 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3766 to be output. Howeve, since Exim outputs its version number before the
3767 error, it didn't break the script. It just looked ugly. I fixed this by
3768 always allowing "-C /dev/null" if the caller is root.
3770 14. Ignore overlarge ACL variable number when reading spool file - insurance
3771 against a later release with more variables having written the file.
3773 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3774 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3775 this, as well as the form without the "IPv6" on the front.
3777 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3778 OS/Makefile-Darwin file.
3780 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3781 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3783 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3784 variable. However, some OS use ints for inodes. I've added cast to long int
3785 to get rid of the compiler warning.
3787 19. I had forgotten to lock out "/../" in configuration file names when
3788 ALT_CONFIG_PREFIX was set.
3790 20. Routers used for verification do not need to specify transports. However,
3791 if such a router generated a host list, and callout was configured, Exim
3792 crashed, because it could not find a port number from the (non-existent)
3793 transport. It now assumes port 25 in this circumstance.
3795 21. Added the -t option to exigrep.
3797 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3798 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3799 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3800 them is set, LOOKUP_LSEARCH is forced.
3802 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3804 24. Added sender and host information to the "rejected by local_scan()" log
3805 line; previously there was no indication of these.
3807 25. Added .include_if_exists.
3809 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3810 Linux. It turns out that not all file systems support this. Apparently some
3811 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3812 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3813 not supported on the file descriptor, is now ignored when Exim is trying to
3814 sync a directory. This applies only to Linux.
3816 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3818 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3819 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3822 29. Added support for maildirsize files from supplied patch (modified a bit).
3824 30. The use of :fail: followed by an empty string could lead Exim to respond to
3825 sender verification failures with (e.g.):
3827 550 Verification failed for <xxx>
3828 550 Sender verify failed
3830 where the first response line was missing the '-' that indicates it is not
3831 the final line of the response.
3833 31. The loop for finding the name of the user that called Exim had a hardwired
3834 limit of 10; it now uses the value of finduser_retries, which is used for
3835 all other user lookups.
3837 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3840 33. Exim was neglecting to zero errno before one call of strtol() when
3841 expanding a string and expecting an integer value. On some systems this
3842 resulted in spurious "integer overflow" errors. Also, it was casting the
3843 result into an int without checking.
3845 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3846 did not work. The code looks as if it has *never* worked, though it appears
3847 to have been documented since at least releast 1.62. I have made it work.
3849 35. The "timeout_DNS" error in retry rules, also documented since at least
3850 1.62, also never worked. As it isn't clear exactly what this means, and
3851 clearly it isn't a major issue, I have abolished the feature by treating it
3852 as "timeout", and writing a warning to the main and panic logs.
3854 36. The display of retry rules for -brt wasn't always showing the error code
3857 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3858 timeout_connect_A, timeout_connect_MX.
3860 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3861 to the empty sender.
3863 39. The daemon was not analysing the content of -oX till after it had closed
3864 stderr and disconnected from the controlling terminal. This meant that any
3865 syntax errors were only noted on the panic log, and the return code from
3866 the command was 0. By re-arranging the code a little, I've made the
3867 decoding happen first, so such errors now appear on stderr, and the return
3868 code is 1. However, the actual setting up of the sockets still happens in
3869 the disconnected process, so errors there are still only recorded on the
3872 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3873 connections (as happens on some IP stacks) was logged at start up time as
3874 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3875 it from "IPv6 and IPv4", which means that two separate sockets are being
3878 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3879 says whether AF_INET or AF_INET6 was passed as an argument.
3881 42. Exiwhat output was messed up when time zones were included in log
3884 43. Exiwhat now gives more information about the daemon's listening ports,
3885 and whether -tls-on-connect was used.
3887 44. The "port" option of the smtp transport is now expanded.
3889 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3890 silently ignored. Now an error message is written to the main and panic
3893 46. There's a new ACL modifier called "logwrite" which writes to a log file
3894 as soon as it is encountered.
3896 47. Added $local_user_uid and $local_user_gid at routing time.
3898 48. Exim crashed when trying to verify a sender address that was being
3901 49. Exim was recognizing only a space character after ".include". It now also
3902 recognizes a tab character.
3904 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3905 extracting the relevant information from the specification. The man page no
3906 longer contains scrambled data for the -d option, and I've added a section
3907 at the front about calling Exim under different names.
3909 51. Added "extra_headers" argument to the "mail" command in filter files.
3911 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3914 53. Installed eximstats 1.29.
3916 54. Added transport_filter_timeout as a generic transport option.
3918 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3919 Cc: header lines. This was required by RFC 822, but it not required by RFC
3922 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3923 incoming messages that did not have them. Now it does so only if the
3924 message originates locally, that is, if there is no associated remote host
3925 address. When Resent- header lines are present, this applies to the Resent-
3926 lines rather than the non-Resent- lines.
3928 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3929 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3931 58. Messages for configuration errors now include the name of the main
3932 configuration file - useful now that there may be more than one file in a
3933 list (.included file names were always shown).
3935 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3936 for those rare installations that do not start the daemon as root or run it
3937 setuid root. I've cut out the call to initgroups() if the daemon is not
3940 60. The Exim user and group can now be bound into the binary as text strings
3941 that are looked up at the start of Exim's processing.
3943 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3945 62. Added $mailstore_basename variable.
3947 63. Installed patch to sieve.c from Michael Haardt.
3949 64. When Exim failed to open the panic log after failing to open the main log,
3950 the original message it was trying to log was written to stderr and debug
3951 output, but if they were not available (the usual case in production), it
3952 was lost. Now it is written to syslog before the two lines that record the
3953 failures to open the logs.
3955 65. Users' Exim filters run in subprocesses under the user's uid. It is
3956 possible for a "deliver" command or an alias in a "personal" command to
3957 provoke an address rewrite. If logging of address rewriting is configured,
3958 this fails because the process is not running as root or exim. There may be
3959 a better way of dealing with this, but for the moment (because 4.30 needs
3960 to be released), I have disabled address rewrite logging when running a
3961 filter in a non-root, non-exim process.
3967 1. The buildconfig auxiliary program wasn't quoting the value set for
3968 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3969 not defined. This bug was masked in 4.22 by the effect that was fixed in
3972 2. Some messages that were rejected after a message id was allocated were
3973 shown as "incomplete" by exigrep. It no longer does this for messages that
3974 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3976 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3977 have allow_domain_literals set, the ID did not get logged in the <= line.
3978 Domain literals are now always recognized in Message-ID: header lines.
3980 4. The first argument for a ${extract expansion item is the key name or field
3981 number. Leading and trailing spaces in this item were not being ignored,
3982 causing some misleading effects.
3984 5. When deliver_drop_privilege was set, single queue runner processes started
3985 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3986 same command in the process it spins off) were not dropping privilege.
3988 6. When the daemon running as "exim" started a queue runner, it always
3989 re-executed Exim in the spun-off process. This is a waste of effort when
3990 deliver_drop_privilege is set. The new process now just calls the
3991 queue-runner function directly.
3997 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4000 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4001 ignored. Though the use of -oP was forcing the writing of a pid file, it
4002 was always written to the default place.
4004 3. If the message "no IP address found for host xxxx" is generated during
4005 incoming verification, it is now followed by identification of the incoming
4006 connection (so you can more easily find what provoked it).
4008 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4010 5. Added some features to "harden" Exim a bit more against certain attacks:
4012 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4013 be put in Local/Makefile. This is like the never_users runtime option,
4014 but it cannot be overridden. The default setting is "root".
4016 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4017 prefix string with which any file named in a -C command line option
4020 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4021 is retained for -C and -D only if the caller of Exim is root. Without
4022 it, the exim user may also use -C and -D and retain privilege.
4024 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4025 command line option is disabled.
4027 6. Macro names set by the -D option must start with an upper case letter, just
4028 like macro names defined in the configuration file.
4030 7. Added "dereference=" facility to LDAP.
4032 8. Two instances of the typo "uknown" in the source files are fixed.
4034 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4035 the Configure-Makefile script screwed up while processing it.
4037 10. Incorporated PCRE 4.4.
4039 11. The SMTP synchronization check was not operating right at the start of an
4040 SMTP session. For example, it could not catch a HELO sent before the client
4041 waited for the greeting. There is now a check for outstanding input at the
4042 point when the greeting is written. Because of the duplex, asynchronous
4043 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4044 way, but not yet received, when the check is performed.
4046 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4047 on TCP/IP sockets, because this apparently causes some broken clients to
4050 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4051 unchanged) from the Cygwin maintainer.
4053 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4054 was supported instead of testing for mbx. Effectively a typo.
4056 15. The spa authenticator server code was not checking that the input it
4057 received was valid base64.
4059 16. The debug output line for the "set" modifier in ACLs was not showing the
4060 name of the variable that was being set.
4062 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4064 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4065 lookup on its own. Something else was needed to trigger the lookup. For
4066 example, a match in host_lookup or the need for a host name in a host list.
4067 Now, if $sender_host_name is referenced and the host name has not yet been
4068 looked up, a lookup is performed. If the lookup fails, the variable remains
4069 empty, and $host_lookup_failed is set to "1".
4071 19. Added "eqi" as a case-independent comparison operator.
4073 20. The saslauthd authentication condition could segfault if neither service
4074 nor realm was specified.
4076 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4077 error message that was logged was misleading, and incoming SMTP
4078 connections were dropped. The message is now more accurate, and temporary
4079 errors are given to SMTP connections.
4081 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4082 (except RSET) with a 421 error, until QUIT is received. However, it was
4083 failing to send a response to QUIT.
4085 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4086 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4087 ACL is now run after the helo_try_verify_hosts code.
4089 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4090 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4091 case-independent, so other case variants are also recognized). Apparently
4092 some systems use these upper case variants.
4094 25. If more than two messages were waiting for the same host, and a transport
4095 filter was specified for the transport, Exim sent two messages over the
4096 same TCP/IP connection, and then failed with "socket operation on non-
4097 socket" when it tried to send the third.
4099 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4101 27. The extern definition of crypt16() in expand.c was not being excluded when
4102 the OS had its own crypt16() function.
4104 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4105 as a preferred synonym for return_size_limit, both as an option and as an
4108 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4110 30. Changed the default configuration ACL to relax the local part checking rule
4111 for addresses that are not in any local domains. For these addresses,
4112 slashes and pipe symbols are allowed within local parts, but the sequence
4113 /../ is explicitly forbidden.
4115 31. SPA server authentication was not clearing the challenge buffer before
4118 32. log_message in a "warn" ACL statement was writing to the reject log as
4119 well as to the main log, which contradicts the documentation and doesn't
4120 seem right (because no rejection is happening). So I have stopped it.
4122 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4123 However, I am unable to do any testing of this.
4125 34. Fixed an infelicity in the appendfile transport. When checking directories
4126 for a mailbox, to see if any needed to be created, it was accidentally
4127 using path names with one or more superfluous leading slashes; tracing
4128 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4130 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4131 contents are added to the log line that is written for every discarded
4132 recipient. (Previously a log_message setting was ignored.)
4134 36. The ${quote: operator now quotes the string if it is empty.
4136 37. The install script runs exim in order to find its version number. If for
4137 some reason other than non-existence or emptiness, which it checks, it
4138 could not run './exim', it was installing it with an empty version number,
4139 i.e. as "exim-". This error state is now caught, and the installation is
4142 38. An argument was missing from the function that creates an error message
4143 when Exim fails to connect to the socket for saslauthd authentication.
4144 This could cause Exim to crash, or give a corrupted message.
4146 39. Added isip, isip4, and isip6 to ${if conditions.
4148 40. The ACL variables $acl_xx are now saved with the message, and can be
4149 accessed later in routers, transports, and filters.
4151 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4152 strings in the file are not string-expanded.
4154 42. If a MAIL command specified a SIZE value that was too large to fit into an
4155 int variable, the check against message_size_limit failed. Such values are
4156 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4157 day this will have to be increased, but I don't think I want to be around
4158 when emails are that large.
4165 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4166 iconv() is not standard in FreeBSD.
4168 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4169 IPv6 enabled. The observed symptom was a segmentation fault on return from
4170 the function os_common_find_running_interfaces() in src/os.c.
4172 3. In the check_special_case() function in daemon.c I had used "errno" as an
4173 argument name, which causes warnings on some systems. This was basically a
4174 typo, since it was named "eno" in the comments!
4176 4. The code that waits for the clock to tick (at a resolution of some fraction
4177 of a second) so as to ensure message-id uniqueness was always waiting for
4178 at least one whole tick, when it could have waited for less. [This is
4179 almost certainly not relevant at current processor speeds, where it is
4180 unlikely to ever wait at all. But we try to future-proof.]
4182 5. The function that sleeps for a time interval that includes fractions of a
4183 second contained a race. It did not block SIGALRM between setting the
4184 timer, and suspending (a couple of lines later). If the interval was short
4185 and the sigsuspend() was delayed until after it had expired, the suspension
4186 never ended. On busy systems this could lead to processes getting stuck for
4189 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4190 process, before it forks any delivery processes. The open lookup caching
4191 mechanism meant that the open file or database connection was passed into
4192 the delivery process. The problem was that delivery processes always tidy
4193 up cached lookup data. This could cause a problem for the next delivery
4194 process started by the queue runner, because the external queue runner
4195 process does not know about the closure. So the next delivery process
4196 still has data in the lookup cache. In the case of a file lookup, there was
4197 no problem because closing a file descriptor in a subprocess doesn't affect
4198 the parent. However, if the lookup was caching a connection to a database,
4199 the connection was closed, and the second delivery process was likely to
4200 see errors such as "PGSQL: query failed: server closed the connection
4201 unexpectedly". The problem has been fixed by closing all cached lookups
4202 in a queue runner before running a delivery process.
4204 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4205 seem to have the "const" qualifier which it has on other OS. I've
4208 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4209 *of the same type* (client or server) with the same public name that an
4210 error should be diagnosed.
4212 9. When Exim looked up a host name for an IP address, but failed to find the
4213 original IP address when looking up the host name (a safety check), it
4214 output the message "<ip address> does not match any IP for NULL", which was
4215 confusing, to say the least. The bug was that the host name should have
4216 appeared instead of "NULL".
4218 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4219 user that is built into the binary, and the -C or -D options is used, root
4220 privilege is dropped before the configuration file is read. In addition,
4221 logging is switched to stderr instead of the normal log files. If the
4222 configuration then re-defines the Exim user, the unprivileged environment
4223 is probably not what is expected, so Exim logs a panic warning message (but
4226 However, if deliver_drop_privilege is set, the unprivileged state may well
4227 be exactly what is intended, so the warning has been cut out in that case,
4228 and Exim is allowed to try to write to its normal log files.
4234 1. smtp_return_error_details was not giving details for temporary sender
4235 or receiver verification errors.
4237 2. Diagnose a configuration error if two authenticators have the same public
4240 3. Exim used not to create the message log file for a message until the first
4241 delivery attempt. This could be confusing when incoming messages were held
4242 for policy or load reasons. The message log file is now created at the time
4243 the message is received, and an initial "Received" line is written to it.
4245 4. The automatically generated man page for command line options had a minor
4246 bug that caused no ill effects; however, a more serious problem was that
4247 the procedure for building the man page automatically didn't always
4248 operate. Consequently, release 4.20 contains an out-of-date version. This
4249 shouldn't happen again.
4251 5. When building Exim with embedded Perl support, the script that builds the
4252 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4253 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4255 6. The freeze_tell option was not being used for messages that were frozen on
4256 arrival, either by an ACL or by local_scan().
4258 7. Added the smtp_incomplete_transaction log selector.
4260 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4261 was accepting AUTH without a new EHLO.
4263 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4264 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4267 10. Logging of TCP/IP connections (when configured) now happens in the main
4268 daemon process instead of the child process, so that the TCP/IP connection
4269 count is more accurate (but it can never be perfect).
4271 11. The use of "drop" in a nested ACL was not being handled correctly in the
4272 outer ACL. Now, if condition failure induced by the nested "drop" causes
4273 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4274 or "require"), the connection is dropped.
4276 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4277 that yield "discard" can now be used with an "accept" or a "discard" verb,
4278 but an error is generated for any others (because I can't see a useful way
4279 to define what should happen).
4281 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4282 are now processed in the same way as lines in the Exim configuration file.
4283 In particular, continuation lines are supported.
4285 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4287 15. Added -ti meaning -t -i.
4289 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4290 domains, and warn by logging if others are found.
4292 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4293 returned from GIFCONF when Exim is trying to find the list of interfaces on
4294 a host. The code in os.c has been modified to copy each ifreq to an aligned
4295 structure in all cases.
4297 Also, in some cases, the returned ifreq's were being copied to a 'struct
4298 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4299 means the last couple of bytes of an IPv6 address could be chopped if the
4300 ifreq contained only a normal sockaddr (14 bytes storage).
4302 18. Named domain lists were not supported in the hosts_treat_as_local option.
4303 An entry such as +xxxx was not recognized, and was treated as a literal
4306 19. Ensure that header lines added by a DATA ACL are included in the reject log
4307 if the ACL subsequently rejects the message.
4309 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4310 MD5 (which is deprecated).
4312 21. When testing a filter file using -bf, Exim was writing a message when it
4313 took the sender from a "From " line in the message, but it was not doing so
4314 when it took $return_path from a Return-Path: header line. It now does.
4316 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4317 with a valid header line field name (a series of printing characters
4318 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4320 23. Changed "disc" in the source to "disk" to conform to the documentation and
4321 the book and for uniformity.
4323 24. Ignore Sendmail's -Ooption=value command line item.
4325 25. When execve() failed while trying to run a command in a pipe transport,
4326 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4327 could be confused with a return value of 69 from the command itself. This
4328 has been changed to 127, the value the shell returns if it is asked to run
4329 a non-existent command. The wording for the related log line suggests a
4330 non-existent command as the problem.
4332 26. If received_header_text expands to an empty string, do not add a Received:
4333 header line to the message. (Well, it adds a token one on the spool, but
4334 marks it "old" so that it doesn't get used or transmitted.)
4336 27. Installed eximstats 1.28 (addition of -nt option).
4338 28. There was no check for failure on the call to getsockname() in the daemon
4339 code. This can fail if there is a shortage of resources on the system, with
4340 ENOMEM, for example. A temporary error is now given on failure.
4342 29. Contrary to the C standard, it seems that in some environments, the
4343 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4344 program. Exim now does this explicitly; it affects the formatting of
4345 timestamps using strftime().
4347 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4348 complaints. I've now initialized all the variables, to avoid this.
4350 32. Header lines added by a system filter were not being "seen" during
4351 transport-time rewrites.
4353 33. The info_callback() function passed to OpenSSL is set up with type void
4354 (*)(SSL *, int, int), as described somewhere. However, when calling the
4355 function (actually a macro) that sets it up, the type void(*)() is
4356 expected. I've put in a cast to prevent warnings from picky compilers.
4358 34. If a DNS black list lookup found a CNAME record, but there were no A
4359 records associated with the domain it pointed at, Exim crashed.
4361 35. If a DNS black list lookup returned more than one A record, Exim ignored
4362 all but the first. It now scans all returned addresses if a particular IP
4363 value is being sought. In this situation, the contents of the
4364 $dnslist_value variable are a list of all the addresses, separated by a
4367 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4368 to accept a host name and all its aliases if the forward lookup for any of
4369 them yielded the IP address of the incoming connection. Now it accepts only
4370 those names whose forward lookup yields the correct IP address. Any other
4371 names are discarded. This closes a loophole whereby a rogue DNS
4372 administrator could create reverse DNS records to break through a
4373 wildcarded host restriction in an ACL.
4375 37. If a user filter or a system filter that ran in a subprocess used any of
4376 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4377 the wrong values were passed to the pipe command ($thisaddress had the
4378 value of $0, $0 had the value of $1, etc). This bug was introduced by
4379 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4381 38. Improved the line breaking for long SMTP error messages from ACLs.
4382 Previously, if there was no break point between 40 and 75 characters, Exim
4383 left the rest of the message alone. Two changes have been made: (a) I've
4384 reduced the minimum length to 35 characters; (b) if it can't find a break
4385 point between 35 and 75 characters, it looks ahead and uses the first one
4386 that it finds. This may give the occasional overlong line, but at least the
4387 remaining text gets split now.
4389 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4390 file descriptors might be low, and that setting 1000 would always raise it.
4391 It turns out that in some environments, the limit is already over 1000 and
4392 that lowering it causes trouble. So now Exim takes care not to decrease it.
4394 40. When delivering a message, the value of $return_path is set to $sender_
4395 address at the start of routing (routers may change the value). By an
4396 oversight, this default was not being set up when an address was tested by
4397 -bt or -bv, which affected the outcome if any router or filter referred to
4400 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4401 log only when "message" or "log_message" are set. However, if one of the
4402 conditions was an address verification, or a call to a nested ACL, the
4403 messages generated by the underlying test were being passed through. This
4404 no longer happens. The underlying message is available in $acl_verify_
4405 message for both "message" and "log_message" expansions, so it can be
4406 passed through if needed.
4408 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4409 new expansion $bh_ to give the encoded byte string without charset
4410 translation. Translation happens only if iconv() is available; HAVE_ICONV
4411 indicates this at build time. HEADERS_CHARSET gives the charset to
4412 translate to; headers_charset can change it in the configuration, and
4413 "headers charset" can change it in an individual filter file.
4415 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4416 that creates RFC 2047 encoded "words" labels them as that charset instead
4417 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4418 expansion operator; (ii) when Exim creates a From: line for a local
4419 message; (iii) when a header line is rewritten to include a "phrase" part.
4421 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4422 buggy, causing it to skip the first lines of messages whose message ID
4423 ended in 'D'. This would not have bitten before Exim release 4.14, because
4424 message IDs were unlikely to end in 'D' before then. The effect was to have
4425 incorrect size information for certain domains.
4427 45. #include "config.h" was missing at the start of the crypt16.c module. This
4428 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4431 46. If there was a timeout during a "random" callout check, Exim treated it as
4432 a failure of the random address, and carried on sending RSET and the real
4433 address. If the delay was just some slowness somewhere, the response to the
4434 original RCPT would be taken as a response to RSET and so on, causing
4435 mayhem of various kinds.
4437 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4438 when I implemented it. It didn't allow for the fact that some option values
4439 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4440 the right test for positive values.
4442 48. Domain names in DNS records are case-independent. Exim always looks them up
4443 in lower case. Some resolvers return domain names in exactly the case they
4444 appear in the zone file, that is, they may contain uppercase letters. Not
4445 all resolvers do this - some return always lower case. Exim was treating a
4446 change of case by a resolver as a change of domain, similar to a widening
4447 of a domain abbreviation. This triggered its re-routing code and so it was
4448 trying to route what was effectively the same domain again. This normally
4449 caused routing to fail (because the router wouldn't handle the domain
4450 twice). Now Exim checks for this case specially, and just changes the
4451 casing of the domain that it ultimately uses when it transmits the message
4454 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4457 50. If a filter generated a file delivery with a non-absolute name (possible if
4458 no home directory exists for the router), the forbid_file option was not
4461 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4462 the existing equality matching.
4464 52. Exim was using ints instead of ino_t variables in some places where it was
4465 dealing with inode numbers.
4467 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4468 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4469 called TMPDIR, and if it finds it is different, it changes its value.
4471 54. The smtp_printf() function is now made available to local_scan() so
4472 additional output lines can be written before returning. There is also an
4473 smtp_fflush() function to enable the detection of a dropped connection.
4474 The variables smtp_input and smtp_batched_input are exported to
4477 55. Changed the default runtime configuration: the message "Unknown user"
4478 has been removed from the ACL, and instead placed on the localuser router,
4479 using the cannot_route_message feature. This means that any verification
4480 failures that generate their own messages won't get overridden. Similarly,
4481 the "Unrouteable address" message that was in the ACL for unverifiable
4482 relay addresses has also been removed.
4484 56. Added hosts_avoid_esmtp to the smtp transport.
4486 57. The exicyclog script was not checking for the esoteric option
4487 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4488 will work only if exicyclog is run under the appropriate euid.
4490 58. Following a discussion on the list, the rules by which Exim recognises line
4491 endings on incoming messages have been changed. The -dropcr and drop_cr
4492 options are now no-ops, retained only for backwards compatibility. The
4493 following line terminators are recognized: LF CRLF CR. However, special
4494 processing applies to CR:
4496 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4497 nor a local message in the state where . is a terminator.
4499 (ii) If a bare CR is encountered in a header line, an extra space is added
4500 after the line terminator so as not to end the header. The reasoning
4501 behind this is that bare CRs in header lines are most likely either
4502 to be mistakes, or people trying to play silly games.
4504 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4505 was being incorrectly given as 18 bytes larger than it should have been.
4506 This is a VOB (very old bug).
4508 60. This may never have affected anything current, but just in case it has:
4509 When the local host is found other than at the start of a list of hosts,
4510 the local host, those with the same MX, and any that follow, are discarded.
4511 When the list in question was part of a longer list of hosts, the following
4512 hosts (not currently being processed) were also being discarded. This no
4513 longer happens. I'm not sure if this situation could ever has previously
4516 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4519 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4520 header. This is something that is recommended in a new Internet Draft, and
4521 is something that is documented as being done by Sendmail. There are two
4522 possible values. For messages generated by the autoreply transport, Exim
4525 Auto-Submitted: auto-replied
4527 whereas for all other generated messages (e.g. bounces) it adds
4529 Auto-Submitted: auto-generated
4531 63. The "personal" condition in filters now includes a test for the
4532 Auto-Submitted: header. If it contains the string "auto-" the message it
4533 not considered personal.
4535 64. Added rcpt_include_affixes as a generic transport option.
4537 65. Added queue_only_override (default true).
4539 66. Added the syslog_duplication option.
4541 67. If what should have been the first header line of a message consisted of
4542 a space followed by a colon, Exim was mis-interpreting it as a header line.
4543 It isn't of course - it is syntactically invalid and should therefore be
4544 treated as the start of the message body. The misbehaviour could have
4545 caused a number of strange effects, including loss of data in subsequent
4546 header lines, and spool format errors.
4548 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4549 client host had authenticated. This control can now be exercised by an ACL
4550 for more flexibility.
4552 69. By default, callouts do not happen when testing with -bh. There is now a
4553 variant, -bhc, which does actually run the callout code, including
4554 consulting and updating the callout cache.
4556 70. Added support for saslauthd authentication, courtesy of Alexander
4559 71. If statvfs() failed on the spool or log directories while checking their
4560 size for availability, Exim confusingly gave the error "space shortage".
4561 Furthermore, in debugging mode it crashed with a floating point exception.
4562 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4563 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4564 serious problem, Exim now writes to the main and panic logs when this
4565 happens, with details of the failure. It then refuses to accept the
4566 incoming message, giving the message "spool directory problem" or "log
4567 directory problem" with a 421 code for SMTP messages.
4569 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4570 0, 1, and 2 exist, because some OS complain for execs without them (see
4571 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4572 descriptors. However, the code omitted to check that the open succeeded,
4573 causing mysterious errors if for some reason the permissions on /dev/null
4574 got screwed. Now Exim writes a message to the main and panic logs, and
4575 bombs out if it can't open /dev/null.
4577 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4578 interact so that it is all more flexible. It is supposed to remain
4579 backwards compatible. Also added extra_local_interfaces.
4581 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4582 to bomb out with an assertion failure - to the client this appears as a
4583 connection drop. This problem occurs in the part of the code that was taken
4584 from the Samba project. Fortunately, the assertion is in a very simple
4585 function, so I have fixed this by reproducing the function inline in the
4586 one place where it is called, and arranging for authentication to fail
4587 instead of killing the process with assert().
4589 75. The SPA client code was not working when the server requested OEM rather
4590 than Unicode encoding.
4592 76. Added code to make require_files with a specific uid setting more usable in
4593 the case where statting the file as root fails - usually a non-root-mounted
4594 NFS file system. When this happens and the failure is EACCES, Exim now
4595 forks a subprocess and does the per-uid checking as the relevant uid.
4597 77. Added process_log_path.
4599 78. If log_file_path was not explicitly set, a setting of check_log_space or
4600 check_log_inodes was ignored.
4602 79. If a space check for the spool or log partitions fails, the incident is now
4603 logged. Of course, in the latter case the data may get lost...
4605 80. Added the %p formatting code to string_format() so that it can be used to
4606 print addresses in debug_print(). Adjusted all the address printing in the
4607 debugging in store.c to use %p rather than %d.
4609 81. There was a concern that a line of code in smtp_in.c could overflow a
4610 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4611 initially expressed, the concern was not well-founded, because trailing
4612 spaces are removed early. However, if the trailing spaces were followed by
4613 a NULL, they did not get removed, so the overflow was possible. Two fixes
4616 (a) I re-wrote the offending code in a cleaner fashion.
4617 (b) If an incoming SMTP command contains a NULL character, it is rejected
4620 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4621 runs initgroups(), so that if the Exim user is in any additional groups,
4622 they will be used during message reception.
4628 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / jgh / exim.git / blob