1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.486 2007/03/01 11:17:00 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.
112 PH/27 When a message times out after an over-quota error from an Exim-imposed
113 quota, the bounce message says "mailbox is full". This message was not
114 being given when it was a system quota that was exceeded. It now should
117 MH/03 Made $recipients available in local_scan(). local_scan() already has
118 better access to the recipient list through recipients_list[], but
119 $recipients can be useful in postmaster-provided expansion strings.
121 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
122 in the case of a MAIL command with additional options following the
123 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
124 were accidentally chopped off.
126 PH/29 SMTP synchronization checks are implemented when a command is read -
127 there is a check that no more input is waiting when there shouldn't be
128 any. However, for some commands, a delay in an ACL can mean that it is
129 some time before the response is written. In this time, more input might
130 arrive, invalidly. So now there are extra checks after an ACL has run for
131 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
132 pipelining has not been advertised.
134 PH/30 MH's patch to allow iscntrl() characters to be list separators.
136 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
137 returned in the SMTP response when smtp_return_error_details was false.
144 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
145 fixed by 4.65/MH/01 (is this a record?) are fixed:
147 (i) An empty string was always treated as zero by the numeric comparison
148 operators. This behaviour has been restored.
150 (ii) It is documented that the numeric comparison operators always treat
151 their arguments as decimal numbers. This was broken in that numbers
152 starting with 0 were being interpreted as octal.
154 While fixing these problems I realized that there was another issue that
155 hadn't been noticed. Values of message_size_limit (both the global option
156 and the transport option) were treated as octal if they started with 0.
157 The documentation was vague. These values are now always treated as
158 decimal, and I will make that clear in the documentation.
164 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
165 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
168 MH/01 Don't check that the operands of numeric comparison operators are
169 integers when their expansion is in "skipping" mode (fixes bug
170 introduced by 4.64-PH/07).
172 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
173 child addresses, Exim now panics and dies. Previously, because the count
174 is held in a short int, deliveries were likely to be lost. As such a
175 large number of recipients for a single message is ridiculous
176 (performance will be very, very poor), I have chosen to impose a limit
177 rather than extend the field.
183 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
184 leftover -K file (the existence of which was triggered by #402).
185 While we were at it, introduced process PID as part of the -K
186 filename. This should rule out race conditions when creating
189 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
190 processing considerably. Previous code took too long for large mails,
191 triggering a timeout which in turn triggers #401.
193 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
194 in the DK code in transports.c. sendfile() is not really portable,
195 hence the _LINUX specificness.
197 TF/01 In the add_headers option to the mail command in an Exim filter,
198 there was a bug that Exim would claim a syntax error in any
199 header after the first one which had an odd number of characters
202 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
203 callout verification, Exim cached a "reject" for the entire domain. This
204 is correct for most verifications, but it is not correct for a recipient
205 verification with use_sender or use_postmaster set, because in that case
206 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
207 case of MAIL FROM:<> rejection from other early rejections (e.g.
208 rejection of HELO). When verifying a recipient using a non-null MAIL
209 address, the cache is ignored if it shows MAIL FROM:<> rejection.
210 Whatever the result of the callout, the value of the domain cache is
211 left unchanged (for any other kind of callout, getting as far as trying
212 RCPT means that the domain itself is ok).
214 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
217 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
218 manifest itself as EPIPE rather than ECONNECT. When tidying away a
219 session, the daemon ignores ECONNECT errors and logs others; it now
220 ignores EPIPE as well.
222 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
223 (quoted-printable decoding).
225 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
226 later the small subsequent patch to fix an introduced bug.
228 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
230 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
232 PH/08 An error is now given if message_size_limit is specified negative.
234 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
235 to be given (somewhat) arbitrary names.
237 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
240 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
241 miscellaneous code fixes
243 PH/10 Added the log_reject_target ACL modifier to specify where to log
246 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
247 hostname. This is wrong, because it relates to the incoming message (and
248 probably the interface on which it is arriving) and not to the outgoing
249 callout (which could be using a different interface). This has been
250 changed to use the value of the helo_data option from the smtp transport
251 instead - this is what is used when a message is actually being sent. If
252 there is no remote transport (possible with a router that sets up host
253 addresses), $smtp_active_hostname is used.
255 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
256 tweaks were necessary in order to get it to work (see also 21 below):
257 (a) The code assumed that strncpy() returns a negative number on buffer
258 overflow, which isn't the case. Replaced with Exim's string_format()
260 (b) There were several signed/unsigned issues. I just did the minimum
261 hacking in of casts. There is scope for a larger refactoring.
262 (c) The code used strcasecmp() which is not a standard C function.
263 Replaced with Exim's strcmpic() function.
264 (d) The code set only $1; it now sets $auth1 as well.
265 (e) A simple test gave the error "authentication client didn't specify
266 service in request". It would seem that Dovecot has changed its
267 interface. Fortunately there's a specification; I followed it and
268 changed what the client sends and it appears to be working now.
270 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
273 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
274 address A is aliased to B and C, where B exists and C does not. Without
275 -v the output is "A verified" because verification stops after a
276 successful redirection if more than one address is generated. However,
277 with -v the child addresses are also verified. Exim was outputting "A
278 failed to verify" and then showing the successful verification for C,
279 with its parentage. It now outputs "B failed to verify", showing B's
280 parentage before showing the successful verification of C.
282 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
283 look up a TXT record in a specific list after matching in a combined
286 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
287 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
288 they consult the DNS. I had assumed they would set it the way they
289 wanted; and indeed my experiments on Linux seem to show that in some
290 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
291 To be on the safe side, however, I have now made the interface to
292 host_find_byname() similar to host_find_bydns(), with an argument
293 containing the DNS resolver options. The host_find_byname() function now
294 sets these options at its start, just as host_find_bydns() does. The smtp
295 transport options dns_qualify_single and dns_search_parents are passed to
296 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
297 of host_find_byname() use the default settings of RES_DEFNAMES
298 (qualify_single) but not RES_DNSRCH (search_parents).
300 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
301 spool_read_header() do less string testing, by means of a preliminary
302 switch on the second character of optional "-foo" lines. (This is
303 overdue, caused by the large number of possibilities that now exist.
304 Originally there were few.) While I was there, I also converted the
305 str(n)cmp tests so they don't re-test the leading "-" and the first
306 character, in the hope this might squeeze out yet more improvement.
308 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
309 flag allowing group syntax was set by the header_syntax check but not
310 turned off, possible causing trouble later; (2) The flag was not being
311 set at all for the header_verify test, causing "group"-style headers to
312 be rejected. I have now set it in this case, and also caused header_
313 verify to ignore an empty address taken from a group. While doing this, I
314 came across some other cases where the code for allowing group syntax
315 while scanning a header line wasn't quite right (mostly, not resetting
316 the flag correctly in the right place). These bugs could have caused
317 trouble for malformed header lines. I hope it is now all correct.
319 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
320 with the "reply" argument non-NULL. The code, however (which originally
321 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
322 but it didn't always do it. This confused somebody who was copying the
323 code for some other use. I have removed all the tests.
325 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
326 feature that was used to support insecure browsers during the U.S. crypto
327 embargo. It requires special client support, and Exim is probably the
328 only MTA that supported it -- and would never use it because real RSA is
329 always available. This code has been removed, because it had the bad
330 effect of slowing Exim down by computing (never used) parameters for the
331 RSA_EXPORT functionality.
333 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
334 authenticator to fail if there's a tab character in the incoming data
335 (there should never be unless someone is messing about, as it's supposed
336 to be base64-encoded). Also added, on Timo's advice, the "secured" option
337 if the connection is using TLS or if the remote IP is the same as the
338 local IP, and the "valid-client-cert option" if a client certificate has
341 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
342 authenticators. This can be used for authorization after authentication
343 succeeds. (In the case of plaintext, it servers for both authentication
346 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
347 if any retry times were supplied.
349 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
350 connection, where there is no client IP address to check. In this
351 situation, the verify now always succeeds.
353 PH/25 Applied John Jetmore's -Mset patch.
355 PH/26 Added -bem to be like -Mset, but loading a message from a file.
357 PH/27 In a string expansion for a processed (not raw) header when multiple
358 headers of the same name were present, leading whitespace was being
359 removed from all of them, but trailing whitespace was being removed only
360 from the last one. Now trailing whitespace is removed from each header
361 before concatenation. Completely empty headers in a concatenation (as
364 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
365 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
367 PH/29 [Removed. This was a change that I later backed out, and forgot to
368 correct the ChangeLog entry (that I had efficiently created) before
369 committing the later change.]
371 PH/30 Exim was sometimes attempting to deliver messages that had suffered
372 address errors (4xx response to RCPT) over the same connection as other
373 messages routed to the same hosts. Such deliveries are always "forced",
374 so retry times are not inspected. This resulted in far too many retries
375 for the affected addresses. The effect occurred only when there were more
376 hosts than the hosts_max_try setting in the smtp transport when it had
377 the 4xx errors. Those hosts that it had tried were not added to the list
378 of hosts for which the message was waiting, so if all were tried, there
379 was no problem. Two fixes have been applied:
381 (i) If there are any address or message errors in an SMTP delivery, none
382 of the hosts (tried or untried) are now added to the list of hosts
383 for which the message is waiting, so the message should not be a
384 candidate for sending over the same connection that was used for a
385 successful delivery of some other message. This seems entirely
386 reasonable: after all the message is NOT "waiting for some host".
387 This is so "obvious" that I'm not sure why it wasn't done
388 previously. Hope I haven't missed anything, but it can't do any
389 harm, as the worst effect is to miss an optimization.
391 (ii) If, despite (i), such a delivery is accidentally attempted, the
392 routing retry time is respected, so at least it doesn't keep
393 hammering the server.
395 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
396 in ${readsocket because some servers need this prod.
398 PH/32 Added some extra debug output when updating a wait-xxx database.
400 PH/33 The hint "could be header name not terminated by colon", which has been
401 given for certain expansion errors for a long time, was not being given
402 for the ${if def:h_colon_omitted{... case.
404 PH/34 The spec says: "With one important exception, whenever a domain list is
405 being scanned, $domain contains the subject domain." There was at least
406 one case where this was not true.
408 PH/35 The error "getsockname() failed: connection reset by peer" was being
409 written to the panic log as well as the main log, but it isn't really
410 panic-worthy as it just means the connection died rather early on. I have
411 removed the panic log writing for the ECONNRESET error when getsockname()
414 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
415 runs only) independently of the message's sender address. This meant
416 that, if the 4xx error was in fact related to the sender, a different
417 message to the same recipient with a different sender could confuse
418 things. In particualar, this can happen when sending to a greylisting
419 server, but other circumstances could also provoke similar problems.
420 I have changed the default so that the retry time for these errors is now
421 based a combination of the sender and recipient addresses. This change
422 can be overridden by setting address_retry_include_sender=false in the
425 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
426 remote server are returned as part of bounce messages. This was not
427 happening for LMTP over a pipe (the lmtp transport), but now it is the
428 same for both kinds of LMTP.
430 PH/38 Despite being documented as not happening, Exim was rewriting addresses
431 in header lines that were in fact CNAMEs. This is no longer the case.
433 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
434 and queue runs started by the daemon processed all messages. This has
435 been fixed so that -R and -S can now usefully be given with -q<time>.
437 PH/40 Import PCRE release 6.7 (fixes some bugs).
439 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
441 PH/42 Give an error if -q is specified more than once.
443 PH/43 Renamed the variables $interface_address and $interface_port as
444 $received_ip_address and $received_port, to make it clear that these
445 values apply to message reception, and not to the outgoing interface when
446 a message is delivered. (The old names remain recognized, of course.)
448 PH/44 There was no timeout on the connect() call when using a Unix domain
449 socket in the ${readsocket expansion. There now is.
451 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
452 be meaningful with "accept".
454 SC/01 Eximstats V1.43
455 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
457 SC/02 Eximstats V1.44
458 Use a glob alias rather than an array ref in the generated
459 parser. This improves both readability and performance.
461 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
462 Collect SpamAssassin and rejection statistics.
463 Don't display local sender or destination tables unless
464 there is data to show.
465 Added average volumes into the top table text output.
467 SC/04 Eximstats V1.46
468 Collect data on the number of addresses (recipients)
469 as well as the number of messages.
471 SC/05 Eximstats V1.47
472 Added 'Message too big' to the list of mail rejection
473 reasons (thanks to Marco Gaiarin).
475 SC/06 Eximstats V1.48
476 Mainlog lines which have GMT offsets and are too short to
477 have a flag are now skipped.
479 SC/07 Eximstats V1.49 (Alain Williams)
480 Added the -emptyok flag.
482 SC/08 Eximstats V1.50
483 Fixes for obtaining the IP address from reject messages.
485 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
486 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
487 whitesspace changes from 4.64-PH/27
489 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
492 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
493 are found, allow negative numbers in numeric criteria)
495 JJ/06 exipick.20061117.2, added new $message_body_missing variable
497 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
498 to match changes made in 4.64-PH/43
500 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
502 PH/47 Put in an explicit test for a DNS lookup of an address record where the
503 "domain" is actually an IP address, and force a failure. This locks out
504 those revolvers/nameservers that support "A-for-A" lookups, in
505 contravention of the specifications.
507 PH/48 When a host name was looked up from an IP address, and the subsequent
508 forward lookup of the name timed out, the host name was left in
509 $sender_host_name, contrary to the specification.
511 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
512 restricted to single-key lookups, Exim was not diagnosing an error if
513 * or *@ was used with a query-style lookup.
515 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
517 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
518 long ago, but noone interested enough thought of it. Let's just say that
519 the "1.1" means that there are some new functions that weren't there at
520 some point in the past.
522 PH/51 Error processing for expansion failure of helo_data from an smtp
523 transport during callout processing was broken.
525 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
526 tested/used via the -bh/-bhc/-bs options.
528 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
529 bug, fixed in subsequent PCRE releases).
531 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
532 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
534 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
540 SC/01 Use a glob alias rather than an array ref in eximstats generated
541 parser. This improves both readability and performance.
543 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
544 Don't display local sender or destination tables in eximstats unless
545 there is data to show.
546 Added average volumes into the eximstats top table text output.
548 SC/03 Collect data on the number of addresses (recipients) as well
549 as the number of messages in eximstats.
551 TF/01 Correct an error in the documentation for the redirect router. Exim
552 does (usually) call initgroups() when daemonizing.
554 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
555 with consistent privilege compared to when running as a daemon.
557 TF/03 Note in the spec that $authenticated_id is not set for local
558 submissions from trusted users.
560 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
561 Thanks to Dean Brooks <dean@iglou.com> for the patch.
563 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
564 by adding some example configuration directives to the default
565 configuration file. A little bit of work is required to uncomment the
566 directives and define how usernames and passwords are checked, but
567 there is now a framework to start from.
569 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
570 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
571 without this. I don't know how relevant this is to other LDAP libraries.
573 PH/02 Add the verb name to the "unknown ACL verb" error.
575 PH/03 Magnus Holmgren's patch for filter_prepend_home.
577 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
579 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
580 directory not expanded when it should be if an expanded home directory
581 was set for the address (which is overridden by the transport).
583 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
586 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
587 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
588 because it is too late at that time, and has no effect.
590 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
591 security issue with \' (bugzilla #107). I could not use the
592 PQescapeStringConn() function, because it needs a PGconn value as one of
595 PH/08 When testing addresses using -bt, indicate those final addresses that
596 are duplicates that would not cause an additional delivery. At least one
597 person was confused, thinking that -bt output corresponded to deliveries.
598 (Suppressing duplicates isn't a good idea as you lose the information
599 about possibly different redirections that led to the duplicates.)
601 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
602 systems where poll() doesn't work, in particular OS X.
604 PH/10 Added more information to debugging output for retry time not reached.
606 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
607 operations in malware.c.
609 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
612 PH/13 If write_rejectlog was set false when logging was sent to syslog with
613 syslog_duplication set false, log lines that would normally be written
614 both the the main log and to the reject log were not written to syslog at
617 PH/14 In the default configuration, change the use of "message" in ACL warn
618 statements to "add_header".
620 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
621 not followed by a command (e.g. "seen endif").
623 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
624 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
627 PH/17 Added extra conditions to the default value of delay_warning_condition
631 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
632 { match{$h_precedence:}{(?i)bulk|list|junk} } \
633 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
636 The Auto-Submitted: and various List- headers are standardised, whereas I
637 don't think Precedence: ever was.
639 PH/18 Refactored debugging code in route_finduser() to show more information,
640 in particular, the error code if getpwnam() issues one.
642 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
643 This is apparently needed in addition to the PH/07 change above to avoid
644 any possible encoding problems.
646 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
647 but not after initializing Perl.
649 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
650 output them only if debugging. By default they are written stderr,
651 apparently, which is not desirable.
653 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
656 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
659 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
661 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
662 authenticated or an ident call has been made. Suppress the default
663 values for $authenticated_id and $authenticated_sender (but permit -oMai
664 and -oMas) when testing with -bh.
666 PH/24 Re-jigged the order of the tests in the default configuration so that the
667 tests for valid domains and recipients precede the DNS black list and CSA
668 tests, on the grounds that those ones are more expensive.
670 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
671 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
672 This bug exists in every version of Exim that I still have, right back to
675 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
676 However, an attempt to turn on case-sensitivity in a regex key by
677 including (?-i) didn't work because the subject string was already
678 lowercased, and the effects were non-intuitive. It turns out that a
679 one-line patch can be used to allow (?-i) to work as expected.
685 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
686 other effects) broke the use of negated acl sub-conditions.
688 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
691 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
692 "Deny" causes Exim to reject the incoming connection with a 554 error.
693 Unfortunately, if there is a major crisis, such as a disk failure,
694 tcp-wrappers gives "deny", whereas what one would like would be some
695 kind of temporary error. A kludge has been added to help with this.
696 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
697 554 error is used if errno is still zero or contains ENOENT (which occurs
698 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
701 PH/03 Add -lutil to the default FreeBSD LIBS setting.
703 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
704 errors. Otherwise a message that provokes a temporary error (when other
705 messages do not) can cause a whole host to time out.
707 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
708 addresses were routed directly to files or pipes from a redirect router.
709 File deliveries just didn't batch; pipe deliveries might have suffered
712 PH/06 A failure to get a lock for a hints database would erroneously always say
713 "Failed to get write lock", even when it was really a read lock.
715 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
716 of 0600. This has been changed to use the value of the lockfile_mode
717 option (which defaults to 0600).
719 PH/08 Applied small patch from the Sieve maintainer.
721 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
722 folder from quota calculations, a direct delivery into this folder messed
723 up the contents of the maildirsize file. This was because the regex was
724 used only to exclude .Trash (or whatever) when the size of the mailbox
725 was calculated. There was no check that a delivery was happening into an
726 excluded directory. This bug has been fixed by ignoring all quota
727 processing for deliveries into excluded directories.
729 PH/10 Added the maildirfolder_create_regex option to appendfile.
735 PH/01 The code for finding all the local interface addresses on a FreeBSD
736 system running IPv6 was broken. This may well have applied to all BSD
737 systems, as well as to others that have similar system calls. The broken
738 code found IPv4 interfaces correctly, but gave incorrect values for the
739 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
740 that it would not match correctly against @[] and not recognize the IPv6
743 PH/02 The ipliteral router was not recognizing addresses of the form user@
744 [ipv6:....] because it didn't know about the "ipv6:" prefix.
746 PH/03 Added disable_ipv6.
748 PH/04 Changed $reply_address to use the raw form of the headers instead of the
749 decoded form, because it is most often used to construct To: headers
750 lines in autoreplies, and the decoded form may well be syntactically
751 invalid. However, $reply_address has leading white space removed, and all
752 newlines turned into spaces so that the autoreply transport does not
755 PH/05 If group was specified without a user on a router, and no group or user
756 was specified on a transport, the group from the router was ignored.
758 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
759 for visible compile-time settings that can be used to change these
760 numbers, for those that want even more. Backwards compatibility with old
761 spool files has been maintained. However, going back to a previous Exim
762 release will lost any variables that are in spool files.
764 PH/07 Two small changes when running in the test harness: increase delay when
765 passing a TCP/IP connection to a new process, in case the original
766 process has to generate a bounce, and remove special handling of
767 127.0.0.2 (sic), which is no longer necessary.
769 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
770 be the same on different OS.
772 PH/09 Moved a debug statement in filter processing to avoid a race problem when
775 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
776 whether --show-vars was specified or not
778 JJ/02 exipick: Added support for new ACL variable spool format introduced
781 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
782 syntactically invalid From: or Reply-to: line, and a filter used this to
783 generate an autoreply, and therefore failed to obtain an address for the
784 autoreply, Exim could try to deliver to a non-existent relative file
785 name, causing unrelated and misleading errors. What now happens is that
786 it logs this as a hard delivery error, but does not attempt to create a
789 PH/11 The exinext utility has a -C option for testing purposes, but although
790 the given file was scanned by exinext itself; it wasn't being passed on
791 when Exim was called.
793 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
794 an end-of-file indication when reading a command response.
796 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
797 compiled. In many other places in Exim, IPv6 addresses are always
798 recognized, so I have changed this. It also means that IPv4 domain
799 literals of the form [IPV4:n.n.n.n] are now always recognized.
801 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
802 used if the router is not running as root, for example, when verifying at
803 ACL time, or when using -bh. The debugging output from this situation was
804 non-existent - all you got was a failure to exec. I have made two
807 (a) Failures to set uid/gid, the current directory, or a process leader
808 in a subprocess such as that created by queryprogram now generate
809 suitable debugging ouput when -d is set.
811 (b) The queryprogram router detects when it is not running as root,
812 outputs suitable debugging information if -d is set, and then runs
813 the subprocess without attempting to change uid/gid.
815 PH/15 Minor change to Makefile for building test_host (undocumented testing
818 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
819 additional section of a DNS packet that returns MX or SRV records.
820 Instead, it always explicitly searches for A/AAAA records. This avoids
821 major problems that occur when a DNS server includes only records of one
822 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
823 fixed another bug: if SRV records were looked up and the corresponding
824 address records were *not* found in the additional section, the port
825 values from the SRV records were lost.
827 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
828 using the correct key (the original address) when searching the retry
829 rules in order to find which one to use for generating the retry hint.
831 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
832 adding the default one. Similarly, if it contains a Reply-To: header, the
833 errors_reply_to option, if set, is not used.
835 PH/19 When calculating a retry time, Exim used to measure the "time since
836 failure" by looking at the "first failed" field in the retry record. Now
837 it does not use this if it is later than than the arrival time of the
838 message. Instead it uses the arrival time. This makes for better
839 behaviour in cases where some deliveries succeed, thus re-setting the
840 "first failed" field. An example is a quota failure for a huge message
841 when small messages continue to be delivered. Without this change, the
842 "time since failure" will always be short, possible causing more frequent
843 delivery attempts for the huge message than are intended.
844 [Note: This change was subsequently modified - see PH/04 for 4.62.]
846 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
847 $1, $2, $3) because the numerical variables can be reset during some
848 expansion items (e.g. "match"), thereby losing the authentication data.
850 PH/21 Make -bV show the size of off_t variables so that the test suite can
851 decide whether to run tests for quotas > 2G.
853 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
854 mailbox_size, and mailbox_filecount in the appendfile transport. If a
855 filecount value is greater than 2G or if a quota value is greater than 2G
856 on a system where the size of off_t is not greater than 4, a panic error
859 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
860 never match. The debug and -bh output now contains an explicit error
861 message indicating a malformed IPv4 address or mask.
863 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
864 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
867 PH/25 Do not write to syslog when running in the test harness. The only
868 occasion when this arises is a failure to open the main or panic logs
869 (for which there is an explicit test).
871 PH/26 Added the /no_tell option to "control=freeze".
873 PH/27 If a host name lookup failed very early in a connection, for example, if
874 the IP address matched host_lookup and the reverse lookup yielded a name
875 that did not have a forward lookup, an error message of the form "no IP
876 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
877 could be logged. Now it outputs the IP address instead of "NULL".
879 PH/28 An enabling patch from MH: add new function child_open_exim2() which
880 allows the sender and the authenticated sender to be set when
881 submitting a message from within Exim. Since child_open_exim() is
882 documented for local_scan(), the new function should be too.
884 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
885 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
886 results in an empty string is now treated as unset.
888 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
890 PH/31 Added sender_verify_fail logging option.
892 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
893 needed by RFC 822 but not by RFC 2822 was commented out. I have now
894 tidied the source and removed it altogether.
896 PH/33 When a queue run was abandoned because the load average was too high, a
897 log line was always written; now it is written only if the queue_run log
898 selector is set. In addition, the log line for abandonment now contains
899 information about the queue run such as the pid. This is always present
900 in "start" and "stop" lines but was omitted from the "abandon" line.
902 PH/34 Omit spaces between a header name and the colon in the error message that
903 is given when verify = headers_syntax fails (if there are lots of them,
904 the message gets confusing).
906 PH/35 Change the default for dns_check_names_pattern to allow slashes within
907 names, as there are now some PTR records that contain slashes. This check
908 is only to protect against broken name servers that fall over on strange
909 characters, so the fact that it applies to all lookups doesn't matter.
911 PH/36 Now that the new test suite is complete, we can remove some of the
912 special code in Exim that was needed for the old test suite. For example,
913 sorting DNS records because real resolvers return them in an arbitrary
914 order. The new test suite's fake resolver always returns records in the
917 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
918 bounces) except when queue_only is set, to avoid logging races between
919 the different processes.
921 PH/38 Panic-die if .include specifies a non-absolute path.
923 PH/39 A tweak to the "H" retry rule from its user.
925 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
926 a label. They prevented compilation on older perls.
928 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
929 a warning to be raised on newish perls.
931 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
932 on queue. Changes to match documented behaviour of showing count of
933 messages matching specified criteria.
935 PH/40 Changed the default ident timeout from 30s to 5s.
937 PH/41 Added support for the use of login_cap features, on those BSD systems
938 that have them, for controlling the resources used by pipe deliveries.
940 PH/42 The content-scanning code uses fopen() to create files in which to put
941 message data. Previously it was not paying any attention to the mode of
942 the files. Exim runs with umask(0) because the rest of the code creates
943 files with open(), and sets the required mode explicitly. Thus, these
944 files were ending up world-writeable. This was not a big issue, because,
945 being within the spool directory, they were not world-accessible. I have
946 created a function called modefopen, which takes an additional mode
947 argument. It sets umask(777), creates the file, chmods it to the required
948 mode, then resets the umask. All the relevant calls to fopen() in the
949 content scanning code have been changed to use this function.
951 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
952 to 24 hours. This avoids potential overflow problems when processing G
953 and H retry rules. I suspect nobody ever tinkers with this value.
955 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
957 PH/45 When the plaintext authenticator is running as a client, the server's
958 challenges are checked to ensure they are valid base64 strings. By
959 default, the authentication attempt is cancelled if an invalid string is
960 received. Setting client_ignore_invalid_base64 true ignores these errors.
961 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
962 they are received. Thus, the responses can be made to depend on the
963 challenges. If an invalid string is ignored, an empty string is placed in
966 PH/46 Messages that are created by the autoreply transport now contains a
967 References: header, in accordance with RFCs 2822 and 3834.
969 PH/47 Added authenticated_sender_force to the smtp transport.
971 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
973 PH/49 Installed latest patch from the Sieve maintainer.
975 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
976 was also set, the appendfile transport was unnecessarily scanning a
977 directory of message files (e.g. for maildir delivery) to find the count
978 of files (along with the size), even though it did not need this
979 information. It now does the scan only if it needs to find either the
980 size of the count of files.
982 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
984 PH/52 Two bugs concerned with error handling when the smtp transport is
987 (i) Exim was not creating retry information for temporary errors given
988 for individual recipients after the DATA command when the smtp transport
989 was used in LMTP mode. This meant that they could be retried too
990 frequently, and not timed out correctly.
992 (ii) Exim was setting the flag that allows error details to be returned
993 for LMTP errors on RCPT commands, but not for LMTP errors for individual
994 recipients that were returned after the DATA command.
996 PH/53 This is related to PH/52, but is more general: for any failing address,
997 when detailed error information was permitted to be returned to the
998 sender, but the error was temporary, then after the final timeout, only
999 "retry timeout exceeded" was returned. Now it returns the full error as
1000 well as "retry timeout exceeded".
1002 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1003 do this, and (what is worse) MTAs that accept it.
1005 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1006 will now be deprecated.
1008 PH/56 New os.c-cygwin from the Cygwin maintainer.
1010 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1011 formats (previously only available in exim formats via -bpr, -bpru,
1012 and -bpra. Now also available in native and exiqgrep formats)
1014 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1015 with very large, slow to parse queues
1017 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1019 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1021 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1022 responses to authentication challenges, though it was showing the
1023 challenges; (ii) I've removed the CR characters from the debug output for
1026 PH/58 Allow for the insertion of a newline as well as a space when a string
1027 is turned into more than one encoded-word during RFC 2047 encoding. The
1028 Sieve code now uses this.
1030 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1031 data_4xx, lost_connection, tls_required.
1033 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1034 message was being sent as an SMTP response.
1036 PH/61 Add -l and -k options to exicyclog.
1038 PH/62 When verifying, if an address was redirected to one new address, so that
1039 verification continued, and the new address failed or deferred after
1040 having set something in $address_data, the value of $address_data was not
1041 passed back to the ACL. This was different to the case when no
1042 redirection occurred. The value is now passed back in both cases.
1044 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1045 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1046 use login_cap.h, so on its own it isn't the distinguishing feature. The
1047 new name refers directly to the setclassresources() function.
1049 PH/65 Added configuration files for NetBSD3.
1051 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1053 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1054 is preferred over IPv4.
1056 PH/68 The bounce_return_message and bounce_return_body options were not being
1057 honoured for bounces generated during the reception of non-SMTP messages.
1058 In particular, this applied to messages rejected by the ACL. This bug has
1059 been fixed. However, if bounce_return_message is true and bounce_return_
1060 body is false, the headers that are returned for a non-SMTP message
1061 include only those that have been read before the error was detected.
1062 (In the case of an ACL rejection, they have all been read.)
1064 PH/69 The HTML version of the specification is now built in a directory called
1065 spec_html instead of spec.html, because the latter looks like a path with
1066 a MIME-type, and this confuses some software.
1068 PH/70 Catch two compiler warnings in sieve.c.
1070 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1071 function verify_get_ident() calls ip_connect() to connect a socket, but
1072 if the "connect()" function timed out, ip_connect() used to close the
1073 socket. However, verify_get_ident() also closes the socket later, and in
1074 between Exim writes to the log, which may get opened at this point. When
1075 the socket was closed in ip_connect(), the log could get the same file
1076 descriptor number as the socket. This naturally causes chaos. The fix is
1077 not to close the socket in ip_connect(); the socket should be closed by
1078 the function that creates it. There was only one place in the code where
1079 this was missing, in the iplookup router, which I don't think anybody now
1080 uses, but I've fixed it anyway.
1082 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1083 well as to direct DNS lookups. Otherwise the handling of names in host
1084 lists is inconsistent and therefore confusing.
1090 PH/01 Two changes to the default runtime configuration:
1092 (1) Move the checks for relay_from_hosts and authenticated clients from
1093 after to before the (commented out) DNS black list checks.
1095 (2) Add control=submission to the relay_from_hosts and authenticated
1096 clients checks, on the grounds that messages accepted by these
1097 statements are most likely to be submissions.
1099 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1101 (1) Generate an error if the third argument for the ${prvs expansion is
1104 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1107 (3) Reset the variables that are obtained from the first argument of
1108 ${prvscheck and used in the second argument before leaving the code,
1109 because their memory is reclaimed, so using them afterwards may do
1112 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1113 one (it's much easier than Tom thought :-).
1115 (5) Because of (4), we can now allow for the use of $prvscheck_result
1116 inside the third argument.
1118 PH/03 For some reason, the default setting of PATH when running a command from
1119 a pipe transport was just "/usr/bin". I have changed it to
1122 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1123 anything to be listed in the output from -bV.
1125 PH/05 When a filter generated an autoreply, the entire To: header line was
1126 quoted in the delivery log line, like this:
1128 => >A.N.Other <ano@some.domain> <original@ddress> ...
1130 This has been changed so that it extracts the operative address. There
1131 may be more than one such address. If so, they are comma-separated, like
1134 => >ano@some.domain,ona@other.domain <original@ddress> ...
1136 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1137 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1138 not being looked up in the rDNS to get a host name, Exim was showing the
1139 IP address twice in Received: lines, even though the IP addresses were
1140 identical. For example:
1142 Received: from [1.2.3.4] (helo=[1.2.3.4])
1144 However, if the real host name was known, it was omitting the HELO data
1145 if it matched the actual IP address. This has been tidied up so that it
1146 doesn't show the same IP address twice.
1148 PH/07 When both +timestamp and +memory debugging was on, the value given by
1149 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1150 called by the debug printing, thereby overwriting the timestamp buffer.
1151 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1153 PH/08 When the original message was included in an autoreply transport, it
1154 always said "this is a copy of the message, including all the headers",
1155 even if body_only or headers_only was set. It now gives an appropriate
1158 PH/09 Applied a patch from the Sieve maintainer which:
1160 o fixes some comments
1161 o adds the (disabled) notify extension core
1162 o adds some debug output for the result of if/elsif tests
1163 o points to the current vacation draft in the documentation
1164 and documents the missing references header update
1168 o fixes a bug in processing the envelope test (when testing
1169 multiple envelope elements, the last element determinted the
1172 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1173 Electronic Mail") by including:
1175 Auto-submitted: auto-generated
1177 in the messages that it generates (bounce messages and others, such as
1178 warnings). In the case of bounce messages for non-SMTP mesages, there was
1179 also a typo: it was using "Auto_submitted" (underscore instead of
1180 hyphen). Since every message generated by Exim is necessarily in response
1181 to another message, thes have all been changed to:
1183 Auto-Submitted: auto-replied
1185 in accordance with these statements in the RFC:
1187 The auto-replied keyword:
1189 - SHOULD be used on messages sent in direct response to another
1190 message by an automatic process,
1192 - MUST NOT be used on manually-generated messages,
1194 - MAY be used on Delivery Status Notifications (DSNs) and Message
1195 Disposition Notifications (MDNs),
1197 - MUST NOT be used on messages generated by automatic or periodic
1198 processes, except for messages which are automatic responses to
1201 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1202 to the default Received: header definition.
1204 PH/12 Added log selector acl_warn_skipped (default on).
1206 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1207 variables because (a) they are in the wrong storage pool and (b) even if
1208 they were copied, it wouldn't work properly because of the caching.
1210 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1211 checking when decoding. Apparently there are clients that generate
1212 overlong encoded strings. Why am I not surprised?
1214 PH/15 If the first argument of "${if match_address" was not empty, but did not
1215 contain an "@" character, Exim crashed. Now it writes a panic log message
1216 and treats the condition as false.
1218 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1220 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1221 extension "envelope-auth". The code is finished and in agreement with
1222 other implementations, but there is no documentation so far and in fact,
1223 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1224 not changing the active code.
1226 Print executed "if" and "elsif" statements when debugging is used. This
1227 helps a great deal to understand what a filter does.
1229 Document more things not specified clearly in RFC3028. I had all this
1230 sorted out, when out of a sudden new issues came to my mind. Oops."
1232 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1235 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1236 canonical form (as documented). However, after a host name lookup from
1237 the IP address, check_host() was doing a simple string comparison with
1238 addresses acquired from the DNS when checking that the found name did
1239 have the original IP as one of its addresses. Since any found IPv6
1240 addresses are likely to be in abbreviated form, the comparison could
1241 fail. Luckily, there already exists a function for doing the comparison
1242 by converting both addresses to binary, so now that is used instead of
1243 the text comparison.
1245 PH/20 There was another similar case to PH/19, when a complete host name was
1246 given in a host list; looking up its IP address could give an abbreviated
1247 form, whereas the current host's name might or might not be abbreviated.
1248 The same fix has been applied.
1254 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1255 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1258 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1259 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1261 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1263 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1264 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1265 Also, for "Auto-Submitted", treat anything other than "no" as
1266 non-personal, in accordance with RFC 3834. (Previously it treated
1267 anything starting "auto-" as non-personal.)
1269 TF/01 The control=submission/name=... option had a problem with syntax
1270 errors if the name included a slash character. The /name= option
1271 now slurps the rest of the string, so it can include any characters
1272 but it must come last in the list of options (after /sender_retain
1275 PH/05 Some modifications to the interface to the fake nameserver for the new
1283 TK/01 Added the "success_on_redirect" address verification option. See
1284 NewStuff for rationale and an example.
1286 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1288 PH/02 Patch to exigrep to allow it to work on syslog lines.
1290 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1291 fread() to skip over the body file's header line, because in Cygwin the
1292 header line is locked and is inaccessible.
1294 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1295 co-exist for some time) to make it clear that it is the Exim ID that is
1296 referenced, not the Message-ID: header line.
1298 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1299 string_format() function, because snprintf() does not exist on all
1302 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1303 ${stat: expansion item.
1305 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1306 protocol synchronization error", to keep the pedants happy.
1308 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1309 well as for IRIX systems, when gcc is being used. See the host.c source
1312 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1314 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1317 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1318 transport and to the smtp transport in LMTP mode.
1320 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1322 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1323 run a filter in a subprocess. This could lead to confusion in subsequent
1324 lookups in the parent process. There should also be a search_tidyup() at
1325 the end of the subprocess.
1327 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1328 only if the host matched helo_try_verify_hosts, which caused the
1329 verification to occur when the EHLO/HELO command was issued. The ACL just
1330 tested the remembered result. Now, if a previous verification attempt has
1331 not happened, "verify = helo" does it there and then.
1333 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1335 TK/03 Fix log output including CR from clamd.
1337 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1338 exist provoked a memory error which could cause a segfault.
1340 PH/15 Installed PCRE 6.2
1342 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1344 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1345 of the problem. Specifically, suggested +O2 rather than +O1 for the
1348 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1350 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1351 record, the port was not correctly logged when the outgoing_port log
1352 selector was set (it logged the transort's default port).
1354 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1355 fallback_hosts, and "hosts" in the smtp transport.
1357 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1358 host errors such as "Connection refused".
1360 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1361 authentication with radiusclient 0.4.9:
1363 - Error returned from rc_read_config was caught wrongly
1364 - Username/password not passed on to radius server due to wrong length.
1366 The presumption is that some radiusclient API changes for 4.51/PH/17
1367 were not taken care of correctly. The code is still untested by me (my
1368 Linux distribution still has 0.3.2 of radiusclient), but it was
1369 contributed by a Radius user.
1371 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1372 expanding the "port" option of the smtp transport.
1374 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1375 while reading a MIME header. Thanks to Tom Hughes for a patch.
1377 PH/24 Include config.h inside local_scan.h so that configuration settings are
1380 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1381 that in an ACL for RCPT (for example), you can examine exactly what was
1384 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1385 commands, but it was not correctly comparing the address with the actual
1386 client host address. Thus, it would show the EHLO address in Received:
1387 header lines when this was not necessary.
1389 PH/27 Added the % operator to ${eval:}.
1391 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1392 it should be ignoring failures (because with -C, for example, it has lost
1393 privilege). It wasn't ignoring creation failures other than "already
1396 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1399 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1400 because an input line was too long, either on its own, or by virtue of
1401 too many continuations, the temporary file was not being removed, and the
1402 return code was incorrect.
1404 PH/31 Missing "BOOL" in function definition in filtertest.c.
1406 PH/32 Applied Sieve patches from the maintainer.
1408 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1410 PH/33 Added "verify = not_blind".
1412 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1413 Local/Makefile (with some defaults set). These are used in built scripts
1414 such as exicyclog, but they have never been used in the exim_install
1415 script (though there are many overriding facilities there). I have
1416 arranged that the exim_install script now takes note of these two
1419 PH/35 Installed configuration files for Dragonfly.
1421 PH/36 When a locally submitted message by a trusted user did not contain a
1422 From: header, and the sender address was obtained from -f or from an SMTP
1423 MAIL command, and the trusted user did not use -F to supply a sender
1424 name, $originator_name was incorrectly used when constructing a From:
1425 header. Furthermore, $originator_name was used for submission mode
1426 messages from external hosts without From: headers in a similar way,
1427 which is clearly wrong.
1429 PH/37 Added control=suppress_local_fixups.
1431 PH/38 When log_selector = +received_sender was set, and the addition of the
1432 sender made the log line's construction buffer exactly full, or one byte
1433 less than full, an overflow happened when the terminating "\n" was
1436 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1437 when the result of a list match is failure because a DNS lookup failed.
1439 PH/40 RM_COMMAND is now used in the building process.
1441 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1442 the "build-* directories that it finds.
1444 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1445 address in a domain literal was a prefix of an interface address.
1447 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1448 when verifying a sender address, unless rewrite_headers is false.
1450 PH/44 Wrote a long comment about why errors_to addresses are verified as
1451 recipients, not senders.
1453 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1454 the ratelimit ACL was added.
1456 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1458 PH/46 Added extra information about PostgreSQL errors to the error string.
1460 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1461 avoiding the need to install special zones in a real server. This is
1462 backwards compatible; if it can't find the fake resolver, it drops back.
1463 Thus, both old and new test suites can be run.
1465 TF/02 Added util/ratelimit.pl
1467 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1468 clock is set back in time.
1470 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1471 Candler <B.Candler@pobox.com>.
1473 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1474 OK for addresses that are the result of redirections.
1476 PH/48 A number of further additions for the benefit of the new test suite,
1477 including a fake gethostbyname() that interfaces to the fake DNS resolver
1480 TF/06 The fix for widen_domains has also been applied to qualify_single and
1481 search_parents which are the other dnslookup options that can cause
1484 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1487 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1489 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1490 service" if it encountered an MX record with an empty target hostname.
1491 The message is now "an MX or SRV record indicated no SMTP service".
1493 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1494 if the DNS of the sending site is misconfigured. This is quite a
1495 common situation. This change restores the behaviour of treating a
1496 helo verification defer as a failure.
1498 PH/51 If self=fail was set on a router, the bounce message did not include the
1499 actual error message.
1505 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1507 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1508 command itself ended in error, the underlying message about the transport
1509 filter timeout was being overwritten with the pipe command error. Now the
1510 underlying error message should be appended to the second error message.
1512 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1514 PH/02 Reduce the amount of output that "make" produces by default. Full output
1515 can still be requested.
1517 PH/03 The warning log line about a condition test deferring for a "warn" verb
1518 was being output only once per connection, rather than after each
1519 occurrence (because it was using the same function as for successful
1520 "warn" verbs). This seems wrong, so I have changed it.
1522 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1523 it should not have, which might have caused a crash in the right
1524 circumstances, but probably never did.
1526 PH/04 Installed a modified version of Tony Finch's patch to make submission
1527 mode fix the return path as well as the Sender: header line, and to
1528 add a /name= option so that you can make the user's friendly name appear
1531 TF/03 Added the control = fakedefer ACL modifier.
1533 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1534 Mark Lowes for thorough testing.
1536 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1538 TK/03 Merged latest SRS patch from Miles Wilton.
1540 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1541 with the definition in sysexits.h (which is #included earlier).
1542 Fortunately, Exim does not actually use EX_OK. The code used to try to
1543 preserve the sysexits.h value, by assumimg that macro definitions were
1544 scanned for macro replacements. I have been disabused of this notion,
1545 so now the code just undefines EX_OK before #including unistd.h.
1547 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1548 in the smtp transport. When a block could not be written in a single
1549 write() function, the timeout was being re-applied to each part-write.
1550 This seems wrong - if the receiver was accepting one byte at a time it
1551 would take for ever. The timeout is now adjusted when this happens. It
1552 doesn't have to be particularly precise.
1554 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1555 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1557 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1558 without a domain if the check to <postmaster@domain> fails.
1560 SC/01 Eximstats: added -xls and the ability to specify output files
1561 (patch written by Frank Heydlauf).
1563 SC/02 Eximstats: use FileHandles for outputing results.
1565 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1567 SC/04 Eximstats: fixed display of large numbers with -nvr option
1569 SC/05 Eximstats: fixed merging of reports with empty tables.
1571 SC/06 Eximstats: added the -include_original_destination flag
1573 SC/07 Eximstats: removed tabs and trailing whitespace.
1575 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1577 TK/06 MBOX spool code: Add real "From " MBOX separator line
1578 so the .eml file is really in mbox format (even though
1579 most programs do not really care). Patch from Alex Miller.
1581 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1582 The latter is generated from $received_to and is only set if the
1583 message has one envelope recipient. SA can use these headers,
1584 obviously out-of-the-box. Patch from Alex Miller.
1586 PH/08 The ${def test on a variable was returning false if the variable's
1587 value was "0", contrary to what the specification has always said!
1588 The result should be true unless the variable is empty.
1590 PH/09 The syntax error of a character other than { following "${if
1591 def:variable_name" (after optional whitespace) was not being diagnosed.
1592 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1593 accidental colon was present, for example, could give incorrect results.
1595 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1596 result is used (not including appendfile, where other changes are about
1599 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1600 This involved changing a lot of size variables from int to off_t. It
1601 should work with maildirs and everything.
1603 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1604 spamd dying while we are connected to it.
1606 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1609 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1610 function for BDB 4.3.
1612 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1614 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1615 the sources for PCRE have been re-arranged and more files are now
1618 PH/15 The code I had for printing potentially long long variables in PH/11
1619 above was not the best (it lost precision). The length of off_t variables
1620 is now inspected at build time, and an appropriate printing format (%ld
1621 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1622 to be "long long int" or "long int". This is needed for the internal
1623 formatting function string_vformat().
1625 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1626 the configuration file to be ":syslog", then the script "guesses" where
1627 the logs files are, rather than using the compiled in default. In our
1628 case the guess is not the same as the compiled default, so the script
1629 suddenly stopped working when I started to use syslog. The patch checks
1630 to see if log_file_path is "". If so, it attempts to read it from exim
1631 with no configuration file to get the compiled in version, before it
1632 falls back to the previous guessing code."
1634 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1635 implementing BATV in an Exim configuration. See NewStuff for the gory
1638 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1639 Makefile that are specific to HP-UX.
1641 PH/18 If the "use_postmaster" option was set for a recipient callout together
1642 with the "random" option, the postmaster address was used as the MAIL
1643 FROM address for the random test, but not for the subsequent recipient
1644 test. It is now used for both.
1646 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1647 patch removes a few documentation additions to RFC 3028, because the
1648 latest draft now contains them. It adds the new en;ascii-case comparator
1649 and a new error check for 8bit text in MIME parts. Comparator and
1650 require names are now matched exactly. I enabled the subaddress
1651 extension, but it is not well tested yet (read: it works for me)."
1653 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1654 rework some of the code of TK/09 above to avoid the hardwired use of
1655 "%lld" and "long long". Replaced the call to snprintf() with a call to
1658 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1659 records point to non-existent hosts", "retry timeout exceeded", and
1660 "retry time not reached for any host after a long failure period".
1662 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1663 experimental DomainKeys support:
1665 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1666 (2) On an error such as an illegally used "control", the wrong name for
1667 the control was given.
1669 These problems did NOT occur unless DomainKeys support was compiled.
1671 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1673 PH/24 Added ${if match_ip condition.
1675 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1676 databases so that it will be absolutely obvious if a crash occurs in the
1677 DB library. This is a regular occurrence (often caused by mis-matched
1680 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1681 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1682 picked up on a user's system that detects such things. There doesn't seem
1683 to be a gcc warning option for this - only an attribute that has to be
1684 put on the function's prototype. It seems that in Fedora Core 4 they have
1685 set this on a number of new functions. No doubt there will be more in due
1688 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1689 specify a transport. However, if an address that was verified by such a
1690 router was the subject of a callout, Exim crashed because it tried to
1691 read the rcpt_include_affixes from the non-existent transport. Now it
1692 just assumes that the setting of that option is false. This bug was
1693 introduced by 4.51/PH/31.
1695 PH/28 Changed -d+all to exclude +memory, because that information is very
1696 rarely of interest, but it makes the output a lot bigger. People tend to
1697 do -d+all out of habit.
1699 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1700 code in os-type was giving problems when libc.so lives in lib64, like on
1703 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1704 aren't the modern standard, and it seems that some systems' include files
1705 don't always have them. Exim was already checking for some of the newer
1706 ones like T_AAAA, and defining it itself. I've added checks for all the
1707 record types that Exim uses.
1709 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1710 not automatically generating a new one, as it is supposed to. This
1711 prevented TLS from working. If the file did exist, but contained invalid
1712 data, a new version was generated, as expected. It was only the case of a
1713 non-existent file that was broken.
1715 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1716 with a change in libdomainkeys > 0.64.
1718 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1719 from DNS. If the selector record carries the flag, it now has
1720 precedence over the domain-wide flag.
1722 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1724 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1725 the use of a port name that isn't defined in /etc/services) occurred, the
1726 message was deferred as in a normal delivery, and thus remained on the
1727 spool, instead of being failed because of the mua_wrapper setting. This
1728 is now fixed, and I tidied up some of the mua_wrapper messages at the
1731 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1732 the messages in a hash of arrays rather than using individual hashes.
1733 This is a bit cleaner and results in dramatic memory savings, albeit
1734 at a slight CPU cost.
1736 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1737 as requested by Marc Sherman.
1739 SC/10 Eximstats: added histograms for user specified patterns as requested
1742 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1744 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1745 fopen() in the content-scanning modules that did not already have it.
1751 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1752 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1754 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1756 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1758 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1759 to test Sieve filters that use "vacation".
1761 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1762 that changes the way the GnuTLS parameters are stored in the cache file.
1763 The new format can be generated externally. For backward compatibility,
1764 if the data in the cache doesn't make sense, Exim assumes it has read an
1765 old-format file, and it generates new data and writes a new file. This
1766 means that you can't go back to an older release without removing the
1769 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1770 work if there are any delivery delays because "one_time" forces the
1771 parent to be marked "delivered", so its unseen clone is never tried
1772 again. For this reason, Exim now forbids the simultaneous setting of
1775 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1776 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1777 entry if you want to know the details. The fix, however, neglected to
1778 consider the case when local delivery batching is involved. The test for
1779 "previously delivered" was not happening when checking to see if an
1780 address could be batched with a previous (undelivered) one; under
1781 certain circumstances this could lead to multiple deliveries to the same
1784 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1785 in its include files, and this causes problems building Exim.
1787 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1788 header_syntax) but Exim was just ignoring anything given after a slash.
1789 In particular, this caused confusion with an attempt to use "verify =
1790 reverse_host_lookup/defer_ok". An error is now given when options are
1791 supplied for verify items that do not have them. (Maybe reverse_host_
1792 lookup should have a defer_ok option, but that's a different point.)
1794 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1795 defined by RFC 821) to 2048, because there were problems with some AUTH
1796 commands, and RFC 1869 says the size should be increased for extended
1797 SMTP commands that take arguments.
1799 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1802 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1803 "unknown" error; now it says that the functionality isn't in the binary.
1805 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1806 an address' error message when a string expansion fails (syntax or
1807 whatever). Otherwise the password may appear in the log. Following change
1808 PH/42 below, there is no longer a chance of it appearing in a bounce
1811 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1813 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1814 of its addresses was ever tried. (Bugzilla bug #2.)
1816 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1817 the result incorrectly in the debug output. (It correctly added a newline
1818 to what was transported.)
1820 TF/01 Added $received_time.
1822 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1823 commented out examples of how to interface to a virus scanner and to
1824 SpamAssassin. Also added commented examples of av_scanner and
1825 spamd_address settings.
1827 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1828 and controls are allowed in which ACLs. There were a couple of minor
1829 errors. Some of the entries in the conditions table (which is a table of
1830 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1831 negation of where the condition IS allowed.
1833 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1835 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1836 header file does not have a version number, so I've had to invent a new
1837 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1838 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1839 radiusclient), but it was contributed by a Radius user.
1841 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1842 files or directories, for OpenSSL.
1844 PH/19 When an Exim process that is running as root has to create an Exim log
1845 file, it does so in a subprocess that runs as exim:exim so as to get the
1846 ownership right at creation (otherwise, other Exim processes might see
1847 the file with the wrong ownership). There was no test for failure of this
1848 fork() call, which would lead to the process getting stuck as it waited
1849 for a non-existent subprocess. Forks do occasionally fail when resources
1850 run out. I reviewed all the other calls to fork(); they all seem to check
1853 PH/20 When checking for unexpected SMTP input at connect time (before writing
1854 the banner), Exim was not dealing correctly with a non-positive return
1855 from the read() function. If the client had disconnected by this time,
1856 the result was a log entry for a synchronization error with an empty
1857 string after "input=" when read() returned zero. If read() returned -1
1858 (an event I could not check), uninitialized data bytes were printed.
1859 There were reports of junk text (parts of files, etc) appearing after
1862 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1864 PH/22 Added support for macro redefinition, and (re)definition in between
1865 driver and ACL definitions.
1867 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1868 forgetting to use the resulting value; it was using the unexpanded value.
1870 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1871 hadn't been configured. The fix is from Juergen Kreileder, who
1872 understands it better than I do:
1874 "Here's what I see happening with three configured cyrus_sasl
1875 authenticators configured (plain, login, cram-md5):
1877 On startup auth_cyrus_sasl_init() gets called for each of these.
1878 This means three calls to sasl_listmech() without a specified mech_list.
1879 => SASL tests which mechs of all available mechs actually work
1880 => three warnings about OTP not working
1881 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1883 With the patch, sasl_listmech() also gets called three times. But now
1884 SASL's mech_list option is set to the server_mech specified in the the
1885 authenticator. Or in other words, the answer from sasl_listmech()
1886 gets limited to just the mech you're testing for (which is different
1888 => the return list contains just 'plain' or 'login', 'cram-md5' or
1889 nothing depending on the value of ob->server_mech.
1891 I've just tested the patch: Authentication still works fine,
1892 unavailable mechs specified in the exim configuration are still
1893 caught, and the auth.log warnings about OTP are gone."
1895 PH/25 When debugging is enabled, the contents of the command line are added
1896 to the debugging output, even when log_selector=+arguments is not
1899 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1900 answer is "GNU", and only if the return is "GNU/something" is the answer
1903 PH/27 $acl_verify_message is now set immediately after the failure of a
1904 verification in an ACL, and so is available in subsequent modifiers. In
1905 particular, the message can be preserved by coding like this:
1907 warn !verify = sender
1908 set acl_m0 = $acl_verify_message
1910 Previously, $acl_verify_message was set only while expanding "message"
1911 and "log_message" when a very denied access.
1913 PH/28 Modified OS/os.c-Linux with
1915 -#ifndef OS_LOAD_AVERAGE
1916 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1918 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1919 nomenclature these days.)
1921 PH/29 Installed patch from the Sieve maintainer that adds the options
1922 sieve_useraddress and sieve_subaddress to the redirect router.
1924 PH/30 In these circumstances:
1925 . Two addresses routed to the same list of hosts;
1926 . First host does not offer TLS;
1927 . First host accepts first address;
1928 . First host gives temporary error to second address;
1929 . Second host offers TLS and a TLS session is established;
1930 . Second host accepts second address.
1931 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1932 and peerdn, if requested) that were in fact used only for the second
1935 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1936 attention to any local part prefix or suffix that was matched by the
1937 router that accepted the address. It now behaves in the same way as it
1938 does for delivery: the affixes are removed from the local part unless
1939 rcpt_include_affixes is set on the transport.
1941 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1942 timeout during the DATA phase of an incoming message.
1944 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1945 applied a patch sanctioned by the Sieve maintainer.
1947 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1948 the uid or gid is negative. A case of a negative gid caused this to be
1949 noticed. The fix allows for either to be negative.
1951 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1952 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1955 PH/36 The change PH/12 above was broken. Fixed it.
1957 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1958 the grounds that routing the same address twice would always produce the
1959 same answer. This might have been true once, but it is certainly no
1960 longer true now. Routing a child address may depend on the previous
1961 routing that produced that child. Some complicated redirection strategies
1962 went wrong when messages had multiple recipients, and made Exim's
1963 behaviour dependent on the order in which the addresses were given.
1965 I have moved the duplicate checking until after the routing is complete.
1966 Exim scans the addresses that are assigned to local and remote
1967 transports, and removes any duplicates. This means that more work will be
1968 done, as duplicates will always all be routed, but duplicates are
1969 presumably rare, so I don't expect this is of any significance.
1971 For deliveries to pipes, files, and autoreplies, the duplicate checking
1972 still happens during the routing process, since they are not going to be
1975 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1976 It corrects a timeout issue with spamd. This is Ian's comment: "The
1977 background is that sometimes spamd either never reads data from a
1978 connection it has accepted, or it never writes response data. The exiscan
1979 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1980 blindly assumes that writes won't block so it may never time out."
1982 PH/39 Allow G after quota size as well as K and M.
1984 PH/40 The value set for $authenticated_id in an authenticator may not contain
1985 binary zeroes or newlines because the value is written to log lines and
1986 to spool files. There was no check on this. Now the value is run through
1987 the string_printing() function so that such characters are converted to
1988 printable escape sequences.
1990 PH/41 $message_linecount is a new variable that contains the total number of
1991 lines in the message. Compare $body_linecount, which is the count for the
1994 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1995 bounce and delay warning messages, except in certain special cases, which
1998 (a) An SMTP error message from a remote host;
1999 (b) A message specified in a :fail: redirection;
2000 (c) A message specified in a "fail" command in a system filter;
2001 (d) A message specified in a FAIL return from the queryprogram router;
2002 (e) A message specified by the cannot_route_message router option.
2004 In these cases only, Exim does include the error details in bounce and
2005 warning messages. There are also a few cases where bland messages such
2006 as "unrouteable address" or "local delivery error" are given.
2008 PH/43 $value is now also set for the "else" part of a ${run expansion.
2010 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2011 being worked on, but at least Exim now implements the latest version to
2014 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2015 process to complete was treated as a delivery failure, a timeout while
2016 writing the message to the pipe was logged, but erroneously treated as a
2017 successful delivery. Such timeouts include transport filter timeouts. For
2018 consistency with the overall process timeout, these timeouts are now
2019 treated as errors, giving rise to delivery failures by default. However,
2020 there is now a new Boolean option for the pipe transport called
2021 timeout_defer, which, if set TRUE, converts the failures into defers for
2022 both kinds of timeout. A transport filter timeout is now identified in
2025 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2026 systems where "make" and "gmake" are different, calling "gmake" at top
2027 level broke things. I've arranged for the value of $(MAKE) to be passed
2028 from the Makefile to this script so that it can call the same version of
2032 A note about Exim versions 4.44 and 4.50
2033 ----------------------------------------
2035 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2036 changes of various kinds. As a consequence, a big documentation update was
2037 needed. This delayed the release for rather longer than seemed good, especially
2038 in the light of a couple of (minor) security issues. Therefore, the changes
2039 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2040 release. So 4.44 and 4.50 are in effect two different branches that both start
2043 I have left the 4.50 change log unchanged; it contains all the changes since
2044 4.43. The change log for 4.44 is below; many of its items are identical to
2045 those for 4.50. This seems to be the most sensible way to preserve the
2046 historical information.
2052 1. Minor wording change to the doc/README.SIEVE file.
2054 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2055 computation of the current number of files was incorrect.
2057 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2058 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2061 4. Give more explanation in the error message when the command for a transport
2062 filter fails to execute.
2064 5. There are several places where Exim runs a non-Exim command in a
2065 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2066 was being done only for the command run by the queryprogram router. It is
2067 now done for all such subprocesses. The other cases are: ${run, transport
2068 filters, and the commands run by the lmtp and pipe transports.
2070 6. Added CONFIGURE_GROUP build-time option.
2072 7. Some older OS have a limit of 256 on the maximum number of file
2073 descriptors. Exim was using setrlimit() to set 1000 as a large value
2074 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2075 systems. I've change it so that if it can't get 1000, it tries for 256.
2077 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2078 was an oversight, and furthermore, ever since the addition of extra
2079 controls (e.g. 4.43/32), the checks on when to allow different forms of
2080 "control" were broken. There should now be diagnostics for all cases when a
2081 control that does not make sense is encountered.
2083 9. Added the /retain_sender option to "control=submission".
2085 10. $recipients is now available in the predata ACL (oversight).
2087 11. Tidy the search cache before the fork to do a delivery from a message
2088 received from the command line. Otherwise the child will trigger a lookup
2089 failure and thereby defer the delivery if it tries to use (for example) a
2090 cached ldap connection that the parent has called unbind on.
2092 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2093 of $address_data from the recipient verification was clobbered by the
2094 sender verification.
2096 13. The value of address_data from a sender verification is now available in
2097 $sender_address_data in subsequent conditions in the ACL statement.
2099 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2101 15. Added a new option "connect=<time>" to callout options, to set a different
2104 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2105 was its contents. (It was OK if the option was not defined at all.)
2107 17. A "Completed" log line is now written for messages that are removed from
2108 the spool by the -Mrm option.
2110 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2111 information about exactly what failed.
2113 19. Added -dd to debug only the daemon process.
2115 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2116 handles timeouts, both on the server side and network timeouts. Renamed the
2117 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2119 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2120 It is now set to "smtps".
2122 22. $host_address is now set to the target address during the checking of
2123 ignore_target_hosts.
2125 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2126 being passed; this would have caused $sender_host_name to have been used if
2127 matching the list had actually called for a host name (not very likely,
2128 since this list is usually IP addresses). A host name is now passed as
2131 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2132 code that specifies a non-restarting handler (typically sigaction() in
2133 modern systems) in an attempt to fix a rare and obscure crash bug.
2135 25. Narrowed the window for a race in the daemon that could cause it to ignore
2136 SIGCHLD signals. This is not a major problem, because they are used only to
2137 wake it up if nothing else does.
2139 26. A malformed maildirsize file could cause Exim to calculate negative values
2140 for the mailbox size or file count. Odd effects could occur as a result.
2141 The maildirsize information is now recalculated if the size or filecount
2144 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2145 support for a long time. Removed HAVE_SYS_VFS_H.
2147 28. Installed the latest version of exipick from John Jetmore.
2149 29. In an address list, if the pattern was not a regular expression, an empty
2150 subject address (from a bounce message) matched only if the pattern was an
2151 empty string. Non-empty patterns were not even tested. This was the wrong
2152 because it is perfectly reasonable to use an empty address as part of a
2153 database query. An empty address is now tested by patterns that are
2154 lookups. However, all the other forms of pattern expect the subject to
2155 contain a local part and a domain, and therefore, for them, an empty
2156 address still always fails if the pattern is not itself empty.
2158 30. Exim went into a mad DNS loop when attempting to do a callout where the
2159 host was specified on an smtp transport, and looking it up yielded more
2160 than one IP address.
2162 31. Re-factored the code for checking spool and log partition space into a
2163 function that finds that data and another that does the check. The former
2164 is then used to implement four new variables: $spool_space, $log_space,
2165 $spool_inodes, and $log_inodes.
2167 32. The RFC2047 encoding function was originally intended for short strings
2168 such as real names; it was not keeping to the 75-character limit for
2169 encoded words that the RFC imposes. It now respects the limit, and
2170 generates multiple encoded words if necessary. To be on the safe side, I
2171 have increased the buffer size for the ${rfc2047: expansion operator from
2174 33. It is now permitted to omit both strings after an "if" condition; if the
2175 condition is true, the result is "true". As before, when the second string
2176 is omitted, a false condition yields an empty string. This makes it less
2177 cumbersome to write custom ACL and router conditions.
2179 34. Failure to deliver a bounce message always caused it to be frozen, even if
2180 there was an errors_to setting on the router. The errors_to setting is now
2183 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2184 canonical form (fully expanded) before being placed in
2185 $sender_host_address.
2187 36. The table in the code that translates DNS record types into text (T_A to
2188 "A" for instance) was missing entries for NS and CNAME. It is just possible
2189 that this could have caused confusion if both these types were looked up
2190 for the same domain, because the text type is used as part of Exim's
2191 per-process caching. But the chance of anyone hitting this buglet seems
2194 37. The dnsdb lookup has been extended in a number of ways.
2196 (1) There is a new type, "zns", which walks up the domain tree until it
2197 finds some nameserver records. It should be used with care.
2199 (2) There is a new type, "mxh", which is like "mx" except that it returns
2200 just the host names, not the priorities.
2202 (3) It is now possible to give a list of domains (or IP addresses) to be
2203 looked up. The behaviour when one of the lookups defers can be
2204 controlled by a keyword.
2206 (4) It is now possible to specify the separator character for use when
2207 multiple records are returned.
2209 38. The dnslists ACL condition has been extended: it is now possible to supply
2210 a list of IP addresses and/or domains to be looked up in a particular DNS
2213 39. Added log_selector=+queue_time_overall.
2215 40. When running the queue in the test harness, wait just a tad after forking a
2216 delivery process, to get repeatability of debugging output.
2218 41. Include certificate and key file names in error message when GnuTLS fails
2219 to set them up, because the GnuTLS error message doesn't include the name
2220 of the failing file when there is a problem reading it.
2222 42. Allow both -bf and -bF in the same test run.
2224 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2226 44. The "Exiscan patch" is now merged into the mainline Exim source.
2228 45. Sometimes the final signoff response after QUIT could fail to get
2229 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2230 < 0 before doing a fflush(). This bug looks as though it goes back to the
2231 introduction of TLS in release 3.20, but "sometimes" must have been rare
2232 because the tests only now provoked it.
2234 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2235 (this can affect the format of dates).
2237 47. exim_tidydb, when checking for the continued existence of a message for
2238 which it has found a message-specific retry record, was not finding
2239 messages that were in split spool directories. Consequently, it was
2240 deleting retry records that should have stayed in existence.
2242 48. Steve fixed some bugs in eximstats.
2244 49. The SPA authentication driver was not abandoning authentication and moving
2245 on to the next authenticator when an expansion was forced to fail,
2246 contradicting the general specification for all authenticators. Instead it
2247 was generating a temporary error. It now behaves as specified.
2249 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2250 (the order specifies the preference for clients). The order is now AES256,
2251 AES128, 3DES, ARCFOUR128.
2253 51. Small patch to Sieve code - explicitly set From: when generating an
2256 52. Exim crashed if a remote delivery caused a very long error message to be
2257 recorded - for instance if somebody sent an entire SpamAssassin report back
2258 as a large number of 550 error lines. This bug was coincidentally fixed by
2259 increasing the size of one of Exim's internal buffers (big_buffer) that
2260 happened as part of the Exiscan merge. However, to be on the safe side, I
2261 have made the code more robust (and fixed the comments that describe what
2264 53. Now that there can be additional text after "Completed" in log lines (if
2265 the queue_time_overall log selector is set), a one-byte patch to exigrep
2266 was needed to allow it to recognize "Completed" as not the last thing in
2269 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2270 patch that reportedly fixes this has been added. I am not expert enough to
2271 create a test for it. This is what the patch creator wrote:
2273 "I found a little strange behaviour of ldap code when working with
2274 Windows 2003 AD Domain, where users was placed in more than one
2275 Organization Units. When I tried to give exim partial DN, the exit code
2276 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2277 But simultaneously result of request was absolutely normal ldap result,
2278 so I produce this patch..."
2280 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2281 REFERENCE, so I have modified the code to exclude the patch when that macro
2284 55. Some experimental protocols are using DNS PTR records for new purposes. The
2285 keys for these records are domain names, not reversed IP addresses. The
2286 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2287 leaves it alone. Component reversal etc. now happens only for IP addresses.
2290 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2292 57. Double the size of the debug message buffer (to 2048) so that more of very
2293 long debug lines gets shown.
2295 58. The exicyclog utility now does better if the number of log files to keep
2296 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2298 59. Two changes related to the smtp_active_hostname option:
2300 (1) $smtp_active_hostname is now available as a variable.
2301 (2) The default for smtp_banner uses $smtp_active_hostname instead
2302 of $primary_hostname.
2304 60. The host_aton() function is supposed to be passed a string that is known
2305 to be a valid IP address. However, in the case of IPv6 addresses, it was
2306 not checking this. This is a hostage to fortune. Exim now panics and dies
2307 if the condition is not met. A case was found where this could be provoked
2308 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2309 components; fortuitously, this particular loophole had already been fixed
2310 by change 4.50/55 above.
2312 If there are any other similar loopholes, the new check in host_aton()
2313 itself should stop them being exploited. The report I received stated that
2314 data on the command line could provoke the exploit when Exim was running as
2315 exim, but did not say which command line option was involved. All I could
2316 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2317 running as the user.
2320 61. There was a buffer overflow vulnerability in the SPA authentication code
2321 (which came originally from the Samba project). I have added a test to the
2322 spa_base64_to_bits() function which I hope fixes it.
2325 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2326 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2328 63. The daemon start-up calls getloadavg() while still root for those OS that
2329 need the first call to be done as root, but it missed one case: when
2330 deliver_queue_load_max is set with deliver_drop_privilege. This is
2331 necessary for the benefit of the queue runner, because there is no re-exec
2332 when deliver_drop_privilege is set.
2334 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2335 This has been fixed.
2337 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2338 was in use, was not putting the data itself into the right store pool;
2339 consequently, it could be overwritten for a subsequent message in the same
2340 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2343 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2345 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2346 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2347 the function were treating the return as a boolean value, which happened to
2348 work because 0=false and not-0=true, but is not correct code.
2350 68. The host_aton() function was not handling scoped IPv6 addresses (those
2351 with, for example, "%eth0" on the end) correctly.
2353 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2354 negated items (that is, ~something) in unsigned ints. Some compilers
2355 apparently mutter when there is no cast.
2357 70. If an address verification called from an ACL failed, and did not produce a
2358 user-specific message (i.e. there was only a "system" message), nothing was
2359 put in $acl_verify_message. In this situation, it now puts the system
2362 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2363 session; change 4.31/43 added the unwanted input to the log line - except
2364 that it did not do this in the start of session case. It now does.
2366 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2367 This is wrong and can cause the other end to generate a synchronization
2368 error if it is another Exim or anything else that does the synchronization
2369 check. A QUIT command is no longer sent after a timeout.
2371 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2372 during host lookups.
2374 74. The defer_ok option of callout verification was not working if it was used
2375 when verifying addresses in header lines, that is, for this case:
2377 verify = header_sender/callout=defer_ok
2379 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2380 those file descriptors could be used for SMTP connections. If anything
2381 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2382 could be sent to the SMTP client, causing chaos. The daemon now opens
2383 stdin, stdout, and stderr to /dev/null when it puts itself into the
2386 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2387 log by default. The user can override this with suitable Perl magic.
2389 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2390 the log message when discard triggers, was not working for the DATA ACL or
2391 for the non-SMTP ACL.
2393 78. Error message wording change in sieve.c.
2395 79. If smtp_accept_max_per_host was set, the number of connections could be
2396 restricted to fewer than expected, because the daemon was trying to set up
2397 a new connection before checking whether the processes handling previous
2398 connections had finished. The check for completed processes is now done
2399 earlier. On busy systems, this bug wouldn't be noticed because something
2400 else would have woken the daemon, and it would have reaped the completed
2403 80. If a message was submitted locally by a user whose login name contained one
2404 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2405 It caused a spool format error. I have fixed the spool reading code. A
2406 related problem was that the "from" clause in the Received: line became
2407 illegal because of the space(s). It is now covered by ${quote_local_part.
2409 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2412 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2414 83. Patch from Sieve maintainer to fix unterminated string problem in
2415 "vacation" handling.
2417 84. Some minor changes to the Linux configuration files to help with other
2418 OS variants using glibc.
2420 85. One more patch for Sieve to update vacation handling to latest spec.
2423 ----------------------------------------------------
2424 See the note above about the 4.44 and 4.50 releases.
2425 ----------------------------------------------------
2431 1. Change 4.43/35 introduced a bug that caused file counts to be
2432 incorrectly computed when quota_filecount was set in an appendfile
2435 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2436 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2439 3. Give more explanation in the error message when the command for a transport
2440 filter fails to execute.
2442 4. There are several places where Exim runs a non-Exim command in a
2443 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2444 was being done only for the command run by the queryprogram router. It is
2445 now done for all such subprocesses. The other cases are: ${run, transport
2446 filters, and the commands run by the lmtp and pipe transports.
2448 5. Some older OS have a limit of 256 on the maximum number of file
2449 descriptors. Exim was using setrlimit() to set 1000 as a large value
2450 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2451 systems. I've change it so that if it can't get 1000, it tries for 256.
2453 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2454 was an oversight, and furthermore, ever since the addition of extra
2455 controls (e.g. 4.43/32), the checks on when to allow different forms of
2456 "control" were broken. There should now be diagnostics for all cases when a
2457 control that does not make sense is encountered.
2459 7. $recipients is now available in the predata ACL (oversight).
2461 8. Tidy the search cache before the fork to do a delivery from a message
2462 received from the command line. Otherwise the child will trigger a lookup
2463 failure and thereby defer the delivery if it tries to use (for example) a
2464 cached ldap connection that the parent has called unbind on.
2466 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2467 of $address_data from the recipient verification was clobbered by the
2468 sender verification.
2470 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2471 was its contents. (It was OK if the option was not defined at all.)
2473 11. A "Completed" log line is now written for messages that are removed from
2474 the spool by the -Mrm option.
2476 12. $host_address is now set to the target address during the checking of
2477 ignore_target_hosts.
2479 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2480 being passed; this would have caused $sender_host_name to have been used if
2481 matching the list had actually called for a host name (not very likely,
2482 since this list is usually IP addresses). A host name is now passed as
2485 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2486 code that specifies a non-restarting handler (typically sigaction() in
2487 modern systems) in an attempt to fix a rare and obscure crash bug.
2489 15. Narrowed the window for a race in the daemon that could cause it to ignore
2490 SIGCHLD signals. This is not a major problem, because they are used only to
2491 wake it up if nothing else does.
2493 16. A malformed maildirsize file could cause Exim to calculate negative values
2494 for the mailbox size or file count. Odd effects could occur as a result.
2495 The maildirsize information is now recalculated if the size or filecount
2498 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2499 support for a long time. Removed HAVE_SYS_VFS_H.
2501 18. Updated exipick to current release from John Jetmore.
2503 19. Allow an empty sender to be matched against a lookup in an address list.
2504 Previously the only cases considered were a regular expression, or an
2507 20. Exim went into a mad DNS lookup loop when doing a callout where the
2508 host was specified on the transport, if the DNS lookup yielded more than
2511 21. The RFC2047 encoding function was originally intended for short strings
2512 such as real names; it was not keeping to the 75-character limit for
2513 encoded words that the RFC imposes. It now respects the limit, and
2514 generates multiple encoded words if necessary. To be on the safe side, I
2515 have increased the buffer size for the ${rfc2047: expansion operator from
2518 22. Failure to deliver a bounce message always caused it to be frozen, even if
2519 there was an errors_to setting on the router. The errors_to setting is now
2522 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2523 canonical form (fully expanded) before being placed in
2524 $sender_host_address.
2526 24. Updated eximstats to version 1.33
2528 25. Include certificate and key file names in error message when GnuTLS fails
2529 to set them up, because the GnuTLS error message doesn't include the name
2530 of the failing file when there is a problem reading it.
2532 26. Expand error message when OpenSSL has problems setting up cert/key files.
2535 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2536 (this can affect the format of dates).
2538 28. exim_tidydb, when checking for the continued existence of a message for
2539 which it has found a message-specific retry record, was not finding
2540 messages that were in split spool directories. Consequently, it was
2541 deleting retry records that should have stayed in existence.
2543 29. eximstats updated to version 1.35
2544 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2545 1.35 - bugfix such that pie charts by volume are generated correctly
2547 30. The SPA authentication driver was not abandoning authentication and moving
2548 on to the next authenticator when an expansion was forced to fail,
2549 contradicting the general specification for all authenticators. Instead it
2550 was generating a temporary error. It now behaves as specified.
2552 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2553 (the order specifies the preference for clients). The order is now AES256,
2554 AES128, 3DES, ARCFOUR128.
2556 31. Small patch to Sieve code - explicitly set From: when generating an
2559 32. Exim crashed if a remote delivery caused a very long error message to be
2560 recorded - for instance if somebody sent an entire SpamAssassin report back
2561 as a large number of 550 error lines. This bug was coincidentally fixed by
2562 increasing the size of one of Exim's internal buffers (big_buffer) that
2563 happened as part of the Exiscan merge. However, to be on the safe side, I
2564 have made the code more robust (and fixed the comments that describe what
2567 33. Some experimental protocols are using DNS PTR records for new purposes. The
2568 keys for these records are domain names, not reversed IP addresses. The
2569 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2570 leaves it alone. Component reversal etc. now happens only for IP addresses.
2573 34. The host_aton() function is supposed to be passed a string that is known
2574 to be a valid IP address. However, in the case of IPv6 addresses, it was
2575 not checking this. This is a hostage to fortune. Exim now panics and dies
2576 if the condition is not met. A case was found where this could be provoked
2577 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2578 components; fortuitously, this particular loophole had already been fixed
2579 by change 4.50/55 or 4.44/33 above.
2581 If there are any other similar loopholes, the new check in host_aton()
2582 itself should stop them being exploited. The report I received stated that
2583 data on the command line could provoke the exploit when Exim was running as
2584 exim, but did not say which command line option was involved. All I could
2585 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2586 running as the user.
2589 35. There was a buffer overflow vulnerability in the SPA authentication code
2590 (which came originally from the Samba project). I have added a test to the
2591 spa_base64_to_bits() function which I hope fixes it.
2594 36. The daemon start-up calls getloadavg() while still root for those OS that
2595 need the first call to be done as root, but it missed one case: when
2596 deliver_queue_load_max is set with deliver_drop_privilege. This is
2597 necessary for the benefit of the queue runner, because there is no re-exec
2598 when deliver_drop_privilege is set.
2600 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2601 was in use, was not putting the data itself into the right store pool;
2602 consequently, it could be overwritten for a subsequent message in the same
2603 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2606 38. Sometimes the final signoff response after QUIT could fail to get
2607 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2608 < 0 before doing a fflush(). This bug looks as though it goes back to the
2609 introduction of TLS in release 3.20, but "sometimes" must have been rare
2610 because the tests only now provoked it.
2616 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2617 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2618 Now it returns an integer. A number of calls were still expecting a T/F
2619 return. Fortuitously, in all cases, the tests worked in OK situations,
2620 which is the norm. However, things would have gone wrong on any write
2621 failures on the smtp file descriptor. This function is used when sending
2622 messages over SMTP and also when doing verify callouts.
2624 2. When Exim is called to do synchronous delivery of a locally submitted
2625 message (the -odf or -odi options), it no longer closes stderr before doing
2628 3. Implemented the mua_wrapper option.
2630 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2632 5. Implemented the functions header_remove(), header_testname(),
2633 header_add_at_position(), and receive_remove_recipient(), and exported them
2636 6. If an ACL "warn" statement specified the addition of headers, Exim already
2637 inserted X-ACL-Warn: at the start if there was no header name. However, it
2638 was not making this test for the second and subsequent header lines if
2639 there were newlines in the string. This meant that an invalid header could
2640 be inserted if Exim was badly configured.
2642 7. Allow an ACL "warn" statement to add header lines at the start or after all
2643 the Received: headers, as well as at the end.
2645 8. Added the rcpt_4xx retry error code.
2647 9. Added postmaster_mailfrom=xxx to callout verification option.
2649 10. Added mailfrom=xxxx to the callout verification option, for verify=
2652 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2653 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2655 12. Inserted some casts to stop certain compilers warning when using pointer
2656 differences as field lengths or precisions in printf-type calls (mostly
2657 affecting debugging statements).
2659 13. Added optional readline() support for -be (dynamically loaded).
2661 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2662 same clock tick as a message's arrival, so that its received time was the
2663 same as the "first fail" time on the retry record, and that message
2664 remained on the queue past the ultimate address timeout, every queue runner
2665 would try a delivery (because it was past the ultimate address timeout) but
2666 after another failure, the ultimate address timeout, which should have then
2667 bounced the address, did not kick in. This was a "< instead of <=" error;
2668 in most cases the first failure would have been in the next clock tick
2669 after the received time, and all would be well.
2671 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2672 being recognized when the domain list was tested by the match_domain
2673 condition in an expansion string.
2675 16. Added the ${str2b64: operator.
2677 17. Exim was always calling setrlimit() to set a large limit for the number of
2678 processes, without checking whether the existing limit was already
2679 adequate. (It did check for the limit on file descriptors.) Furthermore,
2680 errors from getrlimit() and setrlimit() were being ignored. Now they are
2681 logged to the main and panic logs, but Exim does carry on, to try to do its
2682 job under whatever limits there are.
2684 18. Imported PCRE 5.0.
2686 19. Trivial typo in log message " temporarily refused connection" (the leading
2689 20. If the log selector return_path_on_delivery was set and an address was
2690 redirected to /dev/null, the delivery process crashed because it assumed
2691 that a return path would always be set for a "successful" delivery. In this
2692 case, the whole delivery is bypassed as an optimization, and therefore no
2695 21. Internal re-arrangement: the function for sending a challenge and reading
2696 a response while authentication was assuming a zero-terminated challenge
2697 string. It's now changed to take a pointer and a length, to allow for
2698 binary data in such strings.
2700 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2702 23. Exim was not respecting finduser_retries when seeking the login of the
2703 uid under which it was called; it was always trying 10 times. (The default
2704 setting of finduser_retries is zero.) Also, it was sleeping after the final
2705 failure, which is pointless.
2707 24. Implemented tls_on_connect_ports.
2709 25. Implemented acl_smtp_predata.
2711 26. If the domain in control=submission is set empty, Exim assumes that the
2712 authenticated id is a complete email address when it generates From: or
2713 Sender: header lines.
2715 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2716 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2717 chown and chgrp in /bin and hostname in /usr/bin.
2719 28. Exim was keeping the "process log" file open after each use, just as it
2720 does for the main log. This opens the possibility of it remaining open for
2721 long periods when the USR1 signal hits a daemon. Occasional processlog
2722 errors were reported, that could have been caused by this. Anyway, it seems
2723 much more sensible not to leave this file open at all, so that is what now
2726 29. The long-running daemon process does not normally write to the log once it
2727 has entered its main loop, and it closes the log before doing so. This is
2728 so that log files can straightforwardly be renamed and moved. However,
2729 there are a couple of unusual error situations where the daemon does write
2730 log entries, and I had neglected to close the log afterwards.
2732 30. The text of an SMTP error response that was received during a remote
2733 delivery was being truncated at 512 bytes. This is too short for some of
2734 the long messages that one sometimes sees. I've increased the limit to
2737 31. It is now possible to make retry rules that apply only when a message has a
2738 specific sender, in particular, an empty sender.
2740 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2741 it possible to be selective about when SMTP synchronization is enforced.
2743 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2745 32. Implemented hosts_connection_nolog.
2747 33. Added an ACL for QUIT.
2749 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2752 35. Added mailbox_size and mailbox_filecount to appendfile.
2754 36. Added control = no_multiline_responses to ACLs.
2756 37. There was a bug in the logic of the code that waits for the clock to tick
2757 in the case where the clock went backwards by a substantial amount such
2758 that the microsecond fraction of "now" was more than the microsecond
2759 fraction of "then" (but the whole seconds number was less).
2761 38. Added support for the libradius Radius client library this is found on
2762 FreeBSD (previously only the radiusclient library was supported).
2768 1. When certain lookups returned multiple values in the form name=value, the
2769 quoting of the values was not always being done properly. Specifically:
2770 (a) If the value started with a double quote, but contained no whitespace,
2772 (b) If the value contained whitespace other than a space character (i.e.
2773 tabs or newlines or carriage returns) it was not quoted.
2774 This fix has been applied to the mysql and pgsql lookups by writing a
2775 separate quoting function and calling it from the lookup code. The fix
2776 should probably also be applied to nisplus, ibase and oracle lookups, but
2777 since I cannot test any of those, I have not disturbed their existing code.
2779 2. A hit in the callout cache for a specific address caused a log line with no
2780 reason for rejecting RCPT. Now it says "Previous (cached) callout
2781 verification failure".
2783 3. There was an off-by-one bug in the queryprogram router. An over-long
2784 return line was truncated at 256 instead of 255 characters, thereby
2785 overflowing its buffer with the terminating zero. As well as fixing this, I
2786 have increased the buffer size to 1024 (and made a note to document this).
2788 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2789 when Exim is waiting for an SMTP response from a remote server, Exim
2790 restarts its select() call on the socket, thereby resetting its timeout.
2791 This is not a problem when such interrupts are rare. Somebody set up a cron
2792 job to run exiwhat every 2 minutes, which is less than the normal select()
2793 timeout (5 or 10 minutes). This meant that the select() timeout never
2794 kicked in because it was always reset. I have fixed this by comparing the
2795 time when an interrupt arrives with the time at the start of the first call
2796 to select(). If more time than the timeout has elapsed, the interrupt is
2797 treated as a timeout.
2799 5. Some internal re-factoring in preparation for the addition of Sieve
2800 extensions (by MH). In particular, the "personal" test is moved to a
2801 separate function, and given an option for scanning Cc: and Bcc: (which is
2802 not set for Exim filters).
2804 6. When Exim created an email address using the login of the caller as the
2805 local part (e.g. when creating a From: or Sender: header line), it was not
2806 quoting the local part when it contained special characters such as @.
2808 7. Installed new OpenBSD configuration files.
2810 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2811 try to make them clearer.
2813 9. Callout options, other than the timeout value, were being ignored when
2814 verifying sender addresses in header lines. For example, when using
2816 verify = header_sender/callout=no_cache
2818 the cache was (incorrectly) being used.
2820 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2821 only the Cygwin environment.
2823 11. When return_path_on_delivery was set as a log selector, if different remote
2824 addresses in the same message used different return paths and parallel
2825 remote delivery occurred, the wrong values would sometimes be logged.
2826 (Whenever a remote delivery process finished, the return path value from
2827 the most recently started remote delivery process was logged.)
2829 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2830 lines when AUTH and/or TLS are in use. This is the "received protocol"
2831 field. Exim used to use "asmtp" for authenticated SMTP, without any
2832 indication (in the protocol name) for TLS use. Now it follows the RFC and
2833 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2834 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2835 names appear in log lines as well as in Received: header lines.
2837 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2838 extensions, and comparison tests, and to fix some bugs.
2840 14. Changes to the "personal" filter test:
2842 (1) The test was buggy in that it was just doing the equivalent of
2843 "contains" tests on header lines. For example, if a user's address was
2844 anne@some.where, the "personal" test would incorrectly be true for
2846 To: susanne@some.where
2848 This test is now done by extracting each address from the header in turn,
2849 and checking the entire address. Other tests that are part of "personal"
2850 are now done using regular expressions (for example, to check local parts
2851 of addresses in From: header lines).
2853 (2) The list of non-personal local parts in From: addresses has been
2854 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2855 taken from the Sieve specification recommendations.
2857 (3) If the message contains any header line starting with "List-" it is
2858 treated as non-personal.
2860 (4) The test for "circular" in the Subject: header line has been removed
2861 because it now seems ill-conceived.
2863 15. Minor typos in src/EDITME comments corrected.
2865 16. Installed latest exipick from John Jetmore.
2867 17. If headers_add on a router specified a text string that was too long for
2868 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2869 of string_sprintf() is now avoided.
2871 18. $message_body_size was not set (it was always zero) when running the DATA
2872 ACL and the local_scan() function.
2874 19. For the "mail" command in an Exim filter, no default was being set for
2875 the once_repeat time, causing a random time value to be used if "once" was
2876 specified. (If the value happened to be <= 0, no repeat happened.) The
2877 default is now 0s, meaning "never repeat". The "vacation" command was OK
2878 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2879 (I found it when inspecting the code).
2881 20. There is now an overall timeout for performing a callout verification. It
2882 defaults to 4 times the callout timeout, which applies to individual SMTP
2883 commands during the callout. The overall timeout applies when there is more
2884 than one host that can be tried. The timeout is checked before trying the
2885 next host. This prevents very long delays if there are a large number of
2886 hosts and all are timing out (e.g. when the network connections are timing
2887 out). The value of the overall timeout can be changed by specifying an
2888 additional sub-option for "callout", called "maxwait". For example:
2890 verify = sender/callout=5s,maxwait=20s
2892 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2893 to the end before writing, but this should make it even safer).
2895 22. Exim was forgetting that it had advertised PIPELINING for the second and
2896 subsequent messages on an SMTP connection. It was also not resetting its
2897 memory on STARTTLS and an internal HELO.
2899 23. When Exim logs an SMTP synchronization error within a session, it now
2900 records whether PIPELINING has been advertised or not.
2902 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2903 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2904 rather than long int.
2906 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2908 26. Added the never_mail option to autoreply.
2914 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2915 crash if the getsockname() call failed; this can happen if a connection is
2916 closed very soon after it is established. The problem was simply in the
2917 order in which certain operations were done, causing Exim to try to write
2918 to the SMTP stream before it had set up the file descriptor. The bug has
2919 been fixed by making things happen in the correct order.
2925 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2926 before the connection was closed, thus losing the rejection response.
2928 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2929 some early Solaris releases, but causes trouble in current releases where
2930 socklen_t is defined.
2932 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2935 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2938 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2939 that releasing the top of it at the end releases what was used for sub-
2940 expansions (unless the block got too big). However, discard this block if
2941 the first thing is a variable or header, so that we can use its block when
2942 it is dynamic (useful for very large $message_headers, for example).
2944 6. Lookups now cache *every* query, not just the most recent. A new, separate
2945 store pool is used for this. It can be recovered when all lookup caches are
2946 flushed. Lookups now release memory at the end of their result strings.
2947 This has involved some general refactoring of the lookup sources.
2949 7. Some code has been added to the store_xxx() functions to reduce the amount
2950 of flapping under certain conditions.
2952 8. log_incoming_interface used to affect only the <= reception log lines. Now
2953 it causes the local interface and port to be added to several more SMTP log
2954 lines, for example "SMTP connection from", and rejection lines.
2956 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2958 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2960 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2961 could be overwritten at the end of the current message (or the start of a
2962 new message if it was set in a HELO ACL). The value is now preserved for
2963 the duration of the SMTP connection.
2965 12. If a transport had a headers_rewrite setting, and a matching header line
2966 contained an unqualified address, that address was qualified, even if it
2967 did not match any rewriting rules. The underlying bug was that the values
2968 of the flags that permit the existence of unqualified sender and recipient
2969 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2970 non-local messages, and by -bnq for local messages) were not being
2971 preserved with the message after it was received.
2973 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2974 "next input=" as part of the text comprising the host identity instead of
2975 the correct text. The code was using the same buffer for two different
2976 strings. However, depending on which order the printing function evaluated
2977 its arguments, the bug did not always show up. Under Linux, for example, my
2978 test suite worked just fine.
2980 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2981 doesn't work with some older versions of Perl. It has been changed to "my",
2982 which in any case is probably the better facility to use.
2984 15. A really picky compiler found some instances of statements for creating
2985 error messages that either had too many or two few arguments for the format
2988 16. The size of the buffer for calls to the DNS resolver has been increased
2989 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2990 for addresses that have a lot of PTR records. This alleviates a problem; it
2991 does not fully solve it.
2993 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2994 buffer now truncates the list and logs the incident, which is the same
2995 action as happens when Exim is looking up a host name and its aliases.
2996 Previously in this situation something unpredictable would happen;
2997 sometimes it was "internal error: store_reset failed".
2999 18. If a server dropped the connection unexpectedly when an Exim client was
3000 using GnuTLS and trying to read a response, the client delivery process
3001 crashed while trying to generate an error log message.
3003 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3004 string, for example:
3006 warn message = H1: something\nH2: something
3008 the text was added as a single header line from Exim's point of view
3009 though it ended up OK in the delivered message. However, searching for the
3010 second and subsequent header lines using $h_h2: did not work. This has been
3011 fixed. Similarly, if a system filter added multiple headers in this way,
3012 the routers could not see them.
3014 20. Expanded the error message when iplsearch is called with an invalid key to
3015 suggest using net-iplsearch in a host list.
3017 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3018 ACLs are no longer actually imposed (and a message to that effect is
3021 22. If a "gecos" field in a passwd entry contained escaped characters, in
3022 particular, if it contained a \" sequence, Exim got it wrong when building
3023 a From: or a Sender: header from that name. A second bug also caused
3024 incorrect handling when an unquoted " was present following a character
3025 that needed quoting.
3027 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3028 was not being matched caselessly.
3030 24. Arranged for all hyphens in the exim.8 source to be escaped with
3033 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3034 properly. Recipient callout cache records were still being keyed to include
3035 the sender, even when use_sender was set false. This led to far more
3036 callouts that were necessary. The sender is no longer included in the key
3037 when use_sender is false.
3039 26. Added "control = submission" modifier to ACLs.
3041 27. Added the ${base62d: operator to decode base 62 numbers.
3043 28. dnsdb lookups can now access SRV records.
3045 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3046 the configuration file.
3048 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3049 (-v) mode. This makes the output for a verbose queue run more intelligible.
3051 31. Added a use_postmaster feature to recipient callouts.
3053 32. Added the $body_zerocount variable, containing the number of binary zero
3054 bytes in the message body.
3056 33. The time of last modification of the "new" subdirectory is now used as the
3057 "mailbox time last read" when there is a quota error for a maildir
3060 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3062 35. Added +ignore_unknown as a special item in host lists.
3064 36. Code for decoding IPv6 addresses in host lists is now included, even if
3065 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3066 address was recognized as an IP address, but was then not correctly decoded
3067 into binary, causing unexpected and incorrect effects when compared with
3074 1. Very minor rewording of debugging text in manualroute to say "list of
3075 hosts" instead of "hostlist".
3077 2. If verify=header_syntax was set, and a header line with an unqualified
3078 address (no domain) and a large number of spaces between the end of the
3079 name and the colon was received, the reception process suffered a buffer
3080 overflow, and (when I tested it) crashed. This was caused by some obsolete
3081 code that should have been removed. The fix is to remove it!
3083 3. When running in the test harness, delay a bit after writing a bounce
3084 message to get a bit more predictability in the log output.
3086 4. Added a call to search_tidyup() just before forking a reception process. In
3087 theory, someone could use a lookup in the expansion of smtp_accept_max_
3088 per_host which, without the tidyup, could leave open a database connection.
3090 5. Added the variables $recipient_data and $sender_data which get set from a
3091 lookup success in an ACL "recipients" or "senders" condition, or a router
3092 "senders" option, similar to $domain_data and $local_part_data.
3094 6. Moved the writing of debug_print from before to after the "senders" test
3097 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3098 problems for message scanning, either using a data ACL, or using
3099 local_scan() because the Received: header was not generated till after they
3100 were called (in order to set the time as the time of reception completion).
3101 I have revised the way this works. The header is now generated after the
3102 body is received, but before the ACL or local_scan() are called. After they
3103 are run, the timestamp in the header is updated.
3109 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3110 before starting a queue runner without re-exec. This happened only when
3111 deliver_drop_privilege was set or when the Exim user was set to root. The
3112 effect of the bug was that timeouts during subsequent deliveries caused
3113 crashes instead of being properly handled. The handler is now left at its
3114 default (and expected) setting.
3116 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3117 message, again when deliver_drop_privilege is set or Exim is run as root.
3118 The bug described in (1) was not present in this case, but the tidying up
3119 of the other signals was missing. I have made the two cases consistent.
3121 3. The ignore_target_hosts setting on a manualroute router was being ignored
3122 for hosts that were looked up using the /MX notation.
3124 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3127 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3128 operated on the sender address. After changing the $sender_address to <>
3129 for the sender address verify, Exim was re-instated it as the original
3130 (before rewriting) address, but remembering that it had rewritten it, so it
3131 wasn't rewriting it again. This bug also had the effect of breaking the
3132 sender address verification caching when the sender address was rewritten.
3134 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3135 This has been changed so that if the ip literal address matches
3136 ignore_target_hosts, the router declines.
3138 7. Added expansion conditions match_domain, match_address, and match_local_
3139 part (NOT match_host).
3141 8. The placeholder for the Received: header didn't have a length field set.
3143 9. Added code to Exim itself and to exim_lock to test for a specific race
3144 condition that could lead to file corruption when using MBX delivery. The
3145 issue is with the lockfile that is created in /tmp. If this file is removed
3146 after a process has opened it but before that process has acquired a lock,
3147 there is the potential for a second process to recreate the file and also
3148 acquire a lock. This could lead to two Exim processes writing to the file
3149 at the same time. The added code performs the same test as UW imapd; it
3150 checks after acquiring the lock that its file descriptor still refers to
3151 the same named file.
3153 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3154 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3157 11. Added the smtp_active_hostname option. If used, this will typically be made
3158 to depend on the incoming interface address. Because $interface_address is
3159 not set up until the daemon has forked a reception process, error responses
3160 that can happen earlier (such as "too many connections") no longer contain
3163 12. If an expansion in a condition on a "warn" statement fails because a lookup
3164 defers, the "warn" statement is abandoned, and the next ACL statement is
3165 processed. Previously this caused the whole ACL to be aborted.
3167 13. Added the iplsearch lookup type.
3169 14. Added ident_timeout as a log selector.
3171 15. Added tls_certificate_verified as a log selector.
3173 16. Added a global option tls_require_ciphers (compare the smtp transport
3174 option of the same name). This controls incoming TLS connections.
3176 17. I finally figured out how to make tls_require_ciphers do a similar thing
3177 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3178 before starting the TLS session.
3180 18. Tabs are now shown as \t in -bP output.
3182 19. If the log selector return_path_on_delivery was set, Exim crashed when
3183 bouncing a message because it had too many Received: header lines.
3185 20. If two routers both had headers_remove settings, and the first one included
3186 a superfluous trailing colon, the final name in the first list and the
3187 first name in the second list were incorrectly joined into one item (with a
3188 colon in the middle).
3194 1. Added -C and -D options to the exinext utility, mainly to make it easier
3195 to include in the automated testing, but these could be helpful when
3196 multiple configurations are in use.
3198 2. The exinext utility was not formatting the output nicely when there was
3199 an alternate port involved in the retry record key, nor when there was a
3200 message id as well (for retries that were specific to a specific message
3201 and a specific host). It was also confused by IPv6 addresses, because of
3202 the additional colons they contain. I have fixed the IPv4 problem, and
3203 patched it up to do a reasonable job for IPv6.
3205 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3206 delivery, the log line now contains "pipelined" if PIPELINING was used.
3208 4. An SMTP transport process used to panic and die if the bind() call to set
3209 an explicit outgoing interface failed. This has been changed; it is now
3210 treated in the same way as a connect() failure.
3212 5. A reference to $sender_host_name in the part of a conditional expansion
3213 that was being skipped was still causing a DNS lookup. This no longer
3216 6. The def: expansion condition was not recognizing references to header lines
3217 that used bh_ and bheader_.
3219 7. Added the _cache feature to named lists.
3221 8. The code for checking quota_filecount in the appendfile transport was
3222 allowing one more file than it should have been.
3224 9. For compatibility with Sendmail, the command line option
3232 and sets the incoming protocol and host name (for trusted callers). The
3233 host name and its colon can be omitted when only the protocol is to be set.
3234 Note the Exim already has two private options, -pd and -ps, that refer to
3235 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3236 "s", but I don't think that's a major issue.
3238 10. A number of refactoring changes to the code, none of which should affect
3241 (a) The number of logging options was getting close to filling up the
3242 32-bit word that was used as a bit map. I have split them into two classes:
3243 those that are passed in the argument to log_write(), and those that are
3244 only ever tested independently outside of that function. These are now in
3245 separate 32-bit words, so there is plenty of room for expansion again.
3246 There is no change in the user interface or the logging behaviour.
3248 (b) When building, for example, log lines, the code previously used a
3249 macro that called string_cat() twice, in order to add two strings. This is
3250 not really sufficiently general. Furthermore, there was one instance where
3251 it was actually wrong because one of the argument was used twice, and in
3252 one call a function was used. (As it happened, calling the function twice
3253 did not affect the overall behaviour.) The macro has been replaced by a
3254 function that can join an arbitrary number of extra strings onto a growing
3257 (c) The code for expansion conditions now uses a table and a binary chop
3258 instead of a serial search (which was left over from when there were very
3259 few conditions). Also, it now recognizes conditions like "pam" even when
3260 the relevant support is not compiled in: a suitably worded error message is
3261 given if an attempt is made to use such a condition.
3263 11. Added ${time_interval:xxxxx}.
3265 12. A bug was causing one of the ddress fields not to be passed back correctly
3266 from remote delivery subprocesses. The field in question was not being
3267 subsequently used, so this caused to problems in practice.
3269 13. Added new log selectors queue_time and deliver_time.
3271 14. Might have fixed a bug in maildirsizefile handling that threw up
3272 "unexpected character" debug warnings, and recalculated the data
3273 unnecessarily. In any case, I expanded the warning message to give more
3276 15. Added the message "Restricted characters in address" to the statements in
3277 the default ACL that block characters like @ and % in local parts.
3279 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3280 Three changes have been made:
3282 (a) There was a serious bug; a negative response to MAIL caused the whole
3283 recipient domain to be cached as invalid, thereby blocking all messages
3284 to all local parts at the same domain, from all senders. This bug has
3285 been fixed. The domain is no longer cached after a negative response to
3286 MAIL if the sender used is not empty.
3288 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3291 (c) A new callout option, "use_sender" has been added for people who want
3292 the modified behaviour.
3298 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3301 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3302 indeed breaks things for older releases.
3304 3. Added additional logging to the case where there is a problem reading data
3305 from a filter that is running in a subprocess using a pipe, in order to
3306 try to track down a specific problem.
3308 4. Testing facility fudge: when running in the test harness and attempting
3309 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3310 getting "No route to host". Convert this to a timeout.
3312 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3315 6. Some OS don't have socklen_t but use size_t instead. This affects the
3316 fifth argument of getsockopt() amongst other things. This is now
3317 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3318 can be set for individual OS. I have set it for SunOS5, OSF1, and
3319 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3320 some earlier ones do not.
3322 7. Change 4.30/15 was not doing the test caselessly.
3324 8. The standard form for an IPv6 address literal was being rejected by address
3325 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3326 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3327 this, as well as the form without the "IPv6" on the front (but only when
3328 address literals are enabled, of course).
3330 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3332 10. Exim crashed if a message with an empty sender address specified by -f
3333 encountered a router with an errors_to setting. This could be provoked only
3334 by a command such as
3338 where an empty string was supplied; "<>" did not hit this bug.
3340 11. Installed PCRE release 4.5.
3342 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3343 remained set. It is now erased.
3345 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3346 times from message ids (which are base 36 rather than the normal 62).
3348 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3349 were being counted as actual protocol errors, and logged if the log
3350 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3351 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3352 and DATA following a set of rejected RCPTs do not count as protocol errors.
3353 In other words, Exim assumes they were pipelined, though this may not
3354 actually be the case. Of course, in all cases the client gets an
3355 appropriate error code.
3357 15. If a lookup fails in an ACL condition, a message about the failure may
3358 be available; it is used if testing the ACL cannot continue, because most
3359 such messages specify what the cause of the deferral is. However, some
3360 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3361 that caused an old message to be retained and used if a later statement
3362 caused a defer, replacing the real cause of the deferral.
3364 16. If an IP address had so many PTR records that the DNS lookup buffer
3365 was not large enough to hold them, Exim could crash while trying to process
3366 the truncated data. It now detects and logs this case.
3368 17. Further to 4.21/58, another change has been made: if (and only if) the
3369 first line of a message (the first header line) ends with CRLF, a bare LF
3370 in a subsequent header line has a space inserted after it, so as not to
3371 terminate the header.
3373 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3374 unnecessarily, used atoi() instead of strtol(), and didn't check the
3375 termination when getting file sizes from file names by regex.
3377 19. Completely re-implemented the support for maildirsize files, in the light
3378 of a number of problems with the previous contributed implementation
3379 (4.30/29). In particular:
3381 . If the quota is zero, the maildirsize file is maintained, but no quota is
3384 . If the maildir directory does not exist, it is created before any attempt
3385 to write a maildirsize file.
3387 . The quota value in the file is just a cache; if the quota is changed in
3388 the transport, the new value overrides.
3390 . A regular expression is available for excluding directories from the
3393 20. The autoreply transport checks the characters in options that define the
3394 message's headers; it allows continued headers, but it was checking with
3395 isspace() after an embedded newline instead of explicitly looking for a
3398 21. If all the "regular" hosts to which an address was routed had passed their
3399 expiry times, and had not reached their retry times, the address was
3400 bounced, even if fallback hosts were defined. Now Exim should go on to try
3403 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3404 equivalent code in the SMTP transport. Some hosts send humungous responses
3405 to HELO/EHLO, more than 1024 it seems.
3407 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3408 gives compiler warnings in some environments. I've now done it "properly",
3411 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3412 (because of problems with the built-in one) was declared to return uschar *
3413 instead of char *, causing compiler failure.
3415 25. Fixed a file descriptor leak when processing alias/forward files.
3417 26. Fixed a minor format string issue in dbfn.c.
3419 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3421 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3422 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3425 29. When a sender address is verified, it is cached, to save repeating the test
3426 when there is more than one recipient in a message. However, when the
3427 verification involves a callout, it is possible for different callout
3428 options to be set for different recipients. It is too complicated to keep
3429 track of this in the cache, so now Exim always runs a verification when a
3430 callout is required, relying on the callout cache for the optimization.
3431 The overhead is duplication of the address routing, but this should not be
3434 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3435 to be verified with callout=postmaster, and the main callout worked but the
3436 postmaster check failed, the verification correctly failed. However, if a
3437 subsequent RCPT command asked for sender verification *without* the
3438 postmaster check, incorrect caching caused this verification also to fail,
3441 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3442 it was not caching the DNS options (qualify_single, search_parents) that
3443 were used when the lookup failed. A subsequent lookup with different
3444 options therefore always gave the same answer, though there were cases
3445 where it should not have. (Example: a "domains = !$mx_any" option on a
3446 dnslookup router: the "domains" option is always processed without any
3447 widening, but the router might have qualify_single set.) Now Exim uses the
3448 cached value only when the same options are set.
3450 32. Added John Jetmore's "exipick" utility to the distribution.
3452 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3453 than a timeout (e.g. a certificate is required, and is not provided), an
3454 Exim server now closes the connection immediately. Previously it waited for
3455 the client to close - but if the client is SSL, it seems that they each
3456 wait for each other, leading to a delay before one of them times out.
3458 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3459 maintained 0.8.x compatibility because I don't think many are using it, and
3460 it is clearly obsolete.
3462 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3465 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3466 client certificate was expired. A simple patch fixes this, though I don't
3467 understand the full logic of why the verify callback is called multiple
3470 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3471 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3472 which causes problems with some clients (such as the Certicom SSL Plus
3473 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3474 disables the coutermeasure allowing Eudora to connect."
3476 38. Exim was not checking that a write() to a log file succeeded. This could
3477 lead to Bad Things if a log got too big, in particular if it hit a file
3478 size limit. Exim now panics and dies if it cannot write to a log file, just
3479 as it does if it cannot open a log file.
3481 39. Modified OS/Makefile-Linux so that it now contains
3483 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3485 The two -D definitions ensure that Exim is compiled with large file
3486 support, which makes it possible to handle log files that are bigger than
3489 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3490 instance) a domain was checked against a named list that involved a lookup,
3491 causing $domain_data to be set, then another domain was checked against the
3492 same list, then the first domain was re-checked, the value of $domain_data
3493 after the final check could be wrong. In particular, if the second check
3494 failed, it could be set empty. This bug probably also applied to
3497 41. The strip_trailing_dot option was not being applied to the address given
3498 with the -f command-line option.
3500 42. The code for reading a message's header from the spool was incrementing
3501 $received_count, but never initializing it. This meant that the value was
3502 incorrect (doubled) while delivering a message in the same process in which
3503 it was received. In the most common configuration of Exim, this never
3504 happens - a fresh exec is done - but it can happen when
3505 deliver_drop_privilege is set.
3507 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3508 it now includes up to 150 characters of the unexpected data in the log
3511 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3512 and building data strings. The size of both of these buffers was 10 000
3513 bytes - far larger than anybody would *ever* want, thought I. Needless to
3514 say, somebody hit the limit. I have increased the maximum line length to
3515 20 000 and the maximum data length of concatenated lines to 100 000. I have
3516 also fixed two bugs, because there was no checking on these buffers. Tsk,
3517 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3518 buffer is too small.
3520 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3521 lsearch lookups. Now it does.
3523 46. When parsing a route_list item in a manualroute router, a fixed-length
3524 buffer was used for the list of hosts. I made this 1024 bytes long,
3525 thinking that nobody would ever have a list of hosts that long. Wrong.
3526 Somebody had a whole pile of complicated expansion conditions, and the
3527 string was silently truncated, leading to an expansion error. It turns out
3528 that it is easier to change to an unlimited length (owing to other changes
3529 that have happened since this code was originally written) than to build
3530 structure for giving a limitation error. The length of the item that
3531 expands into the list of hosts is now unlimited.
3533 47. The lsearch lookup could not handle data where the length of text line was
3534 more than 4095 characters. Such lines were truncated, leading to shortened
3535 data being returned. It should now handle lines of any length.
3537 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3538 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3541 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3542 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3543 no longer happens in comment lines. A list of replacements is now placed
3544 at the head of all of the source files, except those whose only change is
3545 to replace PERL_COMMAND in the very first #! line.
3547 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3548 messages on the queue, with a bottom-up merge sort, using code contributed
3549 by Michael Haardt. This should make operations like -bp somewhat faster on
3550 large queues. It won't affect queue runners, except when queue_run_in_order
3553 51. Installed eximstats 1.31 in the distribution.
3555 52. Added support for SRV lookups to the dnslookup router.
3557 53. If an ACL referred to $message_body or $message_body_end, the value was not
3558 reset for any messages that followed in the same SMTP session.
3560 54. The store-handling optimization for building very long strings was not
3561 differentiating between the different store pools. I don't think this
3562 actually made any difference in practice, but I've tidied it.
3564 55. While running the routers to verify a sender address, $sender_address
3565 was still set to the sender address. This is wrong, because when routing to
3566 send a bounce to the sender, it would be empty. Therefore, I have changed
3567 it so that, while verifying a sender address, $sender_address is set to <>.
3568 (There is no change to what happens when verifying a recipient address.)
3570 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3571 target A or AAAA records (if not already returned) without resetting the
3572 qualify_single or search_parents options of the DNS resolver. These are
3573 inappropriate in this case because the targets of MX and SRV records must
3574 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3575 target that, when qualified, matched something in the local domain. These
3576 two options are now turned off when doing these lookups.
3578 57. It seems that at least some releases of Reiserfs (which does not have the
3579 concept of a fixed number of inodes) returns zero and not -1 for the
3580 number of available inodes. This interacted badly with check_spool_inodes,
3581 which assumed that -1 was the "no such thing" setting. What I have done is
3582 to check that the total number of inodes is greater than zero before doing
3583 the test of how many are available.
3585 58. When a "warn" ACL statement has a log_message modifier, the message is
3586 remembered, and not repeated. This is to avoid a lot of repetition when a
3587 message has many recipients that cause the same warning to be written.
3588 Howewer, Exim was preserving the list of already written lines for an
3589 entire SMTP session, which doesn't seem right. The memory is now reset if a
3590 new message is started.
3592 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3593 debugging output unless log_rewrite was also set.
3595 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3596 of (int)(handle) when we know that handle contains (void *)(-1).
3598 61. The Exim daemon panic-logs an error return when it closes the incoming
3599 connection. However "connection reset by peer" seems to be common, and
3600 isn't really an error worthy of noting specially, so that particular error
3603 62. When Exim is trying to find all the local interfaces, it used to panic and
3604 die if the ioctl to get the interface flags failed. However, it seems that
3605 on at least one OS (Solaris 9) it is possible to have an interface that is
3606 included in the list of interfaces, but for which you get a failure error
3607 for this call. This happens when the interface is not "plumbed" into a
3608 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3609 failure of the "get flags" call assumes that the interface is down.
3611 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3612 makes life easier for people who are doing arithmetic on fields extracted
3613 from dates, where you often get leading zeros that should not be
3614 interpreted as octal.
3616 64. Added qualify_domain to the redirect router, to override the global
3619 65. If a pathologically long header line contained very many addresses (the
3620 report of this problem mentioned 10 000) and each of them was rewritten,
3621 Exim could use up a very large amount of memory. (It kept on making new
3622 copies of the header line as it rewrote, and never released the old ones.)
3623 At the expense of a bit more processing, the header rewriting function has
3624 been changed so that it no longer eats memory in this way.
3626 66. The generation of the Received: header has been moved from the time that a
3627 message starts to be received, to the time that it finishes. The timestamp
3628 in the Received: header should now be very close to that of the <= log
3629 line. There are two side-effects of this change:
3631 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3632 logged header lines no longer include the local Received: line, because
3633 it has not yet been created. The same applies to a copy of the message
3634 that is returned to a non-SMTP sender when a message is rejected.
3636 (b) When a filter file is tested using -bf, no additional Received: header
3637 is added to the test message. After some thought, I decided that this
3640 This change does not affect the value of $received_for. It is still set
3641 after address rewriting, but before local_scan() is called.
3643 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3645 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3646 gave an unhelpful panic error message, and a defer error. I have managed to
3647 change this behaviour so that it now rejects any supplied certificate,
3648 which seems right, as the list of acceptable certificates is empty.
3650 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3651 gave an unhelpful defer error. I have not managed to make this reject any
3652 supplied certificates, but the error message it gives is "no certificate
3653 supplied", which is not helpful.
3655 70. exigrep's output now also includes lines that are not associated with any
3656 message, but which match the given pattern. Implemented by a patch from
3657 Martin Sluka, which also tidied up the Perl a bit.
3659 71. Recipient callout verification, like sender verification, was using <> in
3660 the MAIL FROM command. This isn't really the right thing, since the actual
3661 sender may affect whether the remote host accepts the recipient or not. I
3662 have changed it to use the actual sender in the callout; this means that
3663 the cache record is now keyed on a recipient/sender pair, not just the
3664 recipient address. There doesn't seem to be a real danger of callout loops,
3665 since a callout by the remote host to check the sender would use <>.
3666 [SEE ABOVE: changed after hitting problems.]
3668 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3669 temporary errors. However, in the case of such a code being given after
3670 the end of a data transmission (i.e. after ".") Exim was failing to write
3671 a retry record for the message. (Yes, there was some broken host that was
3672 actually sending 8xx at this point.)
3674 73. An unknown lookup type in a host list could cause Exim to panic-die when
3675 the list was checked. (An example that provoked this was putting <; in the
3676 middle of a list instead of at the start.) If this happened during a DATA
3677 ACL check, a -D file could be left lying around. This kind of configuration
3678 error no longer causes Exim to die; instead it causes a defer errror. The
3679 incident is still logged to the main and panic logs.
3681 74. Buglet left over from Exim 3 conversion. The message "too many messages
3682 in one connection" was written to the rejectlog but not the mainlog, except
3683 when address rewriting (yes!) was being logged.
3685 75. Added write_rejectlog option.
3687 76. When a system filter was run not as root (that is, when system_filter_user
3688 was set), the values of the $n variables were not being returned to the
3689 main process; thus, they were not subsequently available in the $sn
3692 77. Added +return_path_on_delivery log selector.
3694 78. A connection timeout was being treated differently from recipients deferred
3695 when testing hosts_max_try with a message that was older than the host's
3696 retry timeout. (The host should not be counted, thus allowing all hosts to
3697 be tried at least once before bouncing.) This may have been the cause of an
3698 occasionally reported bug whereby a message would remain on the queue
3699 longer than the retry timeout, but would be bounced if a delivery was
3700 forced. I say "may" because I never totally pinned down the problem;
3701 setting up timeout/retry tests is difficult. See also the next item.
3703 79. The ultimate address timeout was not being applied to errors that involved
3704 a combination of host plus message (for example, a timeout on a MAIL
3705 command). When an address resolved to a number of possible hosts, and they
3706 were not all tried for each delivery (e.g. because of hosts_max_try), a
3707 message could remain on the queue longer than the retry timeout.
3709 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3712 81. Fixed an obscure SMTP outgoing bug which required at least the following
3713 conditions: (a) there was another message waiting for the same server;
3714 (b) the server returned 5xx to all RCPT commands in the first message so
3715 that the message was not completed; (c) the server dropped the connection
3716 or gave a negative response to the RSET that Exim sends to abort the
3717 transaction. The observed case was a dropped connection after DATA that had
3718 been sent in pipelining mode. That is, the server had advertised PIPELINING
3719 but was not implementing it correctly. The effect of the bug was incorrect
3720 behaviour, such as trying another host, and this could lead to a crash.
3726 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3727 and daemon.c were passed as pointers to ints; they should have been
3728 pointers to socklen_t variables (which are typically unsigned ints).
3730 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3733 3. Fixed a really odd bug that affected only the testing scheme; patching a
3734 certain fixed string in the binary changed the value of another string that
3735 happened to be identical to the end of the original first string.
3737 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3738 name", it returns that address as the IP address. On some operating
3739 systems (e.g. Solaris), it also passes back the IP address string as the
3740 "host name". However, on others (e.g. Linux), it passes back an empty
3741 string. Exim wasn't checking for this, and was changing the host name to an
3742 empty string, assuming it had been canonicized.
3744 5. Although rare, it is permitted to have more than one PTR record for a given
3745 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3746 all the names associated with an address, because they do in Solaris.
3747 However, it seems that they do not in Linux for data that comes from the
3748 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3749 I found this out when I moved to a new Linux workstation and tried to run
3750 the Exim test suite.
3752 To get round this problem I have changed the code so that it now does its
3753 own call to the DNS to look up PTR records when searching for a host name.
3754 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3755 addresses that are only in /etc/hosts are still found.
3757 This behaviour is, however, controlled by an option called host_lookup_
3758 order, which defaults to "bydns:byaddr". If people want to use the other
3759 order, or indeed, just use one or the other means of lookup, they can
3760 specify it in this variable.
3762 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3763 some operating systems, this comes back from gethostbyaddr() as an empty
3764 string, and this is what Exim used to test for. However, it seems that in
3765 other systems, "." is yielded. Exim now tests for this case too.
3767 7. The values of check_spool_space and check_log_space are now held internally
3768 as a number of kilobytes instead of an absolute number of bytes. If a
3769 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3770 kilobyte. This means that much larger values can be stored.
3772 8. Exim monitor: an attempt to get the action menu when not actually pointing
3773 at a message produces an empty menu entitled "No message selected". This
3774 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3775 no entries in it ("Shell widget menu has zero width and/or height"). So I
3776 have added a single, blank menu entry in this case.
3778 9. Added ${quote_local_part.
3780 10. MIME decoding is now applied to the contents of Subject: header lines when
3783 11. Now that a reference to $sender_host_address automatically causes a reverse
3784 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3785 host lookup before query-style lookups in lists that might use this
3786 variable. This has therefore been abolished, and the "net-" prefix is no
3787 longer necessary for query-style lookups.
3789 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3790 have been a typo for LFLAGS, so it has been changed.
3792 13. The install script calls Exim with "-C /dev/null" in order to find the
3793 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3794 to be output. Howeve, since Exim outputs its version number before the
3795 error, it didn't break the script. It just looked ugly. I fixed this by
3796 always allowing "-C /dev/null" if the caller is root.
3798 14. Ignore overlarge ACL variable number when reading spool file - insurance
3799 against a later release with more variables having written the file.
3801 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3802 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3803 this, as well as the form without the "IPv6" on the front.
3805 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3806 OS/Makefile-Darwin file.
3808 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3809 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3811 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3812 variable. However, some OS use ints for inodes. I've added cast to long int
3813 to get rid of the compiler warning.
3815 19. I had forgotten to lock out "/../" in configuration file names when
3816 ALT_CONFIG_PREFIX was set.
3818 20. Routers used for verification do not need to specify transports. However,
3819 if such a router generated a host list, and callout was configured, Exim
3820 crashed, because it could not find a port number from the (non-existent)
3821 transport. It now assumes port 25 in this circumstance.
3823 21. Added the -t option to exigrep.
3825 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3826 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3827 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3828 them is set, LOOKUP_LSEARCH is forced.
3830 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3832 24. Added sender and host information to the "rejected by local_scan()" log
3833 line; previously there was no indication of these.
3835 25. Added .include_if_exists.
3837 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3838 Linux. It turns out that not all file systems support this. Apparently some
3839 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3840 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3841 not supported on the file descriptor, is now ignored when Exim is trying to
3842 sync a directory. This applies only to Linux.
3844 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3846 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3847 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3850 29. Added support for maildirsize files from supplied patch (modified a bit).
3852 30. The use of :fail: followed by an empty string could lead Exim to respond to
3853 sender verification failures with (e.g.):
3855 550 Verification failed for <xxx>
3856 550 Sender verify failed
3858 where the first response line was missing the '-' that indicates it is not
3859 the final line of the response.
3861 31. The loop for finding the name of the user that called Exim had a hardwired
3862 limit of 10; it now uses the value of finduser_retries, which is used for
3863 all other user lookups.
3865 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3868 33. Exim was neglecting to zero errno before one call of strtol() when
3869 expanding a string and expecting an integer value. On some systems this
3870 resulted in spurious "integer overflow" errors. Also, it was casting the
3871 result into an int without checking.
3873 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3874 did not work. The code looks as if it has *never* worked, though it appears
3875 to have been documented since at least releast 1.62. I have made it work.
3877 35. The "timeout_DNS" error in retry rules, also documented since at least
3878 1.62, also never worked. As it isn't clear exactly what this means, and
3879 clearly it isn't a major issue, I have abolished the feature by treating it
3880 as "timeout", and writing a warning to the main and panic logs.
3882 36. The display of retry rules for -brt wasn't always showing the error code
3885 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3886 timeout_connect_A, timeout_connect_MX.
3888 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3889 to the empty sender.
3891 39. The daemon was not analysing the content of -oX till after it had closed
3892 stderr and disconnected from the controlling terminal. This meant that any
3893 syntax errors were only noted on the panic log, and the return code from
3894 the command was 0. By re-arranging the code a little, I've made the
3895 decoding happen first, so such errors now appear on stderr, and the return
3896 code is 1. However, the actual setting up of the sockets still happens in
3897 the disconnected process, so errors there are still only recorded on the
3900 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3901 connections (as happens on some IP stacks) was logged at start up time as
3902 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3903 it from "IPv6 and IPv4", which means that two separate sockets are being
3906 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3907 says whether AF_INET or AF_INET6 was passed as an argument.
3909 42. Exiwhat output was messed up when time zones were included in log
3912 43. Exiwhat now gives more information about the daemon's listening ports,
3913 and whether -tls-on-connect was used.
3915 44. The "port" option of the smtp transport is now expanded.
3917 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3918 silently ignored. Now an error message is written to the main and panic
3921 46. There's a new ACL modifier called "logwrite" which writes to a log file
3922 as soon as it is encountered.
3924 47. Added $local_user_uid and $local_user_gid at routing time.
3926 48. Exim crashed when trying to verify a sender address that was being
3929 49. Exim was recognizing only a space character after ".include". It now also
3930 recognizes a tab character.
3932 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3933 extracting the relevant information from the specification. The man page no
3934 longer contains scrambled data for the -d option, and I've added a section
3935 at the front about calling Exim under different names.
3937 51. Added "extra_headers" argument to the "mail" command in filter files.
3939 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3942 53. Installed eximstats 1.29.
3944 54. Added transport_filter_timeout as a generic transport option.
3946 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3947 Cc: header lines. This was required by RFC 822, but it not required by RFC
3950 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3951 incoming messages that did not have them. Now it does so only if the
3952 message originates locally, that is, if there is no associated remote host
3953 address. When Resent- header lines are present, this applies to the Resent-
3954 lines rather than the non-Resent- lines.
3956 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3957 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3959 58. Messages for configuration errors now include the name of the main
3960 configuration file - useful now that there may be more than one file in a
3961 list (.included file names were always shown).
3963 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3964 for those rare installations that do not start the daemon as root or run it
3965 setuid root. I've cut out the call to initgroups() if the daemon is not
3968 60. The Exim user and group can now be bound into the binary as text strings
3969 that are looked up at the start of Exim's processing.
3971 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3973 62. Added $mailstore_basename variable.
3975 63. Installed patch to sieve.c from Michael Haardt.
3977 64. When Exim failed to open the panic log after failing to open the main log,
3978 the original message it was trying to log was written to stderr and debug
3979 output, but if they were not available (the usual case in production), it
3980 was lost. Now it is written to syslog before the two lines that record the
3981 failures to open the logs.
3983 65. Users' Exim filters run in subprocesses under the user's uid. It is
3984 possible for a "deliver" command or an alias in a "personal" command to
3985 provoke an address rewrite. If logging of address rewriting is configured,
3986 this fails because the process is not running as root or exim. There may be
3987 a better way of dealing with this, but for the moment (because 4.30 needs
3988 to be released), I have disabled address rewrite logging when running a
3989 filter in a non-root, non-exim process.
3995 1. The buildconfig auxiliary program wasn't quoting the value set for
3996 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3997 not defined. This bug was masked in 4.22 by the effect that was fixed in
4000 2. Some messages that were rejected after a message id was allocated were
4001 shown as "incomplete" by exigrep. It no longer does this for messages that
4002 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4004 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4005 have allow_domain_literals set, the ID did not get logged in the <= line.
4006 Domain literals are now always recognized in Message-ID: header lines.
4008 4. The first argument for a ${extract expansion item is the key name or field
4009 number. Leading and trailing spaces in this item were not being ignored,
4010 causing some misleading effects.
4012 5. When deliver_drop_privilege was set, single queue runner processes started
4013 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4014 same command in the process it spins off) were not dropping privilege.
4016 6. When the daemon running as "exim" started a queue runner, it always
4017 re-executed Exim in the spun-off process. This is a waste of effort when
4018 deliver_drop_privilege is set. The new process now just calls the
4019 queue-runner function directly.
4025 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4028 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4029 ignored. Though the use of -oP was forcing the writing of a pid file, it
4030 was always written to the default place.
4032 3. If the message "no IP address found for host xxxx" is generated during
4033 incoming verification, it is now followed by identification of the incoming
4034 connection (so you can more easily find what provoked it).
4036 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4038 5. Added some features to "harden" Exim a bit more against certain attacks:
4040 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4041 be put in Local/Makefile. This is like the never_users runtime option,
4042 but it cannot be overridden. The default setting is "root".
4044 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4045 prefix string with which any file named in a -C command line option
4048 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4049 is retained for -C and -D only if the caller of Exim is root. Without
4050 it, the exim user may also use -C and -D and retain privilege.
4052 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4053 command line option is disabled.
4055 6. Macro names set by the -D option must start with an upper case letter, just
4056 like macro names defined in the configuration file.
4058 7. Added "dereference=" facility to LDAP.
4060 8. Two instances of the typo "uknown" in the source files are fixed.
4062 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4063 the Configure-Makefile script screwed up while processing it.
4065 10. Incorporated PCRE 4.4.
4067 11. The SMTP synchronization check was not operating right at the start of an
4068 SMTP session. For example, it could not catch a HELO sent before the client
4069 waited for the greeting. There is now a check for outstanding input at the
4070 point when the greeting is written. Because of the duplex, asynchronous
4071 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4072 way, but not yet received, when the check is performed.
4074 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4075 on TCP/IP sockets, because this apparently causes some broken clients to
4078 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4079 unchanged) from the Cygwin maintainer.
4081 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4082 was supported instead of testing for mbx. Effectively a typo.
4084 15. The spa authenticator server code was not checking that the input it
4085 received was valid base64.
4087 16. The debug output line for the "set" modifier in ACLs was not showing the
4088 name of the variable that was being set.
4090 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4092 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4093 lookup on its own. Something else was needed to trigger the lookup. For
4094 example, a match in host_lookup or the need for a host name in a host list.
4095 Now, if $sender_host_name is referenced and the host name has not yet been
4096 looked up, a lookup is performed. If the lookup fails, the variable remains
4097 empty, and $host_lookup_failed is set to "1".
4099 19. Added "eqi" as a case-independent comparison operator.
4101 20. The saslauthd authentication condition could segfault if neither service
4102 nor realm was specified.
4104 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4105 error message that was logged was misleading, and incoming SMTP
4106 connections were dropped. The message is now more accurate, and temporary
4107 errors are given to SMTP connections.
4109 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4110 (except RSET) with a 421 error, until QUIT is received. However, it was
4111 failing to send a response to QUIT.
4113 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4114 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4115 ACL is now run after the helo_try_verify_hosts code.
4117 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4118 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4119 case-independent, so other case variants are also recognized). Apparently
4120 some systems use these upper case variants.
4122 25. If more than two messages were waiting for the same host, and a transport
4123 filter was specified for the transport, Exim sent two messages over the
4124 same TCP/IP connection, and then failed with "socket operation on non-
4125 socket" when it tried to send the third.
4127 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4129 27. The extern definition of crypt16() in expand.c was not being excluded when
4130 the OS had its own crypt16() function.
4132 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4133 as a preferred synonym for return_size_limit, both as an option and as an
4136 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4138 30. Changed the default configuration ACL to relax the local part checking rule
4139 for addresses that are not in any local domains. For these addresses,
4140 slashes and pipe symbols are allowed within local parts, but the sequence
4141 /../ is explicitly forbidden.
4143 31. SPA server authentication was not clearing the challenge buffer before
4146 32. log_message in a "warn" ACL statement was writing to the reject log as
4147 well as to the main log, which contradicts the documentation and doesn't
4148 seem right (because no rejection is happening). So I have stopped it.
4150 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4151 However, I am unable to do any testing of this.
4153 34. Fixed an infelicity in the appendfile transport. When checking directories
4154 for a mailbox, to see if any needed to be created, it was accidentally
4155 using path names with one or more superfluous leading slashes; tracing
4156 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4158 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4159 contents are added to the log line that is written for every discarded
4160 recipient. (Previously a log_message setting was ignored.)
4162 36. The ${quote: operator now quotes the string if it is empty.
4164 37. The install script runs exim in order to find its version number. If for
4165 some reason other than non-existence or emptiness, which it checks, it
4166 could not run './exim', it was installing it with an empty version number,
4167 i.e. as "exim-". This error state is now caught, and the installation is
4170 38. An argument was missing from the function that creates an error message
4171 when Exim fails to connect to the socket for saslauthd authentication.
4172 This could cause Exim to crash, or give a corrupted message.
4174 39. Added isip, isip4, and isip6 to ${if conditions.
4176 40. The ACL variables $acl_xx are now saved with the message, and can be
4177 accessed later in routers, transports, and filters.
4179 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4180 strings in the file are not string-expanded.
4182 42. If a MAIL command specified a SIZE value that was too large to fit into an
4183 int variable, the check against message_size_limit failed. Such values are
4184 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4185 day this will have to be increased, but I don't think I want to be around
4186 when emails are that large.
4193 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4194 iconv() is not standard in FreeBSD.
4196 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4197 IPv6 enabled. The observed symptom was a segmentation fault on return from
4198 the function os_common_find_running_interfaces() in src/os.c.
4200 3. In the check_special_case() function in daemon.c I had used "errno" as an
4201 argument name, which causes warnings on some systems. This was basically a
4202 typo, since it was named "eno" in the comments!
4204 4. The code that waits for the clock to tick (at a resolution of some fraction
4205 of a second) so as to ensure message-id uniqueness was always waiting for
4206 at least one whole tick, when it could have waited for less. [This is
4207 almost certainly not relevant at current processor speeds, where it is
4208 unlikely to ever wait at all. But we try to future-proof.]
4210 5. The function that sleeps for a time interval that includes fractions of a
4211 second contained a race. It did not block SIGALRM between setting the
4212 timer, and suspending (a couple of lines later). If the interval was short
4213 and the sigsuspend() was delayed until after it had expired, the suspension
4214 never ended. On busy systems this could lead to processes getting stuck for
4217 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4218 process, before it forks any delivery processes. The open lookup caching
4219 mechanism meant that the open file or database connection was passed into
4220 the delivery process. The problem was that delivery processes always tidy
4221 up cached lookup data. This could cause a problem for the next delivery
4222 process started by the queue runner, because the external queue runner
4223 process does not know about the closure. So the next delivery process
4224 still has data in the lookup cache. In the case of a file lookup, there was
4225 no problem because closing a file descriptor in a subprocess doesn't affect
4226 the parent. However, if the lookup was caching a connection to a database,
4227 the connection was closed, and the second delivery process was likely to
4228 see errors such as "PGSQL: query failed: server closed the connection
4229 unexpectedly". The problem has been fixed by closing all cached lookups
4230 in a queue runner before running a delivery process.
4232 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4233 seem to have the "const" qualifier which it has on other OS. I've
4236 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4237 *of the same type* (client or server) with the same public name that an
4238 error should be diagnosed.
4240 9. When Exim looked up a host name for an IP address, but failed to find the
4241 original IP address when looking up the host name (a safety check), it
4242 output the message "<ip address> does not match any IP for NULL", which was
4243 confusing, to say the least. The bug was that the host name should have
4244 appeared instead of "NULL".
4246 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4247 user that is built into the binary, and the -C or -D options is used, root
4248 privilege is dropped before the configuration file is read. In addition,
4249 logging is switched to stderr instead of the normal log files. If the
4250 configuration then re-defines the Exim user, the unprivileged environment
4251 is probably not what is expected, so Exim logs a panic warning message (but
4254 However, if deliver_drop_privilege is set, the unprivileged state may well
4255 be exactly what is intended, so the warning has been cut out in that case,
4256 and Exim is allowed to try to write to its normal log files.
4262 1. smtp_return_error_details was not giving details for temporary sender
4263 or receiver verification errors.
4265 2. Diagnose a configuration error if two authenticators have the same public
4268 3. Exim used not to create the message log file for a message until the first
4269 delivery attempt. This could be confusing when incoming messages were held
4270 for policy or load reasons. The message log file is now created at the time
4271 the message is received, and an initial "Received" line is written to it.
4273 4. The automatically generated man page for command line options had a minor
4274 bug that caused no ill effects; however, a more serious problem was that
4275 the procedure for building the man page automatically didn't always
4276 operate. Consequently, release 4.20 contains an out-of-date version. This
4277 shouldn't happen again.
4279 5. When building Exim with embedded Perl support, the script that builds the
4280 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4281 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4283 6. The freeze_tell option was not being used for messages that were frozen on
4284 arrival, either by an ACL or by local_scan().
4286 7. Added the smtp_incomplete_transaction log selector.
4288 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4289 was accepting AUTH without a new EHLO.
4291 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4292 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4295 10. Logging of TCP/IP connections (when configured) now happens in the main
4296 daemon process instead of the child process, so that the TCP/IP connection
4297 count is more accurate (but it can never be perfect).
4299 11. The use of "drop" in a nested ACL was not being handled correctly in the
4300 outer ACL. Now, if condition failure induced by the nested "drop" causes
4301 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4302 or "require"), the connection is dropped.
4304 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4305 that yield "discard" can now be used with an "accept" or a "discard" verb,
4306 but an error is generated for any others (because I can't see a useful way
4307 to define what should happen).
4309 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4310 are now processed in the same way as lines in the Exim configuration file.
4311 In particular, continuation lines are supported.
4313 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4315 15. Added -ti meaning -t -i.
4317 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4318 domains, and warn by logging if others are found.
4320 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4321 returned from GIFCONF when Exim is trying to find the list of interfaces on
4322 a host. The code in os.c has been modified to copy each ifreq to an aligned
4323 structure in all cases.
4325 Also, in some cases, the returned ifreq's were being copied to a 'struct
4326 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4327 means the last couple of bytes of an IPv6 address could be chopped if the
4328 ifreq contained only a normal sockaddr (14 bytes storage).
4330 18. Named domain lists were not supported in the hosts_treat_as_local option.
4331 An entry such as +xxxx was not recognized, and was treated as a literal
4334 19. Ensure that header lines added by a DATA ACL are included in the reject log
4335 if the ACL subsequently rejects the message.
4337 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4338 MD5 (which is deprecated).
4340 21. When testing a filter file using -bf, Exim was writing a message when it
4341 took the sender from a "From " line in the message, but it was not doing so
4342 when it took $return_path from a Return-Path: header line. It now does.
4344 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4345 with a valid header line field name (a series of printing characters
4346 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4348 23. Changed "disc" in the source to "disk" to conform to the documentation and
4349 the book and for uniformity.
4351 24. Ignore Sendmail's -Ooption=value command line item.
4353 25. When execve() failed while trying to run a command in a pipe transport,
4354 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4355 could be confused with a return value of 69 from the command itself. This
4356 has been changed to 127, the value the shell returns if it is asked to run
4357 a non-existent command. The wording for the related log line suggests a
4358 non-existent command as the problem.
4360 26. If received_header_text expands to an empty string, do not add a Received:
4361 header line to the message. (Well, it adds a token one on the spool, but
4362 marks it "old" so that it doesn't get used or transmitted.)
4364 27. Installed eximstats 1.28 (addition of -nt option).
4366 28. There was no check for failure on the call to getsockname() in the daemon
4367 code. This can fail if there is a shortage of resources on the system, with
4368 ENOMEM, for example. A temporary error is now given on failure.
4370 29. Contrary to the C standard, it seems that in some environments, the
4371 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4372 program. Exim now does this explicitly; it affects the formatting of
4373 timestamps using strftime().
4375 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4376 complaints. I've now initialized all the variables, to avoid this.
4378 32. Header lines added by a system filter were not being "seen" during
4379 transport-time rewrites.
4381 33. The info_callback() function passed to OpenSSL is set up with type void
4382 (*)(SSL *, int, int), as described somewhere. However, when calling the
4383 function (actually a macro) that sets it up, the type void(*)() is
4384 expected. I've put in a cast to prevent warnings from picky compilers.
4386 34. If a DNS black list lookup found a CNAME record, but there were no A
4387 records associated with the domain it pointed at, Exim crashed.
4389 35. If a DNS black list lookup returned more than one A record, Exim ignored
4390 all but the first. It now scans all returned addresses if a particular IP
4391 value is being sought. In this situation, the contents of the
4392 $dnslist_value variable are a list of all the addresses, separated by a
4395 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4396 to accept a host name and all its aliases if the forward lookup for any of
4397 them yielded the IP address of the incoming connection. Now it accepts only
4398 those names whose forward lookup yields the correct IP address. Any other
4399 names are discarded. This closes a loophole whereby a rogue DNS
4400 administrator could create reverse DNS records to break through a
4401 wildcarded host restriction in an ACL.
4403 37. If a user filter or a system filter that ran in a subprocess used any of
4404 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4405 the wrong values were passed to the pipe command ($thisaddress had the
4406 value of $0, $0 had the value of $1, etc). This bug was introduced by
4407 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4409 38. Improved the line breaking for long SMTP error messages from ACLs.
4410 Previously, if there was no break point between 40 and 75 characters, Exim
4411 left the rest of the message alone. Two changes have been made: (a) I've
4412 reduced the minimum length to 35 characters; (b) if it can't find a break
4413 point between 35 and 75 characters, it looks ahead and uses the first one
4414 that it finds. This may give the occasional overlong line, but at least the
4415 remaining text gets split now.
4417 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4418 file descriptors might be low, and that setting 1000 would always raise it.
4419 It turns out that in some environments, the limit is already over 1000 and
4420 that lowering it causes trouble. So now Exim takes care not to decrease it.
4422 40. When delivering a message, the value of $return_path is set to $sender_
4423 address at the start of routing (routers may change the value). By an
4424 oversight, this default was not being set up when an address was tested by
4425 -bt or -bv, which affected the outcome if any router or filter referred to
4428 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4429 log only when "message" or "log_message" are set. However, if one of the
4430 conditions was an address verification, or a call to a nested ACL, the
4431 messages generated by the underlying test were being passed through. This
4432 no longer happens. The underlying message is available in $acl_verify_
4433 message for both "message" and "log_message" expansions, so it can be
4434 passed through if needed.
4436 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4437 new expansion $bh_ to give the encoded byte string without charset
4438 translation. Translation happens only if iconv() is available; HAVE_ICONV
4439 indicates this at build time. HEADERS_CHARSET gives the charset to
4440 translate to; headers_charset can change it in the configuration, and
4441 "headers charset" can change it in an individual filter file.
4443 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4444 that creates RFC 2047 encoded "words" labels them as that charset instead
4445 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4446 expansion operator; (ii) when Exim creates a From: line for a local
4447 message; (iii) when a header line is rewritten to include a "phrase" part.
4449 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4450 buggy, causing it to skip the first lines of messages whose message ID
4451 ended in 'D'. This would not have bitten before Exim release 4.14, because
4452 message IDs were unlikely to end in 'D' before then. The effect was to have
4453 incorrect size information for certain domains.
4455 45. #include "config.h" was missing at the start of the crypt16.c module. This
4456 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4459 46. If there was a timeout during a "random" callout check, Exim treated it as
4460 a failure of the random address, and carried on sending RSET and the real
4461 address. If the delay was just some slowness somewhere, the response to the
4462 original RCPT would be taken as a response to RSET and so on, causing
4463 mayhem of various kinds.
4465 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4466 when I implemented it. It didn't allow for the fact that some option values
4467 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4468 the right test for positive values.
4470 48. Domain names in DNS records are case-independent. Exim always looks them up
4471 in lower case. Some resolvers return domain names in exactly the case they
4472 appear in the zone file, that is, they may contain uppercase letters. Not
4473 all resolvers do this - some return always lower case. Exim was treating a
4474 change of case by a resolver as a change of domain, similar to a widening
4475 of a domain abbreviation. This triggered its re-routing code and so it was
4476 trying to route what was effectively the same domain again. This normally
4477 caused routing to fail (because the router wouldn't handle the domain
4478 twice). Now Exim checks for this case specially, and just changes the
4479 casing of the domain that it ultimately uses when it transmits the message
4482 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4485 50. If a filter generated a file delivery with a non-absolute name (possible if
4486 no home directory exists for the router), the forbid_file option was not
4489 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4490 the existing equality matching.
4492 52. Exim was using ints instead of ino_t variables in some places where it was
4493 dealing with inode numbers.
4495 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4496 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4497 called TMPDIR, and if it finds it is different, it changes its value.
4499 54. The smtp_printf() function is now made available to local_scan() so
4500 additional output lines can be written before returning. There is also an
4501 smtp_fflush() function to enable the detection of a dropped connection.
4502 The variables smtp_input and smtp_batched_input are exported to
4505 55. Changed the default runtime configuration: the message "Unknown user"
4506 has been removed from the ACL, and instead placed on the localuser router,
4507 using the cannot_route_message feature. This means that any verification
4508 failures that generate their own messages won't get overridden. Similarly,
4509 the "Unrouteable address" message that was in the ACL for unverifiable
4510 relay addresses has also been removed.
4512 56. Added hosts_avoid_esmtp to the smtp transport.
4514 57. The exicyclog script was not checking for the esoteric option
4515 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4516 will work only if exicyclog is run under the appropriate euid.
4518 58. Following a discussion on the list, the rules by which Exim recognises line
4519 endings on incoming messages have been changed. The -dropcr and drop_cr
4520 options are now no-ops, retained only for backwards compatibility. The
4521 following line terminators are recognized: LF CRLF CR. However, special
4522 processing applies to CR:
4524 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4525 nor a local message in the state where . is a terminator.
4527 (ii) If a bare CR is encountered in a header line, an extra space is added
4528 after the line terminator so as not to end the header. The reasoning
4529 behind this is that bare CRs in header lines are most likely either
4530 to be mistakes, or people trying to play silly games.
4532 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4533 was being incorrectly given as 18 bytes larger than it should have been.
4534 This is a VOB (very old bug).
4536 60. This may never have affected anything current, but just in case it has:
4537 When the local host is found other than at the start of a list of hosts,
4538 the local host, those with the same MX, and any that follow, are discarded.
4539 When the list in question was part of a longer list of hosts, the following
4540 hosts (not currently being processed) were also being discarded. This no
4541 longer happens. I'm not sure if this situation could ever has previously
4544 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4547 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4548 header. This is something that is recommended in a new Internet Draft, and
4549 is something that is documented as being done by Sendmail. There are two
4550 possible values. For messages generated by the autoreply transport, Exim
4553 Auto-Submitted: auto-replied
4555 whereas for all other generated messages (e.g. bounces) it adds
4557 Auto-Submitted: auto-generated
4559 63. The "personal" condition in filters now includes a test for the
4560 Auto-Submitted: header. If it contains the string "auto-" the message it
4561 not considered personal.
4563 64. Added rcpt_include_affixes as a generic transport option.
4565 65. Added queue_only_override (default true).
4567 66. Added the syslog_duplication option.
4569 67. If what should have been the first header line of a message consisted of
4570 a space followed by a colon, Exim was mis-interpreting it as a header line.
4571 It isn't of course - it is syntactically invalid and should therefore be
4572 treated as the start of the message body. The misbehaviour could have
4573 caused a number of strange effects, including loss of data in subsequent
4574 header lines, and spool format errors.
4576 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4577 client host had authenticated. This control can now be exercised by an ACL
4578 for more flexibility.
4580 69. By default, callouts do not happen when testing with -bh. There is now a
4581 variant, -bhc, which does actually run the callout code, including
4582 consulting and updating the callout cache.
4584 70. Added support for saslauthd authentication, courtesy of Alexander
4587 71. If statvfs() failed on the spool or log directories while checking their
4588 size for availability, Exim confusingly gave the error "space shortage".
4589 Furthermore, in debugging mode it crashed with a floating point exception.
4590 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4591 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4592 serious problem, Exim now writes to the main and panic logs when this
4593 happens, with details of the failure. It then refuses to accept the
4594 incoming message, giving the message "spool directory problem" or "log
4595 directory problem" with a 421 code for SMTP messages.
4597 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4598 0, 1, and 2 exist, because some OS complain for execs without them (see
4599 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4600 descriptors. However, the code omitted to check that the open succeeded,
4601 causing mysterious errors if for some reason the permissions on /dev/null
4602 got screwed. Now Exim writes a message to the main and panic logs, and
4603 bombs out if it can't open /dev/null.
4605 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4606 interact so that it is all more flexible. It is supposed to remain
4607 backwards compatible. Also added extra_local_interfaces.
4609 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4610 to bomb out with an assertion failure - to the client this appears as a
4611 connection drop. This problem occurs in the part of the code that was taken
4612 from the Samba project. Fortunately, the assertion is in a very simple
4613 function, so I have fixed this by reproducing the function inline in the
4614 one place where it is called, and arranging for authentication to fail
4615 instead of killing the process with assert().
4617 75. The SPA client code was not working when the server requested OEM rather
4618 than Unicode encoding.
4620 76. Added code to make require_files with a specific uid setting more usable in
4621 the case where statting the file as root fails - usually a non-root-mounted
4622 NFS file system. When this happens and the failure is EACCES, Exim now
4623 forks a subprocess and does the per-uid checking as the relevant uid.
4625 77. Added process_log_path.
4627 78. If log_file_path was not explicitly set, a setting of check_log_space or
4628 check_log_inodes was ignored.
4630 79. If a space check for the spool or log partitions fails, the incident is now
4631 logged. Of course, in the latter case the data may get lost...
4633 80. Added the %p formatting code to string_format() so that it can be used to
4634 print addresses in debug_print(). Adjusted all the address printing in the
4635 debugging in store.c to use %p rather than %d.
4637 81. There was a concern that a line of code in smtp_in.c could overflow a
4638 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4639 initially expressed, the concern was not well-founded, because trailing
4640 spaces are removed early. However, if the trailing spaces were followed by
4641 a NULL, they did not get removed, so the overflow was possible. Two fixes
4644 (a) I re-wrote the offending code in a cleaner fashion.
4645 (b) If an incoming SMTP command contains a NULL character, it is rejected
4648 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4649 runs initgroups(), so that if the Exim user is in any additional groups,
4650 they will be used during message reception.
4656 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob