1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.487 2007/03/01 14:06:56 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.
140 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
141 instead of the C I/O that was originally supplied, because problems were
148 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
149 fixed by 4.65/MH/01 (is this a record?) are fixed:
151 (i) An empty string was always treated as zero by the numeric comparison
152 operators. This behaviour has been restored.
154 (ii) It is documented that the numeric comparison operators always treat
155 their arguments as decimal numbers. This was broken in that numbers
156 starting with 0 were being interpreted as octal.
158 While fixing these problems I realized that there was another issue that
159 hadn't been noticed. Values of message_size_limit (both the global option
160 and the transport option) were treated as octal if they started with 0.
161 The documentation was vague. These values are now always treated as
162 decimal, and I will make that clear in the documentation.
168 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
169 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
172 MH/01 Don't check that the operands of numeric comparison operators are
173 integers when their expansion is in "skipping" mode (fixes bug
174 introduced by 4.64-PH/07).
176 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
177 child addresses, Exim now panics and dies. Previously, because the count
178 is held in a short int, deliveries were likely to be lost. As such a
179 large number of recipients for a single message is ridiculous
180 (performance will be very, very poor), I have chosen to impose a limit
181 rather than extend the field.
187 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
188 leftover -K file (the existence of which was triggered by #402).
189 While we were at it, introduced process PID as part of the -K
190 filename. This should rule out race conditions when creating
193 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
194 processing considerably. Previous code took too long for large mails,
195 triggering a timeout which in turn triggers #401.
197 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
198 in the DK code in transports.c. sendfile() is not really portable,
199 hence the _LINUX specificness.
201 TF/01 In the add_headers option to the mail command in an Exim filter,
202 there was a bug that Exim would claim a syntax error in any
203 header after the first one which had an odd number of characters
206 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
207 callout verification, Exim cached a "reject" for the entire domain. This
208 is correct for most verifications, but it is not correct for a recipient
209 verification with use_sender or use_postmaster set, because in that case
210 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
211 case of MAIL FROM:<> rejection from other early rejections (e.g.
212 rejection of HELO). When verifying a recipient using a non-null MAIL
213 address, the cache is ignored if it shows MAIL FROM:<> rejection.
214 Whatever the result of the callout, the value of the domain cache is
215 left unchanged (for any other kind of callout, getting as far as trying
216 RCPT means that the domain itself is ok).
218 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
221 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
222 manifest itself as EPIPE rather than ECONNECT. When tidying away a
223 session, the daemon ignores ECONNECT errors and logs others; it now
224 ignores EPIPE as well.
226 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
227 (quoted-printable decoding).
229 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
230 later the small subsequent patch to fix an introduced bug.
232 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
234 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
236 PH/08 An error is now given if message_size_limit is specified negative.
238 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
239 to be given (somewhat) arbitrary names.
241 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
244 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
245 miscellaneous code fixes
247 PH/10 Added the log_reject_target ACL modifier to specify where to log
250 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
251 hostname. This is wrong, because it relates to the incoming message (and
252 probably the interface on which it is arriving) and not to the outgoing
253 callout (which could be using a different interface). This has been
254 changed to use the value of the helo_data option from the smtp transport
255 instead - this is what is used when a message is actually being sent. If
256 there is no remote transport (possible with a router that sets up host
257 addresses), $smtp_active_hostname is used.
259 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
260 tweaks were necessary in order to get it to work (see also 21 below):
261 (a) The code assumed that strncpy() returns a negative number on buffer
262 overflow, which isn't the case. Replaced with Exim's string_format()
264 (b) There were several signed/unsigned issues. I just did the minimum
265 hacking in of casts. There is scope for a larger refactoring.
266 (c) The code used strcasecmp() which is not a standard C function.
267 Replaced with Exim's strcmpic() function.
268 (d) The code set only $1; it now sets $auth1 as well.
269 (e) A simple test gave the error "authentication client didn't specify
270 service in request". It would seem that Dovecot has changed its
271 interface. Fortunately there's a specification; I followed it and
272 changed what the client sends and it appears to be working now.
274 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
277 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
278 address A is aliased to B and C, where B exists and C does not. Without
279 -v the output is "A verified" because verification stops after a
280 successful redirection if more than one address is generated. However,
281 with -v the child addresses are also verified. Exim was outputting "A
282 failed to verify" and then showing the successful verification for C,
283 with its parentage. It now outputs "B failed to verify", showing B's
284 parentage before showing the successful verification of C.
286 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
287 look up a TXT record in a specific list after matching in a combined
290 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
291 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
292 they consult the DNS. I had assumed they would set it the way they
293 wanted; and indeed my experiments on Linux seem to show that in some
294 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
295 To be on the safe side, however, I have now made the interface to
296 host_find_byname() similar to host_find_bydns(), with an argument
297 containing the DNS resolver options. The host_find_byname() function now
298 sets these options at its start, just as host_find_bydns() does. The smtp
299 transport options dns_qualify_single and dns_search_parents are passed to
300 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
301 of host_find_byname() use the default settings of RES_DEFNAMES
302 (qualify_single) but not RES_DNSRCH (search_parents).
304 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
305 spool_read_header() do less string testing, by means of a preliminary
306 switch on the second character of optional "-foo" lines. (This is
307 overdue, caused by the large number of possibilities that now exist.
308 Originally there were few.) While I was there, I also converted the
309 str(n)cmp tests so they don't re-test the leading "-" and the first
310 character, in the hope this might squeeze out yet more improvement.
312 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
313 flag allowing group syntax was set by the header_syntax check but not
314 turned off, possible causing trouble later; (2) The flag was not being
315 set at all for the header_verify test, causing "group"-style headers to
316 be rejected. I have now set it in this case, and also caused header_
317 verify to ignore an empty address taken from a group. While doing this, I
318 came across some other cases where the code for allowing group syntax
319 while scanning a header line wasn't quite right (mostly, not resetting
320 the flag correctly in the right place). These bugs could have caused
321 trouble for malformed header lines. I hope it is now all correct.
323 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
324 with the "reply" argument non-NULL. The code, however (which originally
325 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
326 but it didn't always do it. This confused somebody who was copying the
327 code for some other use. I have removed all the tests.
329 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
330 feature that was used to support insecure browsers during the U.S. crypto
331 embargo. It requires special client support, and Exim is probably the
332 only MTA that supported it -- and would never use it because real RSA is
333 always available. This code has been removed, because it had the bad
334 effect of slowing Exim down by computing (never used) parameters for the
335 RSA_EXPORT functionality.
337 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
338 authenticator to fail if there's a tab character in the incoming data
339 (there should never be unless someone is messing about, as it's supposed
340 to be base64-encoded). Also added, on Timo's advice, the "secured" option
341 if the connection is using TLS or if the remote IP is the same as the
342 local IP, and the "valid-client-cert option" if a client certificate has
345 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
346 authenticators. This can be used for authorization after authentication
347 succeeds. (In the case of plaintext, it servers for both authentication
350 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
351 if any retry times were supplied.
353 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
354 connection, where there is no client IP address to check. In this
355 situation, the verify now always succeeds.
357 PH/25 Applied John Jetmore's -Mset patch.
359 PH/26 Added -bem to be like -Mset, but loading a message from a file.
361 PH/27 In a string expansion for a processed (not raw) header when multiple
362 headers of the same name were present, leading whitespace was being
363 removed from all of them, but trailing whitespace was being removed only
364 from the last one. Now trailing whitespace is removed from each header
365 before concatenation. Completely empty headers in a concatenation (as
368 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
369 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
371 PH/29 [Removed. This was a change that I later backed out, and forgot to
372 correct the ChangeLog entry (that I had efficiently created) before
373 committing the later change.]
375 PH/30 Exim was sometimes attempting to deliver messages that had suffered
376 address errors (4xx response to RCPT) over the same connection as other
377 messages routed to the same hosts. Such deliveries are always "forced",
378 so retry times are not inspected. This resulted in far too many retries
379 for the affected addresses. The effect occurred only when there were more
380 hosts than the hosts_max_try setting in the smtp transport when it had
381 the 4xx errors. Those hosts that it had tried were not added to the list
382 of hosts for which the message was waiting, so if all were tried, there
383 was no problem. Two fixes have been applied:
385 (i) If there are any address or message errors in an SMTP delivery, none
386 of the hosts (tried or untried) are now added to the list of hosts
387 for which the message is waiting, so the message should not be a
388 candidate for sending over the same connection that was used for a
389 successful delivery of some other message. This seems entirely
390 reasonable: after all the message is NOT "waiting for some host".
391 This is so "obvious" that I'm not sure why it wasn't done
392 previously. Hope I haven't missed anything, but it can't do any
393 harm, as the worst effect is to miss an optimization.
395 (ii) If, despite (i), such a delivery is accidentally attempted, the
396 routing retry time is respected, so at least it doesn't keep
397 hammering the server.
399 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
400 in ${readsocket because some servers need this prod.
402 PH/32 Added some extra debug output when updating a wait-xxx database.
404 PH/33 The hint "could be header name not terminated by colon", which has been
405 given for certain expansion errors for a long time, was not being given
406 for the ${if def:h_colon_omitted{... case.
408 PH/34 The spec says: "With one important exception, whenever a domain list is
409 being scanned, $domain contains the subject domain." There was at least
410 one case where this was not true.
412 PH/35 The error "getsockname() failed: connection reset by peer" was being
413 written to the panic log as well as the main log, but it isn't really
414 panic-worthy as it just means the connection died rather early on. I have
415 removed the panic log writing for the ECONNRESET error when getsockname()
418 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
419 runs only) independently of the message's sender address. This meant
420 that, if the 4xx error was in fact related to the sender, a different
421 message to the same recipient with a different sender could confuse
422 things. In particualar, this can happen when sending to a greylisting
423 server, but other circumstances could also provoke similar problems.
424 I have changed the default so that the retry time for these errors is now
425 based a combination of the sender and recipient addresses. This change
426 can be overridden by setting address_retry_include_sender=false in the
429 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
430 remote server are returned as part of bounce messages. This was not
431 happening for LMTP over a pipe (the lmtp transport), but now it is the
432 same for both kinds of LMTP.
434 PH/38 Despite being documented as not happening, Exim was rewriting addresses
435 in header lines that were in fact CNAMEs. This is no longer the case.
437 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
438 and queue runs started by the daemon processed all messages. This has
439 been fixed so that -R and -S can now usefully be given with -q<time>.
441 PH/40 Import PCRE release 6.7 (fixes some bugs).
443 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
445 PH/42 Give an error if -q is specified more than once.
447 PH/43 Renamed the variables $interface_address and $interface_port as
448 $received_ip_address and $received_port, to make it clear that these
449 values apply to message reception, and not to the outgoing interface when
450 a message is delivered. (The old names remain recognized, of course.)
452 PH/44 There was no timeout on the connect() call when using a Unix domain
453 socket in the ${readsocket expansion. There now is.
455 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
456 be meaningful with "accept".
458 SC/01 Eximstats V1.43
459 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
461 SC/02 Eximstats V1.44
462 Use a glob alias rather than an array ref in the generated
463 parser. This improves both readability and performance.
465 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
466 Collect SpamAssassin and rejection statistics.
467 Don't display local sender or destination tables unless
468 there is data to show.
469 Added average volumes into the top table text output.
471 SC/04 Eximstats V1.46
472 Collect data on the number of addresses (recipients)
473 as well as the number of messages.
475 SC/05 Eximstats V1.47
476 Added 'Message too big' to the list of mail rejection
477 reasons (thanks to Marco Gaiarin).
479 SC/06 Eximstats V1.48
480 Mainlog lines which have GMT offsets and are too short to
481 have a flag are now skipped.
483 SC/07 Eximstats V1.49 (Alain Williams)
484 Added the -emptyok flag.
486 SC/08 Eximstats V1.50
487 Fixes for obtaining the IP address from reject messages.
489 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
490 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
491 whitesspace changes from 4.64-PH/27
493 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
496 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
497 are found, allow negative numbers in numeric criteria)
499 JJ/06 exipick.20061117.2, added new $message_body_missing variable
501 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
502 to match changes made in 4.64-PH/43
504 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
506 PH/47 Put in an explicit test for a DNS lookup of an address record where the
507 "domain" is actually an IP address, and force a failure. This locks out
508 those revolvers/nameservers that support "A-for-A" lookups, in
509 contravention of the specifications.
511 PH/48 When a host name was looked up from an IP address, and the subsequent
512 forward lookup of the name timed out, the host name was left in
513 $sender_host_name, contrary to the specification.
515 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
516 restricted to single-key lookups, Exim was not diagnosing an error if
517 * or *@ was used with a query-style lookup.
519 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
521 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
522 long ago, but noone interested enough thought of it. Let's just say that
523 the "1.1" means that there are some new functions that weren't there at
524 some point in the past.
526 PH/51 Error processing for expansion failure of helo_data from an smtp
527 transport during callout processing was broken.
529 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
530 tested/used via the -bh/-bhc/-bs options.
532 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
533 bug, fixed in subsequent PCRE releases).
535 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
536 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
538 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
544 SC/01 Use a glob alias rather than an array ref in eximstats generated
545 parser. This improves both readability and performance.
547 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
548 Don't display local sender or destination tables in eximstats unless
549 there is data to show.
550 Added average volumes into the eximstats top table text output.
552 SC/03 Collect data on the number of addresses (recipients) as well
553 as the number of messages in eximstats.
555 TF/01 Correct an error in the documentation for the redirect router. Exim
556 does (usually) call initgroups() when daemonizing.
558 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
559 with consistent privilege compared to when running as a daemon.
561 TF/03 Note in the spec that $authenticated_id is not set for local
562 submissions from trusted users.
564 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
565 Thanks to Dean Brooks <dean@iglou.com> for the patch.
567 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
568 by adding some example configuration directives to the default
569 configuration file. A little bit of work is required to uncomment the
570 directives and define how usernames and passwords are checked, but
571 there is now a framework to start from.
573 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
574 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
575 without this. I don't know how relevant this is to other LDAP libraries.
577 PH/02 Add the verb name to the "unknown ACL verb" error.
579 PH/03 Magnus Holmgren's patch for filter_prepend_home.
581 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
583 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
584 directory not expanded when it should be if an expanded home directory
585 was set for the address (which is overridden by the transport).
587 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
590 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
591 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
592 because it is too late at that time, and has no effect.
594 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
595 security issue with \' (bugzilla #107). I could not use the
596 PQescapeStringConn() function, because it needs a PGconn value as one of
599 PH/08 When testing addresses using -bt, indicate those final addresses that
600 are duplicates that would not cause an additional delivery. At least one
601 person was confused, thinking that -bt output corresponded to deliveries.
602 (Suppressing duplicates isn't a good idea as you lose the information
603 about possibly different redirections that led to the duplicates.)
605 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
606 systems where poll() doesn't work, in particular OS X.
608 PH/10 Added more information to debugging output for retry time not reached.
610 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
611 operations in malware.c.
613 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
616 PH/13 If write_rejectlog was set false when logging was sent to syslog with
617 syslog_duplication set false, log lines that would normally be written
618 both the the main log and to the reject log were not written to syslog at
621 PH/14 In the default configuration, change the use of "message" in ACL warn
622 statements to "add_header".
624 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
625 not followed by a command (e.g. "seen endif").
627 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
628 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
631 PH/17 Added extra conditions to the default value of delay_warning_condition
635 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
636 { match{$h_precedence:}{(?i)bulk|list|junk} } \
637 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
640 The Auto-Submitted: and various List- headers are standardised, whereas I
641 don't think Precedence: ever was.
643 PH/18 Refactored debugging code in route_finduser() to show more information,
644 in particular, the error code if getpwnam() issues one.
646 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
647 This is apparently needed in addition to the PH/07 change above to avoid
648 any possible encoding problems.
650 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
651 but not after initializing Perl.
653 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
654 output them only if debugging. By default they are written stderr,
655 apparently, which is not desirable.
657 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
660 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
663 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
665 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
666 authenticated or an ident call has been made. Suppress the default
667 values for $authenticated_id and $authenticated_sender (but permit -oMai
668 and -oMas) when testing with -bh.
670 PH/24 Re-jigged the order of the tests in the default configuration so that the
671 tests for valid domains and recipients precede the DNS black list and CSA
672 tests, on the grounds that those ones are more expensive.
674 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
675 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
676 This bug exists in every version of Exim that I still have, right back to
679 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
680 However, an attempt to turn on case-sensitivity in a regex key by
681 including (?-i) didn't work because the subject string was already
682 lowercased, and the effects were non-intuitive. It turns out that a
683 one-line patch can be used to allow (?-i) to work as expected.
689 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
690 other effects) broke the use of negated acl sub-conditions.
692 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
695 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
696 "Deny" causes Exim to reject the incoming connection with a 554 error.
697 Unfortunately, if there is a major crisis, such as a disk failure,
698 tcp-wrappers gives "deny", whereas what one would like would be some
699 kind of temporary error. A kludge has been added to help with this.
700 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
701 554 error is used if errno is still zero or contains ENOENT (which occurs
702 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
705 PH/03 Add -lutil to the default FreeBSD LIBS setting.
707 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
708 errors. Otherwise a message that provokes a temporary error (when other
709 messages do not) can cause a whole host to time out.
711 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
712 addresses were routed directly to files or pipes from a redirect router.
713 File deliveries just didn't batch; pipe deliveries might have suffered
716 PH/06 A failure to get a lock for a hints database would erroneously always say
717 "Failed to get write lock", even when it was really a read lock.
719 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
720 of 0600. This has been changed to use the value of the lockfile_mode
721 option (which defaults to 0600).
723 PH/08 Applied small patch from the Sieve maintainer.
725 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
726 folder from quota calculations, a direct delivery into this folder messed
727 up the contents of the maildirsize file. This was because the regex was
728 used only to exclude .Trash (or whatever) when the size of the mailbox
729 was calculated. There was no check that a delivery was happening into an
730 excluded directory. This bug has been fixed by ignoring all quota
731 processing for deliveries into excluded directories.
733 PH/10 Added the maildirfolder_create_regex option to appendfile.
739 PH/01 The code for finding all the local interface addresses on a FreeBSD
740 system running IPv6 was broken. This may well have applied to all BSD
741 systems, as well as to others that have similar system calls. The broken
742 code found IPv4 interfaces correctly, but gave incorrect values for the
743 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
744 that it would not match correctly against @[] and not recognize the IPv6
747 PH/02 The ipliteral router was not recognizing addresses of the form user@
748 [ipv6:....] because it didn't know about the "ipv6:" prefix.
750 PH/03 Added disable_ipv6.
752 PH/04 Changed $reply_address to use the raw form of the headers instead of the
753 decoded form, because it is most often used to construct To: headers
754 lines in autoreplies, and the decoded form may well be syntactically
755 invalid. However, $reply_address has leading white space removed, and all
756 newlines turned into spaces so that the autoreply transport does not
759 PH/05 If group was specified without a user on a router, and no group or user
760 was specified on a transport, the group from the router was ignored.
762 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
763 for visible compile-time settings that can be used to change these
764 numbers, for those that want even more. Backwards compatibility with old
765 spool files has been maintained. However, going back to a previous Exim
766 release will lost any variables that are in spool files.
768 PH/07 Two small changes when running in the test harness: increase delay when
769 passing a TCP/IP connection to a new process, in case the original
770 process has to generate a bounce, and remove special handling of
771 127.0.0.2 (sic), which is no longer necessary.
773 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
774 be the same on different OS.
776 PH/09 Moved a debug statement in filter processing to avoid a race problem when
779 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
780 whether --show-vars was specified or not
782 JJ/02 exipick: Added support for new ACL variable spool format introduced
785 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
786 syntactically invalid From: or Reply-to: line, and a filter used this to
787 generate an autoreply, and therefore failed to obtain an address for the
788 autoreply, Exim could try to deliver to a non-existent relative file
789 name, causing unrelated and misleading errors. What now happens is that
790 it logs this as a hard delivery error, but does not attempt to create a
793 PH/11 The exinext utility has a -C option for testing purposes, but although
794 the given file was scanned by exinext itself; it wasn't being passed on
795 when Exim was called.
797 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
798 an end-of-file indication when reading a command response.
800 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
801 compiled. In many other places in Exim, IPv6 addresses are always
802 recognized, so I have changed this. It also means that IPv4 domain
803 literals of the form [IPV4:n.n.n.n] are now always recognized.
805 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
806 used if the router is not running as root, for example, when verifying at
807 ACL time, or when using -bh. The debugging output from this situation was
808 non-existent - all you got was a failure to exec. I have made two
811 (a) Failures to set uid/gid, the current directory, or a process leader
812 in a subprocess such as that created by queryprogram now generate
813 suitable debugging ouput when -d is set.
815 (b) The queryprogram router detects when it is not running as root,
816 outputs suitable debugging information if -d is set, and then runs
817 the subprocess without attempting to change uid/gid.
819 PH/15 Minor change to Makefile for building test_host (undocumented testing
822 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
823 additional section of a DNS packet that returns MX or SRV records.
824 Instead, it always explicitly searches for A/AAAA records. This avoids
825 major problems that occur when a DNS server includes only records of one
826 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
827 fixed another bug: if SRV records were looked up and the corresponding
828 address records were *not* found in the additional section, the port
829 values from the SRV records were lost.
831 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
832 using the correct key (the original address) when searching the retry
833 rules in order to find which one to use for generating the retry hint.
835 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
836 adding the default one. Similarly, if it contains a Reply-To: header, the
837 errors_reply_to option, if set, is not used.
839 PH/19 When calculating a retry time, Exim used to measure the "time since
840 failure" by looking at the "first failed" field in the retry record. Now
841 it does not use this if it is later than than the arrival time of the
842 message. Instead it uses the arrival time. This makes for better
843 behaviour in cases where some deliveries succeed, thus re-setting the
844 "first failed" field. An example is a quota failure for a huge message
845 when small messages continue to be delivered. Without this change, the
846 "time since failure" will always be short, possible causing more frequent
847 delivery attempts for the huge message than are intended.
848 [Note: This change was subsequently modified - see PH/04 for 4.62.]
850 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
851 $1, $2, $3) because the numerical variables can be reset during some
852 expansion items (e.g. "match"), thereby losing the authentication data.
854 PH/21 Make -bV show the size of off_t variables so that the test suite can
855 decide whether to run tests for quotas > 2G.
857 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
858 mailbox_size, and mailbox_filecount in the appendfile transport. If a
859 filecount value is greater than 2G or if a quota value is greater than 2G
860 on a system where the size of off_t is not greater than 4, a panic error
863 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
864 never match. The debug and -bh output now contains an explicit error
865 message indicating a malformed IPv4 address or mask.
867 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
868 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
871 PH/25 Do not write to syslog when running in the test harness. The only
872 occasion when this arises is a failure to open the main or panic logs
873 (for which there is an explicit test).
875 PH/26 Added the /no_tell option to "control=freeze".
877 PH/27 If a host name lookup failed very early in a connection, for example, if
878 the IP address matched host_lookup and the reverse lookup yielded a name
879 that did not have a forward lookup, an error message of the form "no IP
880 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
881 could be logged. Now it outputs the IP address instead of "NULL".
883 PH/28 An enabling patch from MH: add new function child_open_exim2() which
884 allows the sender and the authenticated sender to be set when
885 submitting a message from within Exim. Since child_open_exim() is
886 documented for local_scan(), the new function should be too.
888 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
889 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
890 results in an empty string is now treated as unset.
892 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
894 PH/31 Added sender_verify_fail logging option.
896 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
897 needed by RFC 822 but not by RFC 2822 was commented out. I have now
898 tidied the source and removed it altogether.
900 PH/33 When a queue run was abandoned because the load average was too high, a
901 log line was always written; now it is written only if the queue_run log
902 selector is set. In addition, the log line for abandonment now contains
903 information about the queue run such as the pid. This is always present
904 in "start" and "stop" lines but was omitted from the "abandon" line.
906 PH/34 Omit spaces between a header name and the colon in the error message that
907 is given when verify = headers_syntax fails (if there are lots of them,
908 the message gets confusing).
910 PH/35 Change the default for dns_check_names_pattern to allow slashes within
911 names, as there are now some PTR records that contain slashes. This check
912 is only to protect against broken name servers that fall over on strange
913 characters, so the fact that it applies to all lookups doesn't matter.
915 PH/36 Now that the new test suite is complete, we can remove some of the
916 special code in Exim that was needed for the old test suite. For example,
917 sorting DNS records because real resolvers return them in an arbitrary
918 order. The new test suite's fake resolver always returns records in the
921 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
922 bounces) except when queue_only is set, to avoid logging races between
923 the different processes.
925 PH/38 Panic-die if .include specifies a non-absolute path.
927 PH/39 A tweak to the "H" retry rule from its user.
929 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
930 a label. They prevented compilation on older perls.
932 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
933 a warning to be raised on newish perls.
935 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
936 on queue. Changes to match documented behaviour of showing count of
937 messages matching specified criteria.
939 PH/40 Changed the default ident timeout from 30s to 5s.
941 PH/41 Added support for the use of login_cap features, on those BSD systems
942 that have them, for controlling the resources used by pipe deliveries.
944 PH/42 The content-scanning code uses fopen() to create files in which to put
945 message data. Previously it was not paying any attention to the mode of
946 the files. Exim runs with umask(0) because the rest of the code creates
947 files with open(), and sets the required mode explicitly. Thus, these
948 files were ending up world-writeable. This was not a big issue, because,
949 being within the spool directory, they were not world-accessible. I have
950 created a function called modefopen, which takes an additional mode
951 argument. It sets umask(777), creates the file, chmods it to the required
952 mode, then resets the umask. All the relevant calls to fopen() in the
953 content scanning code have been changed to use this function.
955 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
956 to 24 hours. This avoids potential overflow problems when processing G
957 and H retry rules. I suspect nobody ever tinkers with this value.
959 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
961 PH/45 When the plaintext authenticator is running as a client, the server's
962 challenges are checked to ensure they are valid base64 strings. By
963 default, the authentication attempt is cancelled if an invalid string is
964 received. Setting client_ignore_invalid_base64 true ignores these errors.
965 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
966 they are received. Thus, the responses can be made to depend on the
967 challenges. If an invalid string is ignored, an empty string is placed in
970 PH/46 Messages that are created by the autoreply transport now contains a
971 References: header, in accordance with RFCs 2822 and 3834.
973 PH/47 Added authenticated_sender_force to the smtp transport.
975 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
977 PH/49 Installed latest patch from the Sieve maintainer.
979 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
980 was also set, the appendfile transport was unnecessarily scanning a
981 directory of message files (e.g. for maildir delivery) to find the count
982 of files (along with the size), even though it did not need this
983 information. It now does the scan only if it needs to find either the
984 size of the count of files.
986 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
988 PH/52 Two bugs concerned with error handling when the smtp transport is
991 (i) Exim was not creating retry information for temporary errors given
992 for individual recipients after the DATA command when the smtp transport
993 was used in LMTP mode. This meant that they could be retried too
994 frequently, and not timed out correctly.
996 (ii) Exim was setting the flag that allows error details to be returned
997 for LMTP errors on RCPT commands, but not for LMTP errors for individual
998 recipients that were returned after the DATA command.
1000 PH/53 This is related to PH/52, but is more general: for any failing address,
1001 when detailed error information was permitted to be returned to the
1002 sender, but the error was temporary, then after the final timeout, only
1003 "retry timeout exceeded" was returned. Now it returns the full error as
1004 well as "retry timeout exceeded".
1006 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1007 do this, and (what is worse) MTAs that accept it.
1009 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1010 will now be deprecated.
1012 PH/56 New os.c-cygwin from the Cygwin maintainer.
1014 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1015 formats (previously only available in exim formats via -bpr, -bpru,
1016 and -bpra. Now also available in native and exiqgrep formats)
1018 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1019 with very large, slow to parse queues
1021 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1023 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1025 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1026 responses to authentication challenges, though it was showing the
1027 challenges; (ii) I've removed the CR characters from the debug output for
1030 PH/58 Allow for the insertion of a newline as well as a space when a string
1031 is turned into more than one encoded-word during RFC 2047 encoding. The
1032 Sieve code now uses this.
1034 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1035 data_4xx, lost_connection, tls_required.
1037 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1038 message was being sent as an SMTP response.
1040 PH/61 Add -l and -k options to exicyclog.
1042 PH/62 When verifying, if an address was redirected to one new address, so that
1043 verification continued, and the new address failed or deferred after
1044 having set something in $address_data, the value of $address_data was not
1045 passed back to the ACL. This was different to the case when no
1046 redirection occurred. The value is now passed back in both cases.
1048 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1049 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1050 use login_cap.h, so on its own it isn't the distinguishing feature. The
1051 new name refers directly to the setclassresources() function.
1053 PH/65 Added configuration files for NetBSD3.
1055 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1057 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1058 is preferred over IPv4.
1060 PH/68 The bounce_return_message and bounce_return_body options were not being
1061 honoured for bounces generated during the reception of non-SMTP messages.
1062 In particular, this applied to messages rejected by the ACL. This bug has
1063 been fixed. However, if bounce_return_message is true and bounce_return_
1064 body is false, the headers that are returned for a non-SMTP message
1065 include only those that have been read before the error was detected.
1066 (In the case of an ACL rejection, they have all been read.)
1068 PH/69 The HTML version of the specification is now built in a directory called
1069 spec_html instead of spec.html, because the latter looks like a path with
1070 a MIME-type, and this confuses some software.
1072 PH/70 Catch two compiler warnings in sieve.c.
1074 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1075 function verify_get_ident() calls ip_connect() to connect a socket, but
1076 if the "connect()" function timed out, ip_connect() used to close the
1077 socket. However, verify_get_ident() also closes the socket later, and in
1078 between Exim writes to the log, which may get opened at this point. When
1079 the socket was closed in ip_connect(), the log could get the same file
1080 descriptor number as the socket. This naturally causes chaos. The fix is
1081 not to close the socket in ip_connect(); the socket should be closed by
1082 the function that creates it. There was only one place in the code where
1083 this was missing, in the iplookup router, which I don't think anybody now
1084 uses, but I've fixed it anyway.
1086 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1087 well as to direct DNS lookups. Otherwise the handling of names in host
1088 lists is inconsistent and therefore confusing.
1094 PH/01 Two changes to the default runtime configuration:
1096 (1) Move the checks for relay_from_hosts and authenticated clients from
1097 after to before the (commented out) DNS black list checks.
1099 (2) Add control=submission to the relay_from_hosts and authenticated
1100 clients checks, on the grounds that messages accepted by these
1101 statements are most likely to be submissions.
1103 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1105 (1) Generate an error if the third argument for the ${prvs expansion is
1108 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1111 (3) Reset the variables that are obtained from the first argument of
1112 ${prvscheck and used in the second argument before leaving the code,
1113 because their memory is reclaimed, so using them afterwards may do
1116 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1117 one (it's much easier than Tom thought :-).
1119 (5) Because of (4), we can now allow for the use of $prvscheck_result
1120 inside the third argument.
1122 PH/03 For some reason, the default setting of PATH when running a command from
1123 a pipe transport was just "/usr/bin". I have changed it to
1126 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1127 anything to be listed in the output from -bV.
1129 PH/05 When a filter generated an autoreply, the entire To: header line was
1130 quoted in the delivery log line, like this:
1132 => >A.N.Other <ano@some.domain> <original@ddress> ...
1134 This has been changed so that it extracts the operative address. There
1135 may be more than one such address. If so, they are comma-separated, like
1138 => >ano@some.domain,ona@other.domain <original@ddress> ...
1140 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1141 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1142 not being looked up in the rDNS to get a host name, Exim was showing the
1143 IP address twice in Received: lines, even though the IP addresses were
1144 identical. For example:
1146 Received: from [1.2.3.4] (helo=[1.2.3.4])
1148 However, if the real host name was known, it was omitting the HELO data
1149 if it matched the actual IP address. This has been tidied up so that it
1150 doesn't show the same IP address twice.
1152 PH/07 When both +timestamp and +memory debugging was on, the value given by
1153 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1154 called by the debug printing, thereby overwriting the timestamp buffer.
1155 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1157 PH/08 When the original message was included in an autoreply transport, it
1158 always said "this is a copy of the message, including all the headers",
1159 even if body_only or headers_only was set. It now gives an appropriate
1162 PH/09 Applied a patch from the Sieve maintainer which:
1164 o fixes some comments
1165 o adds the (disabled) notify extension core
1166 o adds some debug output for the result of if/elsif tests
1167 o points to the current vacation draft in the documentation
1168 and documents the missing references header update
1172 o fixes a bug in processing the envelope test (when testing
1173 multiple envelope elements, the last element determinted the
1176 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1177 Electronic Mail") by including:
1179 Auto-submitted: auto-generated
1181 in the messages that it generates (bounce messages and others, such as
1182 warnings). In the case of bounce messages for non-SMTP mesages, there was
1183 also a typo: it was using "Auto_submitted" (underscore instead of
1184 hyphen). Since every message generated by Exim is necessarily in response
1185 to another message, thes have all been changed to:
1187 Auto-Submitted: auto-replied
1189 in accordance with these statements in the RFC:
1191 The auto-replied keyword:
1193 - SHOULD be used on messages sent in direct response to another
1194 message by an automatic process,
1196 - MUST NOT be used on manually-generated messages,
1198 - MAY be used on Delivery Status Notifications (DSNs) and Message
1199 Disposition Notifications (MDNs),
1201 - MUST NOT be used on messages generated by automatic or periodic
1202 processes, except for messages which are automatic responses to
1205 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1206 to the default Received: header definition.
1208 PH/12 Added log selector acl_warn_skipped (default on).
1210 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1211 variables because (a) they are in the wrong storage pool and (b) even if
1212 they were copied, it wouldn't work properly because of the caching.
1214 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1215 checking when decoding. Apparently there are clients that generate
1216 overlong encoded strings. Why am I not surprised?
1218 PH/15 If the first argument of "${if match_address" was not empty, but did not
1219 contain an "@" character, Exim crashed. Now it writes a panic log message
1220 and treats the condition as false.
1222 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1224 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1225 extension "envelope-auth". The code is finished and in agreement with
1226 other implementations, but there is no documentation so far and in fact,
1227 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1228 not changing the active code.
1230 Print executed "if" and "elsif" statements when debugging is used. This
1231 helps a great deal to understand what a filter does.
1233 Document more things not specified clearly in RFC3028. I had all this
1234 sorted out, when out of a sudden new issues came to my mind. Oops."
1236 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1239 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1240 canonical form (as documented). However, after a host name lookup from
1241 the IP address, check_host() was doing a simple string comparison with
1242 addresses acquired from the DNS when checking that the found name did
1243 have the original IP as one of its addresses. Since any found IPv6
1244 addresses are likely to be in abbreviated form, the comparison could
1245 fail. Luckily, there already exists a function for doing the comparison
1246 by converting both addresses to binary, so now that is used instead of
1247 the text comparison.
1249 PH/20 There was another similar case to PH/19, when a complete host name was
1250 given in a host list; looking up its IP address could give an abbreviated
1251 form, whereas the current host's name might or might not be abbreviated.
1252 The same fix has been applied.
1258 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1259 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1262 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1263 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1265 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1267 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1268 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1269 Also, for "Auto-Submitted", treat anything other than "no" as
1270 non-personal, in accordance with RFC 3834. (Previously it treated
1271 anything starting "auto-" as non-personal.)
1273 TF/01 The control=submission/name=... option had a problem with syntax
1274 errors if the name included a slash character. The /name= option
1275 now slurps the rest of the string, so it can include any characters
1276 but it must come last in the list of options (after /sender_retain
1279 PH/05 Some modifications to the interface to the fake nameserver for the new
1287 TK/01 Added the "success_on_redirect" address verification option. See
1288 NewStuff for rationale and an example.
1290 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1292 PH/02 Patch to exigrep to allow it to work on syslog lines.
1294 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1295 fread() to skip over the body file's header line, because in Cygwin the
1296 header line is locked and is inaccessible.
1298 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1299 co-exist for some time) to make it clear that it is the Exim ID that is
1300 referenced, not the Message-ID: header line.
1302 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1303 string_format() function, because snprintf() does not exist on all
1306 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1307 ${stat: expansion item.
1309 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1310 protocol synchronization error", to keep the pedants happy.
1312 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1313 well as for IRIX systems, when gcc is being used. See the host.c source
1316 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1318 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1321 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1322 transport and to the smtp transport in LMTP mode.
1324 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1326 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1327 run a filter in a subprocess. This could lead to confusion in subsequent
1328 lookups in the parent process. There should also be a search_tidyup() at
1329 the end of the subprocess.
1331 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1332 only if the host matched helo_try_verify_hosts, which caused the
1333 verification to occur when the EHLO/HELO command was issued. The ACL just
1334 tested the remembered result. Now, if a previous verification attempt has
1335 not happened, "verify = helo" does it there and then.
1337 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1339 TK/03 Fix log output including CR from clamd.
1341 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1342 exist provoked a memory error which could cause a segfault.
1344 PH/15 Installed PCRE 6.2
1346 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1348 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1349 of the problem. Specifically, suggested +O2 rather than +O1 for the
1352 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1354 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1355 record, the port was not correctly logged when the outgoing_port log
1356 selector was set (it logged the transort's default port).
1358 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1359 fallback_hosts, and "hosts" in the smtp transport.
1361 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1362 host errors such as "Connection refused".
1364 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1365 authentication with radiusclient 0.4.9:
1367 - Error returned from rc_read_config was caught wrongly
1368 - Username/password not passed on to radius server due to wrong length.
1370 The presumption is that some radiusclient API changes for 4.51/PH/17
1371 were not taken care of correctly. The code is still untested by me (my
1372 Linux distribution still has 0.3.2 of radiusclient), but it was
1373 contributed by a Radius user.
1375 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1376 expanding the "port" option of the smtp transport.
1378 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1379 while reading a MIME header. Thanks to Tom Hughes for a patch.
1381 PH/24 Include config.h inside local_scan.h so that configuration settings are
1384 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1385 that in an ACL for RCPT (for example), you can examine exactly what was
1388 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1389 commands, but it was not correctly comparing the address with the actual
1390 client host address. Thus, it would show the EHLO address in Received:
1391 header lines when this was not necessary.
1393 PH/27 Added the % operator to ${eval:}.
1395 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1396 it should be ignoring failures (because with -C, for example, it has lost
1397 privilege). It wasn't ignoring creation failures other than "already
1400 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1403 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1404 because an input line was too long, either on its own, or by virtue of
1405 too many continuations, the temporary file was not being removed, and the
1406 return code was incorrect.
1408 PH/31 Missing "BOOL" in function definition in filtertest.c.
1410 PH/32 Applied Sieve patches from the maintainer.
1412 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1414 PH/33 Added "verify = not_blind".
1416 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1417 Local/Makefile (with some defaults set). These are used in built scripts
1418 such as exicyclog, but they have never been used in the exim_install
1419 script (though there are many overriding facilities there). I have
1420 arranged that the exim_install script now takes note of these two
1423 PH/35 Installed configuration files for Dragonfly.
1425 PH/36 When a locally submitted message by a trusted user did not contain a
1426 From: header, and the sender address was obtained from -f or from an SMTP
1427 MAIL command, and the trusted user did not use -F to supply a sender
1428 name, $originator_name was incorrectly used when constructing a From:
1429 header. Furthermore, $originator_name was used for submission mode
1430 messages from external hosts without From: headers in a similar way,
1431 which is clearly wrong.
1433 PH/37 Added control=suppress_local_fixups.
1435 PH/38 When log_selector = +received_sender was set, and the addition of the
1436 sender made the log line's construction buffer exactly full, or one byte
1437 less than full, an overflow happened when the terminating "\n" was
1440 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1441 when the result of a list match is failure because a DNS lookup failed.
1443 PH/40 RM_COMMAND is now used in the building process.
1445 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1446 the "build-* directories that it finds.
1448 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1449 address in a domain literal was a prefix of an interface address.
1451 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1452 when verifying a sender address, unless rewrite_headers is false.
1454 PH/44 Wrote a long comment about why errors_to addresses are verified as
1455 recipients, not senders.
1457 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1458 the ratelimit ACL was added.
1460 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1462 PH/46 Added extra information about PostgreSQL errors to the error string.
1464 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1465 avoiding the need to install special zones in a real server. This is
1466 backwards compatible; if it can't find the fake resolver, it drops back.
1467 Thus, both old and new test suites can be run.
1469 TF/02 Added util/ratelimit.pl
1471 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1472 clock is set back in time.
1474 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1475 Candler <B.Candler@pobox.com>.
1477 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1478 OK for addresses that are the result of redirections.
1480 PH/48 A number of further additions for the benefit of the new test suite,
1481 including a fake gethostbyname() that interfaces to the fake DNS resolver
1484 TF/06 The fix for widen_domains has also been applied to qualify_single and
1485 search_parents which are the other dnslookup options that can cause
1488 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1491 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1493 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1494 service" if it encountered an MX record with an empty target hostname.
1495 The message is now "an MX or SRV record indicated no SMTP service".
1497 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1498 if the DNS of the sending site is misconfigured. This is quite a
1499 common situation. This change restores the behaviour of treating a
1500 helo verification defer as a failure.
1502 PH/51 If self=fail was set on a router, the bounce message did not include the
1503 actual error message.
1509 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1511 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1512 command itself ended in error, the underlying message about the transport
1513 filter timeout was being overwritten with the pipe command error. Now the
1514 underlying error message should be appended to the second error message.
1516 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1518 PH/02 Reduce the amount of output that "make" produces by default. Full output
1519 can still be requested.
1521 PH/03 The warning log line about a condition test deferring for a "warn" verb
1522 was being output only once per connection, rather than after each
1523 occurrence (because it was using the same function as for successful
1524 "warn" verbs). This seems wrong, so I have changed it.
1526 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1527 it should not have, which might have caused a crash in the right
1528 circumstances, but probably never did.
1530 PH/04 Installed a modified version of Tony Finch's patch to make submission
1531 mode fix the return path as well as the Sender: header line, and to
1532 add a /name= option so that you can make the user's friendly name appear
1535 TF/03 Added the control = fakedefer ACL modifier.
1537 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1538 Mark Lowes for thorough testing.
1540 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1542 TK/03 Merged latest SRS patch from Miles Wilton.
1544 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1545 with the definition in sysexits.h (which is #included earlier).
1546 Fortunately, Exim does not actually use EX_OK. The code used to try to
1547 preserve the sysexits.h value, by assumimg that macro definitions were
1548 scanned for macro replacements. I have been disabused of this notion,
1549 so now the code just undefines EX_OK before #including unistd.h.
1551 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1552 in the smtp transport. When a block could not be written in a single
1553 write() function, the timeout was being re-applied to each part-write.
1554 This seems wrong - if the receiver was accepting one byte at a time it
1555 would take for ever. The timeout is now adjusted when this happens. It
1556 doesn't have to be particularly precise.
1558 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1559 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1561 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1562 without a domain if the check to <postmaster@domain> fails.
1564 SC/01 Eximstats: added -xls and the ability to specify output files
1565 (patch written by Frank Heydlauf).
1567 SC/02 Eximstats: use FileHandles for outputing results.
1569 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1571 SC/04 Eximstats: fixed display of large numbers with -nvr option
1573 SC/05 Eximstats: fixed merging of reports with empty tables.
1575 SC/06 Eximstats: added the -include_original_destination flag
1577 SC/07 Eximstats: removed tabs and trailing whitespace.
1579 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1581 TK/06 MBOX spool code: Add real "From " MBOX separator line
1582 so the .eml file is really in mbox format (even though
1583 most programs do not really care). Patch from Alex Miller.
1585 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1586 The latter is generated from $received_to and is only set if the
1587 message has one envelope recipient. SA can use these headers,
1588 obviously out-of-the-box. Patch from Alex Miller.
1590 PH/08 The ${def test on a variable was returning false if the variable's
1591 value was "0", contrary to what the specification has always said!
1592 The result should be true unless the variable is empty.
1594 PH/09 The syntax error of a character other than { following "${if
1595 def:variable_name" (after optional whitespace) was not being diagnosed.
1596 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1597 accidental colon was present, for example, could give incorrect results.
1599 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1600 result is used (not including appendfile, where other changes are about
1603 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1604 This involved changing a lot of size variables from int to off_t. It
1605 should work with maildirs and everything.
1607 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1608 spamd dying while we are connected to it.
1610 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1613 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1614 function for BDB 4.3.
1616 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1618 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1619 the sources for PCRE have been re-arranged and more files are now
1622 PH/15 The code I had for printing potentially long long variables in PH/11
1623 above was not the best (it lost precision). The length of off_t variables
1624 is now inspected at build time, and an appropriate printing format (%ld
1625 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1626 to be "long long int" or "long int". This is needed for the internal
1627 formatting function string_vformat().
1629 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1630 the configuration file to be ":syslog", then the script "guesses" where
1631 the logs files are, rather than using the compiled in default. In our
1632 case the guess is not the same as the compiled default, so the script
1633 suddenly stopped working when I started to use syslog. The patch checks
1634 to see if log_file_path is "". If so, it attempts to read it from exim
1635 with no configuration file to get the compiled in version, before it
1636 falls back to the previous guessing code."
1638 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1639 implementing BATV in an Exim configuration. See NewStuff for the gory
1642 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1643 Makefile that are specific to HP-UX.
1645 PH/18 If the "use_postmaster" option was set for a recipient callout together
1646 with the "random" option, the postmaster address was used as the MAIL
1647 FROM address for the random test, but not for the subsequent recipient
1648 test. It is now used for both.
1650 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1651 patch removes a few documentation additions to RFC 3028, because the
1652 latest draft now contains them. It adds the new en;ascii-case comparator
1653 and a new error check for 8bit text in MIME parts. Comparator and
1654 require names are now matched exactly. I enabled the subaddress
1655 extension, but it is not well tested yet (read: it works for me)."
1657 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1658 rework some of the code of TK/09 above to avoid the hardwired use of
1659 "%lld" and "long long". Replaced the call to snprintf() with a call to
1662 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1663 records point to non-existent hosts", "retry timeout exceeded", and
1664 "retry time not reached for any host after a long failure period".
1666 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1667 experimental DomainKeys support:
1669 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1670 (2) On an error such as an illegally used "control", the wrong name for
1671 the control was given.
1673 These problems did NOT occur unless DomainKeys support was compiled.
1675 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1677 PH/24 Added ${if match_ip condition.
1679 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1680 databases so that it will be absolutely obvious if a crash occurs in the
1681 DB library. This is a regular occurrence (often caused by mis-matched
1684 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1685 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1686 picked up on a user's system that detects such things. There doesn't seem
1687 to be a gcc warning option for this - only an attribute that has to be
1688 put on the function's prototype. It seems that in Fedora Core 4 they have
1689 set this on a number of new functions. No doubt there will be more in due
1692 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1693 specify a transport. However, if an address that was verified by such a
1694 router was the subject of a callout, Exim crashed because it tried to
1695 read the rcpt_include_affixes from the non-existent transport. Now it
1696 just assumes that the setting of that option is false. This bug was
1697 introduced by 4.51/PH/31.
1699 PH/28 Changed -d+all to exclude +memory, because that information is very
1700 rarely of interest, but it makes the output a lot bigger. People tend to
1701 do -d+all out of habit.
1703 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1704 code in os-type was giving problems when libc.so lives in lib64, like on
1707 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1708 aren't the modern standard, and it seems that some systems' include files
1709 don't always have them. Exim was already checking for some of the newer
1710 ones like T_AAAA, and defining it itself. I've added checks for all the
1711 record types that Exim uses.
1713 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1714 not automatically generating a new one, as it is supposed to. This
1715 prevented TLS from working. If the file did exist, but contained invalid
1716 data, a new version was generated, as expected. It was only the case of a
1717 non-existent file that was broken.
1719 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1720 with a change in libdomainkeys > 0.64.
1722 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1723 from DNS. If the selector record carries the flag, it now has
1724 precedence over the domain-wide flag.
1726 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1728 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1729 the use of a port name that isn't defined in /etc/services) occurred, the
1730 message was deferred as in a normal delivery, and thus remained on the
1731 spool, instead of being failed because of the mua_wrapper setting. This
1732 is now fixed, and I tidied up some of the mua_wrapper messages at the
1735 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1736 the messages in a hash of arrays rather than using individual hashes.
1737 This is a bit cleaner and results in dramatic memory savings, albeit
1738 at a slight CPU cost.
1740 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1741 as requested by Marc Sherman.
1743 SC/10 Eximstats: added histograms for user specified patterns as requested
1746 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1748 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1749 fopen() in the content-scanning modules that did not already have it.
1755 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1756 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1758 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1760 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1762 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1763 to test Sieve filters that use "vacation".
1765 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1766 that changes the way the GnuTLS parameters are stored in the cache file.
1767 The new format can be generated externally. For backward compatibility,
1768 if the data in the cache doesn't make sense, Exim assumes it has read an
1769 old-format file, and it generates new data and writes a new file. This
1770 means that you can't go back to an older release without removing the
1773 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1774 work if there are any delivery delays because "one_time" forces the
1775 parent to be marked "delivered", so its unseen clone is never tried
1776 again. For this reason, Exim now forbids the simultaneous setting of
1779 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1780 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1781 entry if you want to know the details. The fix, however, neglected to
1782 consider the case when local delivery batching is involved. The test for
1783 "previously delivered" was not happening when checking to see if an
1784 address could be batched with a previous (undelivered) one; under
1785 certain circumstances this could lead to multiple deliveries to the same
1788 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1789 in its include files, and this causes problems building Exim.
1791 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1792 header_syntax) but Exim was just ignoring anything given after a slash.
1793 In particular, this caused confusion with an attempt to use "verify =
1794 reverse_host_lookup/defer_ok". An error is now given when options are
1795 supplied for verify items that do not have them. (Maybe reverse_host_
1796 lookup should have a defer_ok option, but that's a different point.)
1798 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1799 defined by RFC 821) to 2048, because there were problems with some AUTH
1800 commands, and RFC 1869 says the size should be increased for extended
1801 SMTP commands that take arguments.
1803 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1806 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1807 "unknown" error; now it says that the functionality isn't in the binary.
1809 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1810 an address' error message when a string expansion fails (syntax or
1811 whatever). Otherwise the password may appear in the log. Following change
1812 PH/42 below, there is no longer a chance of it appearing in a bounce
1815 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1817 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1818 of its addresses was ever tried. (Bugzilla bug #2.)
1820 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1821 the result incorrectly in the debug output. (It correctly added a newline
1822 to what was transported.)
1824 TF/01 Added $received_time.
1826 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1827 commented out examples of how to interface to a virus scanner and to
1828 SpamAssassin. Also added commented examples of av_scanner and
1829 spamd_address settings.
1831 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1832 and controls are allowed in which ACLs. There were a couple of minor
1833 errors. Some of the entries in the conditions table (which is a table of
1834 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1835 negation of where the condition IS allowed.
1837 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1839 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1840 header file does not have a version number, so I've had to invent a new
1841 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1842 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1843 radiusclient), but it was contributed by a Radius user.
1845 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1846 files or directories, for OpenSSL.
1848 PH/19 When an Exim process that is running as root has to create an Exim log
1849 file, it does so in a subprocess that runs as exim:exim so as to get the
1850 ownership right at creation (otherwise, other Exim processes might see
1851 the file with the wrong ownership). There was no test for failure of this
1852 fork() call, which would lead to the process getting stuck as it waited
1853 for a non-existent subprocess. Forks do occasionally fail when resources
1854 run out. I reviewed all the other calls to fork(); they all seem to check
1857 PH/20 When checking for unexpected SMTP input at connect time (before writing
1858 the banner), Exim was not dealing correctly with a non-positive return
1859 from the read() function. If the client had disconnected by this time,
1860 the result was a log entry for a synchronization error with an empty
1861 string after "input=" when read() returned zero. If read() returned -1
1862 (an event I could not check), uninitialized data bytes were printed.
1863 There were reports of junk text (parts of files, etc) appearing after
1866 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1868 PH/22 Added support for macro redefinition, and (re)definition in between
1869 driver and ACL definitions.
1871 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1872 forgetting to use the resulting value; it was using the unexpanded value.
1874 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1875 hadn't been configured. The fix is from Juergen Kreileder, who
1876 understands it better than I do:
1878 "Here's what I see happening with three configured cyrus_sasl
1879 authenticators configured (plain, login, cram-md5):
1881 On startup auth_cyrus_sasl_init() gets called for each of these.
1882 This means three calls to sasl_listmech() without a specified mech_list.
1883 => SASL tests which mechs of all available mechs actually work
1884 => three warnings about OTP not working
1885 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1887 With the patch, sasl_listmech() also gets called three times. But now
1888 SASL's mech_list option is set to the server_mech specified in the the
1889 authenticator. Or in other words, the answer from sasl_listmech()
1890 gets limited to just the mech you're testing for (which is different
1892 => the return list contains just 'plain' or 'login', 'cram-md5' or
1893 nothing depending on the value of ob->server_mech.
1895 I've just tested the patch: Authentication still works fine,
1896 unavailable mechs specified in the exim configuration are still
1897 caught, and the auth.log warnings about OTP are gone."
1899 PH/25 When debugging is enabled, the contents of the command line are added
1900 to the debugging output, even when log_selector=+arguments is not
1903 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1904 answer is "GNU", and only if the return is "GNU/something" is the answer
1907 PH/27 $acl_verify_message is now set immediately after the failure of a
1908 verification in an ACL, and so is available in subsequent modifiers. In
1909 particular, the message can be preserved by coding like this:
1911 warn !verify = sender
1912 set acl_m0 = $acl_verify_message
1914 Previously, $acl_verify_message was set only while expanding "message"
1915 and "log_message" when a very denied access.
1917 PH/28 Modified OS/os.c-Linux with
1919 -#ifndef OS_LOAD_AVERAGE
1920 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1922 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1923 nomenclature these days.)
1925 PH/29 Installed patch from the Sieve maintainer that adds the options
1926 sieve_useraddress and sieve_subaddress to the redirect router.
1928 PH/30 In these circumstances:
1929 . Two addresses routed to the same list of hosts;
1930 . First host does not offer TLS;
1931 . First host accepts first address;
1932 . First host gives temporary error to second address;
1933 . Second host offers TLS and a TLS session is established;
1934 . Second host accepts second address.
1935 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1936 and peerdn, if requested) that were in fact used only for the second
1939 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1940 attention to any local part prefix or suffix that was matched by the
1941 router that accepted the address. It now behaves in the same way as it
1942 does for delivery: the affixes are removed from the local part unless
1943 rcpt_include_affixes is set on the transport.
1945 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1946 timeout during the DATA phase of an incoming message.
1948 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1949 applied a patch sanctioned by the Sieve maintainer.
1951 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1952 the uid or gid is negative. A case of a negative gid caused this to be
1953 noticed. The fix allows for either to be negative.
1955 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1956 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1959 PH/36 The change PH/12 above was broken. Fixed it.
1961 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1962 the grounds that routing the same address twice would always produce the
1963 same answer. This might have been true once, but it is certainly no
1964 longer true now. Routing a child address may depend on the previous
1965 routing that produced that child. Some complicated redirection strategies
1966 went wrong when messages had multiple recipients, and made Exim's
1967 behaviour dependent on the order in which the addresses were given.
1969 I have moved the duplicate checking until after the routing is complete.
1970 Exim scans the addresses that are assigned to local and remote
1971 transports, and removes any duplicates. This means that more work will be
1972 done, as duplicates will always all be routed, but duplicates are
1973 presumably rare, so I don't expect this is of any significance.
1975 For deliveries to pipes, files, and autoreplies, the duplicate checking
1976 still happens during the routing process, since they are not going to be
1979 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1980 It corrects a timeout issue with spamd. This is Ian's comment: "The
1981 background is that sometimes spamd either never reads data from a
1982 connection it has accepted, or it never writes response data. The exiscan
1983 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1984 blindly assumes that writes won't block so it may never time out."
1986 PH/39 Allow G after quota size as well as K and M.
1988 PH/40 The value set for $authenticated_id in an authenticator may not contain
1989 binary zeroes or newlines because the value is written to log lines and
1990 to spool files. There was no check on this. Now the value is run through
1991 the string_printing() function so that such characters are converted to
1992 printable escape sequences.
1994 PH/41 $message_linecount is a new variable that contains the total number of
1995 lines in the message. Compare $body_linecount, which is the count for the
1998 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1999 bounce and delay warning messages, except in certain special cases, which
2002 (a) An SMTP error message from a remote host;
2003 (b) A message specified in a :fail: redirection;
2004 (c) A message specified in a "fail" command in a system filter;
2005 (d) A message specified in a FAIL return from the queryprogram router;
2006 (e) A message specified by the cannot_route_message router option.
2008 In these cases only, Exim does include the error details in bounce and
2009 warning messages. There are also a few cases where bland messages such
2010 as "unrouteable address" or "local delivery error" are given.
2012 PH/43 $value is now also set for the "else" part of a ${run expansion.
2014 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2015 being worked on, but at least Exim now implements the latest version to
2018 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2019 process to complete was treated as a delivery failure, a timeout while
2020 writing the message to the pipe was logged, but erroneously treated as a
2021 successful delivery. Such timeouts include transport filter timeouts. For
2022 consistency with the overall process timeout, these timeouts are now
2023 treated as errors, giving rise to delivery failures by default. However,
2024 there is now a new Boolean option for the pipe transport called
2025 timeout_defer, which, if set TRUE, converts the failures into defers for
2026 both kinds of timeout. A transport filter timeout is now identified in
2029 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2030 systems where "make" and "gmake" are different, calling "gmake" at top
2031 level broke things. I've arranged for the value of $(MAKE) to be passed
2032 from the Makefile to this script so that it can call the same version of
2036 A note about Exim versions 4.44 and 4.50
2037 ----------------------------------------
2039 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2040 changes of various kinds. As a consequence, a big documentation update was
2041 needed. This delayed the release for rather longer than seemed good, especially
2042 in the light of a couple of (minor) security issues. Therefore, the changes
2043 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2044 release. So 4.44 and 4.50 are in effect two different branches that both start
2047 I have left the 4.50 change log unchanged; it contains all the changes since
2048 4.43. The change log for 4.44 is below; many of its items are identical to
2049 those for 4.50. This seems to be the most sensible way to preserve the
2050 historical information.
2056 1. Minor wording change to the doc/README.SIEVE file.
2058 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2059 computation of the current number of files was incorrect.
2061 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2062 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2065 4. Give more explanation in the error message when the command for a transport
2066 filter fails to execute.
2068 5. There are several places where Exim runs a non-Exim command in a
2069 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2070 was being done only for the command run by the queryprogram router. It is
2071 now done for all such subprocesses. The other cases are: ${run, transport
2072 filters, and the commands run by the lmtp and pipe transports.
2074 6. Added CONFIGURE_GROUP build-time option.
2076 7. Some older OS have a limit of 256 on the maximum number of file
2077 descriptors. Exim was using setrlimit() to set 1000 as a large value
2078 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2079 systems. I've change it so that if it can't get 1000, it tries for 256.
2081 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2082 was an oversight, and furthermore, ever since the addition of extra
2083 controls (e.g. 4.43/32), the checks on when to allow different forms of
2084 "control" were broken. There should now be diagnostics for all cases when a
2085 control that does not make sense is encountered.
2087 9. Added the /retain_sender option to "control=submission".
2089 10. $recipients is now available in the predata ACL (oversight).
2091 11. Tidy the search cache before the fork to do a delivery from a message
2092 received from the command line. Otherwise the child will trigger a lookup
2093 failure and thereby defer the delivery if it tries to use (for example) a
2094 cached ldap connection that the parent has called unbind on.
2096 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2097 of $address_data from the recipient verification was clobbered by the
2098 sender verification.
2100 13. The value of address_data from a sender verification is now available in
2101 $sender_address_data in subsequent conditions in the ACL statement.
2103 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2105 15. Added a new option "connect=<time>" to callout options, to set a different
2108 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2109 was its contents. (It was OK if the option was not defined at all.)
2111 17. A "Completed" log line is now written for messages that are removed from
2112 the spool by the -Mrm option.
2114 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2115 information about exactly what failed.
2117 19. Added -dd to debug only the daemon process.
2119 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2120 handles timeouts, both on the server side and network timeouts. Renamed the
2121 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2123 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2124 It is now set to "smtps".
2126 22. $host_address is now set to the target address during the checking of
2127 ignore_target_hosts.
2129 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2130 being passed; this would have caused $sender_host_name to have been used if
2131 matching the list had actually called for a host name (not very likely,
2132 since this list is usually IP addresses). A host name is now passed as
2135 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2136 code that specifies a non-restarting handler (typically sigaction() in
2137 modern systems) in an attempt to fix a rare and obscure crash bug.
2139 25. Narrowed the window for a race in the daemon that could cause it to ignore
2140 SIGCHLD signals. This is not a major problem, because they are used only to
2141 wake it up if nothing else does.
2143 26. A malformed maildirsize file could cause Exim to calculate negative values
2144 for the mailbox size or file count. Odd effects could occur as a result.
2145 The maildirsize information is now recalculated if the size or filecount
2148 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2149 support for a long time. Removed HAVE_SYS_VFS_H.
2151 28. Installed the latest version of exipick from John Jetmore.
2153 29. In an address list, if the pattern was not a regular expression, an empty
2154 subject address (from a bounce message) matched only if the pattern was an
2155 empty string. Non-empty patterns were not even tested. This was the wrong
2156 because it is perfectly reasonable to use an empty address as part of a
2157 database query. An empty address is now tested by patterns that are
2158 lookups. However, all the other forms of pattern expect the subject to
2159 contain a local part and a domain, and therefore, for them, an empty
2160 address still always fails if the pattern is not itself empty.
2162 30. Exim went into a mad DNS loop when attempting to do a callout where the
2163 host was specified on an smtp transport, and looking it up yielded more
2164 than one IP address.
2166 31. Re-factored the code for checking spool and log partition space into a
2167 function that finds that data and another that does the check. The former
2168 is then used to implement four new variables: $spool_space, $log_space,
2169 $spool_inodes, and $log_inodes.
2171 32. The RFC2047 encoding function was originally intended for short strings
2172 such as real names; it was not keeping to the 75-character limit for
2173 encoded words that the RFC imposes. It now respects the limit, and
2174 generates multiple encoded words if necessary. To be on the safe side, I
2175 have increased the buffer size for the ${rfc2047: expansion operator from
2178 33. It is now permitted to omit both strings after an "if" condition; if the
2179 condition is true, the result is "true". As before, when the second string
2180 is omitted, a false condition yields an empty string. This makes it less
2181 cumbersome to write custom ACL and router conditions.
2183 34. Failure to deliver a bounce message always caused it to be frozen, even if
2184 there was an errors_to setting on the router. The errors_to setting is now
2187 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2188 canonical form (fully expanded) before being placed in
2189 $sender_host_address.
2191 36. The table in the code that translates DNS record types into text (T_A to
2192 "A" for instance) was missing entries for NS and CNAME. It is just possible
2193 that this could have caused confusion if both these types were looked up
2194 for the same domain, because the text type is used as part of Exim's
2195 per-process caching. But the chance of anyone hitting this buglet seems
2198 37. The dnsdb lookup has been extended in a number of ways.
2200 (1) There is a new type, "zns", which walks up the domain tree until it
2201 finds some nameserver records. It should be used with care.
2203 (2) There is a new type, "mxh", which is like "mx" except that it returns
2204 just the host names, not the priorities.
2206 (3) It is now possible to give a list of domains (or IP addresses) to be
2207 looked up. The behaviour when one of the lookups defers can be
2208 controlled by a keyword.
2210 (4) It is now possible to specify the separator character for use when
2211 multiple records are returned.
2213 38. The dnslists ACL condition has been extended: it is now possible to supply
2214 a list of IP addresses and/or domains to be looked up in a particular DNS
2217 39. Added log_selector=+queue_time_overall.
2219 40. When running the queue in the test harness, wait just a tad after forking a
2220 delivery process, to get repeatability of debugging output.
2222 41. Include certificate and key file names in error message when GnuTLS fails
2223 to set them up, because the GnuTLS error message doesn't include the name
2224 of the failing file when there is a problem reading it.
2226 42. Allow both -bf and -bF in the same test run.
2228 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2230 44. The "Exiscan patch" is now merged into the mainline Exim source.
2232 45. Sometimes the final signoff response after QUIT could fail to get
2233 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2234 < 0 before doing a fflush(). This bug looks as though it goes back to the
2235 introduction of TLS in release 3.20, but "sometimes" must have been rare
2236 because the tests only now provoked it.
2238 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2239 (this can affect the format of dates).
2241 47. exim_tidydb, when checking for the continued existence of a message for
2242 which it has found a message-specific retry record, was not finding
2243 messages that were in split spool directories. Consequently, it was
2244 deleting retry records that should have stayed in existence.
2246 48. Steve fixed some bugs in eximstats.
2248 49. The SPA authentication driver was not abandoning authentication and moving
2249 on to the next authenticator when an expansion was forced to fail,
2250 contradicting the general specification for all authenticators. Instead it
2251 was generating a temporary error. It now behaves as specified.
2253 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2254 (the order specifies the preference for clients). The order is now AES256,
2255 AES128, 3DES, ARCFOUR128.
2257 51. Small patch to Sieve code - explicitly set From: when generating an
2260 52. Exim crashed if a remote delivery caused a very long error message to be
2261 recorded - for instance if somebody sent an entire SpamAssassin report back
2262 as a large number of 550 error lines. This bug was coincidentally fixed by
2263 increasing the size of one of Exim's internal buffers (big_buffer) that
2264 happened as part of the Exiscan merge. However, to be on the safe side, I
2265 have made the code more robust (and fixed the comments that describe what
2268 53. Now that there can be additional text after "Completed" in log lines (if
2269 the queue_time_overall log selector is set), a one-byte patch to exigrep
2270 was needed to allow it to recognize "Completed" as not the last thing in
2273 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2274 patch that reportedly fixes this has been added. I am not expert enough to
2275 create a test for it. This is what the patch creator wrote:
2277 "I found a little strange behaviour of ldap code when working with
2278 Windows 2003 AD Domain, where users was placed in more than one
2279 Organization Units. When I tried to give exim partial DN, the exit code
2280 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2281 But simultaneously result of request was absolutely normal ldap result,
2282 so I produce this patch..."
2284 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2285 REFERENCE, so I have modified the code to exclude the patch when that macro
2288 55. Some experimental protocols are using DNS PTR records for new purposes. The
2289 keys for these records are domain names, not reversed IP addresses. The
2290 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2291 leaves it alone. Component reversal etc. now happens only for IP addresses.
2294 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2296 57. Double the size of the debug message buffer (to 2048) so that more of very
2297 long debug lines gets shown.
2299 58. The exicyclog utility now does better if the number of log files to keep
2300 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2302 59. Two changes related to the smtp_active_hostname option:
2304 (1) $smtp_active_hostname is now available as a variable.
2305 (2) The default for smtp_banner uses $smtp_active_hostname instead
2306 of $primary_hostname.
2308 60. The host_aton() function is supposed to be passed a string that is known
2309 to be a valid IP address. However, in the case of IPv6 addresses, it was
2310 not checking this. This is a hostage to fortune. Exim now panics and dies
2311 if the condition is not met. A case was found where this could be provoked
2312 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2313 components; fortuitously, this particular loophole had already been fixed
2314 by change 4.50/55 above.
2316 If there are any other similar loopholes, the new check in host_aton()
2317 itself should stop them being exploited. The report I received stated that
2318 data on the command line could provoke the exploit when Exim was running as
2319 exim, but did not say which command line option was involved. All I could
2320 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2321 running as the user.
2324 61. There was a buffer overflow vulnerability in the SPA authentication code
2325 (which came originally from the Samba project). I have added a test to the
2326 spa_base64_to_bits() function which I hope fixes it.
2329 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2330 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2332 63. The daemon start-up calls getloadavg() while still root for those OS that
2333 need the first call to be done as root, but it missed one case: when
2334 deliver_queue_load_max is set with deliver_drop_privilege. This is
2335 necessary for the benefit of the queue runner, because there is no re-exec
2336 when deliver_drop_privilege is set.
2338 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2339 This has been fixed.
2341 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2342 was in use, was not putting the data itself into the right store pool;
2343 consequently, it could be overwritten for a subsequent message in the same
2344 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2347 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2349 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2350 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2351 the function were treating the return as a boolean value, which happened to
2352 work because 0=false and not-0=true, but is not correct code.
2354 68. The host_aton() function was not handling scoped IPv6 addresses (those
2355 with, for example, "%eth0" on the end) correctly.
2357 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2358 negated items (that is, ~something) in unsigned ints. Some compilers
2359 apparently mutter when there is no cast.
2361 70. If an address verification called from an ACL failed, and did not produce a
2362 user-specific message (i.e. there was only a "system" message), nothing was
2363 put in $acl_verify_message. In this situation, it now puts the system
2366 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2367 session; change 4.31/43 added the unwanted input to the log line - except
2368 that it did not do this in the start of session case. It now does.
2370 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2371 This is wrong and can cause the other end to generate a synchronization
2372 error if it is another Exim or anything else that does the synchronization
2373 check. A QUIT command is no longer sent after a timeout.
2375 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2376 during host lookups.
2378 74. The defer_ok option of callout verification was not working if it was used
2379 when verifying addresses in header lines, that is, for this case:
2381 verify = header_sender/callout=defer_ok
2383 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2384 those file descriptors could be used for SMTP connections. If anything
2385 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2386 could be sent to the SMTP client, causing chaos. The daemon now opens
2387 stdin, stdout, and stderr to /dev/null when it puts itself into the
2390 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2391 log by default. The user can override this with suitable Perl magic.
2393 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2394 the log message when discard triggers, was not working for the DATA ACL or
2395 for the non-SMTP ACL.
2397 78. Error message wording change in sieve.c.
2399 79. If smtp_accept_max_per_host was set, the number of connections could be
2400 restricted to fewer than expected, because the daemon was trying to set up
2401 a new connection before checking whether the processes handling previous
2402 connections had finished. The check for completed processes is now done
2403 earlier. On busy systems, this bug wouldn't be noticed because something
2404 else would have woken the daemon, and it would have reaped the completed
2407 80. If a message was submitted locally by a user whose login name contained one
2408 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2409 It caused a spool format error. I have fixed the spool reading code. A
2410 related problem was that the "from" clause in the Received: line became
2411 illegal because of the space(s). It is now covered by ${quote_local_part.
2413 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2416 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2418 83. Patch from Sieve maintainer to fix unterminated string problem in
2419 "vacation" handling.
2421 84. Some minor changes to the Linux configuration files to help with other
2422 OS variants using glibc.
2424 85. One more patch for Sieve to update vacation handling to latest spec.
2427 ----------------------------------------------------
2428 See the note above about the 4.44 and 4.50 releases.
2429 ----------------------------------------------------
2435 1. Change 4.43/35 introduced a bug that caused file counts to be
2436 incorrectly computed when quota_filecount was set in an appendfile
2439 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2440 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2443 3. Give more explanation in the error message when the command for a transport
2444 filter fails to execute.
2446 4. There are several places where Exim runs a non-Exim command in a
2447 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2448 was being done only for the command run by the queryprogram router. It is
2449 now done for all such subprocesses. The other cases are: ${run, transport
2450 filters, and the commands run by the lmtp and pipe transports.
2452 5. Some older OS have a limit of 256 on the maximum number of file
2453 descriptors. Exim was using setrlimit() to set 1000 as a large value
2454 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2455 systems. I've change it so that if it can't get 1000, it tries for 256.
2457 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2458 was an oversight, and furthermore, ever since the addition of extra
2459 controls (e.g. 4.43/32), the checks on when to allow different forms of
2460 "control" were broken. There should now be diagnostics for all cases when a
2461 control that does not make sense is encountered.
2463 7. $recipients is now available in the predata ACL (oversight).
2465 8. Tidy the search cache before the fork to do a delivery from a message
2466 received from the command line. Otherwise the child will trigger a lookup
2467 failure and thereby defer the delivery if it tries to use (for example) a
2468 cached ldap connection that the parent has called unbind on.
2470 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2471 of $address_data from the recipient verification was clobbered by the
2472 sender verification.
2474 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2475 was its contents. (It was OK if the option was not defined at all.)
2477 11. A "Completed" log line is now written for messages that are removed from
2478 the spool by the -Mrm option.
2480 12. $host_address is now set to the target address during the checking of
2481 ignore_target_hosts.
2483 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2484 being passed; this would have caused $sender_host_name to have been used if
2485 matching the list had actually called for a host name (not very likely,
2486 since this list is usually IP addresses). A host name is now passed as
2489 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2490 code that specifies a non-restarting handler (typically sigaction() in
2491 modern systems) in an attempt to fix a rare and obscure crash bug.
2493 15. Narrowed the window for a race in the daemon that could cause it to ignore
2494 SIGCHLD signals. This is not a major problem, because they are used only to
2495 wake it up if nothing else does.
2497 16. A malformed maildirsize file could cause Exim to calculate negative values
2498 for the mailbox size or file count. Odd effects could occur as a result.
2499 The maildirsize information is now recalculated if the size or filecount
2502 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2503 support for a long time. Removed HAVE_SYS_VFS_H.
2505 18. Updated exipick to current release from John Jetmore.
2507 19. Allow an empty sender to be matched against a lookup in an address list.
2508 Previously the only cases considered were a regular expression, or an
2511 20. Exim went into a mad DNS lookup loop when doing a callout where the
2512 host was specified on the transport, if the DNS lookup yielded more than
2515 21. The RFC2047 encoding function was originally intended for short strings
2516 such as real names; it was not keeping to the 75-character limit for
2517 encoded words that the RFC imposes. It now respects the limit, and
2518 generates multiple encoded words if necessary. To be on the safe side, I
2519 have increased the buffer size for the ${rfc2047: expansion operator from
2522 22. Failure to deliver a bounce message always caused it to be frozen, even if
2523 there was an errors_to setting on the router. The errors_to setting is now
2526 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2527 canonical form (fully expanded) before being placed in
2528 $sender_host_address.
2530 24. Updated eximstats to version 1.33
2532 25. Include certificate and key file names in error message when GnuTLS fails
2533 to set them up, because the GnuTLS error message doesn't include the name
2534 of the failing file when there is a problem reading it.
2536 26. Expand error message when OpenSSL has problems setting up cert/key files.
2539 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2540 (this can affect the format of dates).
2542 28. exim_tidydb, when checking for the continued existence of a message for
2543 which it has found a message-specific retry record, was not finding
2544 messages that were in split spool directories. Consequently, it was
2545 deleting retry records that should have stayed in existence.
2547 29. eximstats updated to version 1.35
2548 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2549 1.35 - bugfix such that pie charts by volume are generated correctly
2551 30. The SPA authentication driver was not abandoning authentication and moving
2552 on to the next authenticator when an expansion was forced to fail,
2553 contradicting the general specification for all authenticators. Instead it
2554 was generating a temporary error. It now behaves as specified.
2556 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2557 (the order specifies the preference for clients). The order is now AES256,
2558 AES128, 3DES, ARCFOUR128.
2560 31. Small patch to Sieve code - explicitly set From: when generating an
2563 32. Exim crashed if a remote delivery caused a very long error message to be
2564 recorded - for instance if somebody sent an entire SpamAssassin report back
2565 as a large number of 550 error lines. This bug was coincidentally fixed by
2566 increasing the size of one of Exim's internal buffers (big_buffer) that
2567 happened as part of the Exiscan merge. However, to be on the safe side, I
2568 have made the code more robust (and fixed the comments that describe what
2571 33. Some experimental protocols are using DNS PTR records for new purposes. The
2572 keys for these records are domain names, not reversed IP addresses. The
2573 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2574 leaves it alone. Component reversal etc. now happens only for IP addresses.
2577 34. The host_aton() function is supposed to be passed a string that is known
2578 to be a valid IP address. However, in the case of IPv6 addresses, it was
2579 not checking this. This is a hostage to fortune. Exim now panics and dies
2580 if the condition is not met. A case was found where this could be provoked
2581 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2582 components; fortuitously, this particular loophole had already been fixed
2583 by change 4.50/55 or 4.44/33 above.
2585 If there are any other similar loopholes, the new check in host_aton()
2586 itself should stop them being exploited. The report I received stated that
2587 data on the command line could provoke the exploit when Exim was running as
2588 exim, but did not say which command line option was involved. All I could
2589 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2590 running as the user.
2593 35. There was a buffer overflow vulnerability in the SPA authentication code
2594 (which came originally from the Samba project). I have added a test to the
2595 spa_base64_to_bits() function which I hope fixes it.
2598 36. The daemon start-up calls getloadavg() while still root for those OS that
2599 need the first call to be done as root, but it missed one case: when
2600 deliver_queue_load_max is set with deliver_drop_privilege. This is
2601 necessary for the benefit of the queue runner, because there is no re-exec
2602 when deliver_drop_privilege is set.
2604 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2605 was in use, was not putting the data itself into the right store pool;
2606 consequently, it could be overwritten for a subsequent message in the same
2607 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2610 38. Sometimes the final signoff response after QUIT could fail to get
2611 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2612 < 0 before doing a fflush(). This bug looks as though it goes back to the
2613 introduction of TLS in release 3.20, but "sometimes" must have been rare
2614 because the tests only now provoked it.
2620 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2621 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2622 Now it returns an integer. A number of calls were still expecting a T/F
2623 return. Fortuitously, in all cases, the tests worked in OK situations,
2624 which is the norm. However, things would have gone wrong on any write
2625 failures on the smtp file descriptor. This function is used when sending
2626 messages over SMTP and also when doing verify callouts.
2628 2. When Exim is called to do synchronous delivery of a locally submitted
2629 message (the -odf or -odi options), it no longer closes stderr before doing
2632 3. Implemented the mua_wrapper option.
2634 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2636 5. Implemented the functions header_remove(), header_testname(),
2637 header_add_at_position(), and receive_remove_recipient(), and exported them
2640 6. If an ACL "warn" statement specified the addition of headers, Exim already
2641 inserted X-ACL-Warn: at the start if there was no header name. However, it
2642 was not making this test for the second and subsequent header lines if
2643 there were newlines in the string. This meant that an invalid header could
2644 be inserted if Exim was badly configured.
2646 7. Allow an ACL "warn" statement to add header lines at the start or after all
2647 the Received: headers, as well as at the end.
2649 8. Added the rcpt_4xx retry error code.
2651 9. Added postmaster_mailfrom=xxx to callout verification option.
2653 10. Added mailfrom=xxxx to the callout verification option, for verify=
2656 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2657 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2659 12. Inserted some casts to stop certain compilers warning when using pointer
2660 differences as field lengths or precisions in printf-type calls (mostly
2661 affecting debugging statements).
2663 13. Added optional readline() support for -be (dynamically loaded).
2665 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2666 same clock tick as a message's arrival, so that its received time was the
2667 same as the "first fail" time on the retry record, and that message
2668 remained on the queue past the ultimate address timeout, every queue runner
2669 would try a delivery (because it was past the ultimate address timeout) but
2670 after another failure, the ultimate address timeout, which should have then
2671 bounced the address, did not kick in. This was a "< instead of <=" error;
2672 in most cases the first failure would have been in the next clock tick
2673 after the received time, and all would be well.
2675 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2676 being recognized when the domain list was tested by the match_domain
2677 condition in an expansion string.
2679 16. Added the ${str2b64: operator.
2681 17. Exim was always calling setrlimit() to set a large limit for the number of
2682 processes, without checking whether the existing limit was already
2683 adequate. (It did check for the limit on file descriptors.) Furthermore,
2684 errors from getrlimit() and setrlimit() were being ignored. Now they are
2685 logged to the main and panic logs, but Exim does carry on, to try to do its
2686 job under whatever limits there are.
2688 18. Imported PCRE 5.0.
2690 19. Trivial typo in log message " temporarily refused connection" (the leading
2693 20. If the log selector return_path_on_delivery was set and an address was
2694 redirected to /dev/null, the delivery process crashed because it assumed
2695 that a return path would always be set for a "successful" delivery. In this
2696 case, the whole delivery is bypassed as an optimization, and therefore no
2699 21. Internal re-arrangement: the function for sending a challenge and reading
2700 a response while authentication was assuming a zero-terminated challenge
2701 string. It's now changed to take a pointer and a length, to allow for
2702 binary data in such strings.
2704 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2706 23. Exim was not respecting finduser_retries when seeking the login of the
2707 uid under which it was called; it was always trying 10 times. (The default
2708 setting of finduser_retries is zero.) Also, it was sleeping after the final
2709 failure, which is pointless.
2711 24. Implemented tls_on_connect_ports.
2713 25. Implemented acl_smtp_predata.
2715 26. If the domain in control=submission is set empty, Exim assumes that the
2716 authenticated id is a complete email address when it generates From: or
2717 Sender: header lines.
2719 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2720 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2721 chown and chgrp in /bin and hostname in /usr/bin.
2723 28. Exim was keeping the "process log" file open after each use, just as it
2724 does for the main log. This opens the possibility of it remaining open for
2725 long periods when the USR1 signal hits a daemon. Occasional processlog
2726 errors were reported, that could have been caused by this. Anyway, it seems
2727 much more sensible not to leave this file open at all, so that is what now
2730 29. The long-running daemon process does not normally write to the log once it
2731 has entered its main loop, and it closes the log before doing so. This is
2732 so that log files can straightforwardly be renamed and moved. However,
2733 there are a couple of unusual error situations where the daemon does write
2734 log entries, and I had neglected to close the log afterwards.
2736 30. The text of an SMTP error response that was received during a remote
2737 delivery was being truncated at 512 bytes. This is too short for some of
2738 the long messages that one sometimes sees. I've increased the limit to
2741 31. It is now possible to make retry rules that apply only when a message has a
2742 specific sender, in particular, an empty sender.
2744 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2745 it possible to be selective about when SMTP synchronization is enforced.
2747 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2749 32. Implemented hosts_connection_nolog.
2751 33. Added an ACL for QUIT.
2753 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2756 35. Added mailbox_size and mailbox_filecount to appendfile.
2758 36. Added control = no_multiline_responses to ACLs.
2760 37. There was a bug in the logic of the code that waits for the clock to tick
2761 in the case where the clock went backwards by a substantial amount such
2762 that the microsecond fraction of "now" was more than the microsecond
2763 fraction of "then" (but the whole seconds number was less).
2765 38. Added support for the libradius Radius client library this is found on
2766 FreeBSD (previously only the radiusclient library was supported).
2772 1. When certain lookups returned multiple values in the form name=value, the
2773 quoting of the values was not always being done properly. Specifically:
2774 (a) If the value started with a double quote, but contained no whitespace,
2776 (b) If the value contained whitespace other than a space character (i.e.
2777 tabs or newlines or carriage returns) it was not quoted.
2778 This fix has been applied to the mysql and pgsql lookups by writing a
2779 separate quoting function and calling it from the lookup code. The fix
2780 should probably also be applied to nisplus, ibase and oracle lookups, but
2781 since I cannot test any of those, I have not disturbed their existing code.
2783 2. A hit in the callout cache for a specific address caused a log line with no
2784 reason for rejecting RCPT. Now it says "Previous (cached) callout
2785 verification failure".
2787 3. There was an off-by-one bug in the queryprogram router. An over-long
2788 return line was truncated at 256 instead of 255 characters, thereby
2789 overflowing its buffer with the terminating zero. As well as fixing this, I
2790 have increased the buffer size to 1024 (and made a note to document this).
2792 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2793 when Exim is waiting for an SMTP response from a remote server, Exim
2794 restarts its select() call on the socket, thereby resetting its timeout.
2795 This is not a problem when such interrupts are rare. Somebody set up a cron
2796 job to run exiwhat every 2 minutes, which is less than the normal select()
2797 timeout (5 or 10 minutes). This meant that the select() timeout never
2798 kicked in because it was always reset. I have fixed this by comparing the
2799 time when an interrupt arrives with the time at the start of the first call
2800 to select(). If more time than the timeout has elapsed, the interrupt is
2801 treated as a timeout.
2803 5. Some internal re-factoring in preparation for the addition of Sieve
2804 extensions (by MH). In particular, the "personal" test is moved to a
2805 separate function, and given an option for scanning Cc: and Bcc: (which is
2806 not set for Exim filters).
2808 6. When Exim created an email address using the login of the caller as the
2809 local part (e.g. when creating a From: or Sender: header line), it was not
2810 quoting the local part when it contained special characters such as @.
2812 7. Installed new OpenBSD configuration files.
2814 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2815 try to make them clearer.
2817 9. Callout options, other than the timeout value, were being ignored when
2818 verifying sender addresses in header lines. For example, when using
2820 verify = header_sender/callout=no_cache
2822 the cache was (incorrectly) being used.
2824 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2825 only the Cygwin environment.
2827 11. When return_path_on_delivery was set as a log selector, if different remote
2828 addresses in the same message used different return paths and parallel
2829 remote delivery occurred, the wrong values would sometimes be logged.
2830 (Whenever a remote delivery process finished, the return path value from
2831 the most recently started remote delivery process was logged.)
2833 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2834 lines when AUTH and/or TLS are in use. This is the "received protocol"
2835 field. Exim used to use "asmtp" for authenticated SMTP, without any
2836 indication (in the protocol name) for TLS use. Now it follows the RFC and
2837 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2838 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2839 names appear in log lines as well as in Received: header lines.
2841 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2842 extensions, and comparison tests, and to fix some bugs.
2844 14. Changes to the "personal" filter test:
2846 (1) The test was buggy in that it was just doing the equivalent of
2847 "contains" tests on header lines. For example, if a user's address was
2848 anne@some.where, the "personal" test would incorrectly be true for
2850 To: susanne@some.where
2852 This test is now done by extracting each address from the header in turn,
2853 and checking the entire address. Other tests that are part of "personal"
2854 are now done using regular expressions (for example, to check local parts
2855 of addresses in From: header lines).
2857 (2) The list of non-personal local parts in From: addresses has been
2858 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2859 taken from the Sieve specification recommendations.
2861 (3) If the message contains any header line starting with "List-" it is
2862 treated as non-personal.
2864 (4) The test for "circular" in the Subject: header line has been removed
2865 because it now seems ill-conceived.
2867 15. Minor typos in src/EDITME comments corrected.
2869 16. Installed latest exipick from John Jetmore.
2871 17. If headers_add on a router specified a text string that was too long for
2872 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2873 of string_sprintf() is now avoided.
2875 18. $message_body_size was not set (it was always zero) when running the DATA
2876 ACL and the local_scan() function.
2878 19. For the "mail" command in an Exim filter, no default was being set for
2879 the once_repeat time, causing a random time value to be used if "once" was
2880 specified. (If the value happened to be <= 0, no repeat happened.) The
2881 default is now 0s, meaning "never repeat". The "vacation" command was OK
2882 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2883 (I found it when inspecting the code).
2885 20. There is now an overall timeout for performing a callout verification. It
2886 defaults to 4 times the callout timeout, which applies to individual SMTP
2887 commands during the callout. The overall timeout applies when there is more
2888 than one host that can be tried. The timeout is checked before trying the
2889 next host. This prevents very long delays if there are a large number of
2890 hosts and all are timing out (e.g. when the network connections are timing
2891 out). The value of the overall timeout can be changed by specifying an
2892 additional sub-option for "callout", called "maxwait". For example:
2894 verify = sender/callout=5s,maxwait=20s
2896 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2897 to the end before writing, but this should make it even safer).
2899 22. Exim was forgetting that it had advertised PIPELINING for the second and
2900 subsequent messages on an SMTP connection. It was also not resetting its
2901 memory on STARTTLS and an internal HELO.
2903 23. When Exim logs an SMTP synchronization error within a session, it now
2904 records whether PIPELINING has been advertised or not.
2906 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2907 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2908 rather than long int.
2910 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2912 26. Added the never_mail option to autoreply.
2918 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2919 crash if the getsockname() call failed; this can happen if a connection is
2920 closed very soon after it is established. The problem was simply in the
2921 order in which certain operations were done, causing Exim to try to write
2922 to the SMTP stream before it had set up the file descriptor. The bug has
2923 been fixed by making things happen in the correct order.
2929 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2930 before the connection was closed, thus losing the rejection response.
2932 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2933 some early Solaris releases, but causes trouble in current releases where
2934 socklen_t is defined.
2936 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2939 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2942 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2943 that releasing the top of it at the end releases what was used for sub-
2944 expansions (unless the block got too big). However, discard this block if
2945 the first thing is a variable or header, so that we can use its block when
2946 it is dynamic (useful for very large $message_headers, for example).
2948 6. Lookups now cache *every* query, not just the most recent. A new, separate
2949 store pool is used for this. It can be recovered when all lookup caches are
2950 flushed. Lookups now release memory at the end of their result strings.
2951 This has involved some general refactoring of the lookup sources.
2953 7. Some code has been added to the store_xxx() functions to reduce the amount
2954 of flapping under certain conditions.
2956 8. log_incoming_interface used to affect only the <= reception log lines. Now
2957 it causes the local interface and port to be added to several more SMTP log
2958 lines, for example "SMTP connection from", and rejection lines.
2960 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2962 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2964 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2965 could be overwritten at the end of the current message (or the start of a
2966 new message if it was set in a HELO ACL). The value is now preserved for
2967 the duration of the SMTP connection.
2969 12. If a transport had a headers_rewrite setting, and a matching header line
2970 contained an unqualified address, that address was qualified, even if it
2971 did not match any rewriting rules. The underlying bug was that the values
2972 of the flags that permit the existence of unqualified sender and recipient
2973 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2974 non-local messages, and by -bnq for local messages) were not being
2975 preserved with the message after it was received.
2977 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2978 "next input=" as part of the text comprising the host identity instead of
2979 the correct text. The code was using the same buffer for two different
2980 strings. However, depending on which order the printing function evaluated
2981 its arguments, the bug did not always show up. Under Linux, for example, my
2982 test suite worked just fine.
2984 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2985 doesn't work with some older versions of Perl. It has been changed to "my",
2986 which in any case is probably the better facility to use.
2988 15. A really picky compiler found some instances of statements for creating
2989 error messages that either had too many or two few arguments for the format
2992 16. The size of the buffer for calls to the DNS resolver has been increased
2993 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2994 for addresses that have a lot of PTR records. This alleviates a problem; it
2995 does not fully solve it.
2997 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2998 buffer now truncates the list and logs the incident, which is the same
2999 action as happens when Exim is looking up a host name and its aliases.
3000 Previously in this situation something unpredictable would happen;
3001 sometimes it was "internal error: store_reset failed".
3003 18. If a server dropped the connection unexpectedly when an Exim client was
3004 using GnuTLS and trying to read a response, the client delivery process
3005 crashed while trying to generate an error log message.
3007 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3008 string, for example:
3010 warn message = H1: something\nH2: something
3012 the text was added as a single header line from Exim's point of view
3013 though it ended up OK in the delivered message. However, searching for the
3014 second and subsequent header lines using $h_h2: did not work. This has been
3015 fixed. Similarly, if a system filter added multiple headers in this way,
3016 the routers could not see them.
3018 20. Expanded the error message when iplsearch is called with an invalid key to
3019 suggest using net-iplsearch in a host list.
3021 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3022 ACLs are no longer actually imposed (and a message to that effect is
3025 22. If a "gecos" field in a passwd entry contained escaped characters, in
3026 particular, if it contained a \" sequence, Exim got it wrong when building
3027 a From: or a Sender: header from that name. A second bug also caused
3028 incorrect handling when an unquoted " was present following a character
3029 that needed quoting.
3031 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3032 was not being matched caselessly.
3034 24. Arranged for all hyphens in the exim.8 source to be escaped with
3037 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3038 properly. Recipient callout cache records were still being keyed to include
3039 the sender, even when use_sender was set false. This led to far more
3040 callouts that were necessary. The sender is no longer included in the key
3041 when use_sender is false.
3043 26. Added "control = submission" modifier to ACLs.
3045 27. Added the ${base62d: operator to decode base 62 numbers.
3047 28. dnsdb lookups can now access SRV records.
3049 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3050 the configuration file.
3052 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3053 (-v) mode. This makes the output for a verbose queue run more intelligible.
3055 31. Added a use_postmaster feature to recipient callouts.
3057 32. Added the $body_zerocount variable, containing the number of binary zero
3058 bytes in the message body.
3060 33. The time of last modification of the "new" subdirectory is now used as the
3061 "mailbox time last read" when there is a quota error for a maildir
3064 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3066 35. Added +ignore_unknown as a special item in host lists.
3068 36. Code for decoding IPv6 addresses in host lists is now included, even if
3069 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3070 address was recognized as an IP address, but was then not correctly decoded
3071 into binary, causing unexpected and incorrect effects when compared with
3078 1. Very minor rewording of debugging text in manualroute to say "list of
3079 hosts" instead of "hostlist".
3081 2. If verify=header_syntax was set, and a header line with an unqualified
3082 address (no domain) and a large number of spaces between the end of the
3083 name and the colon was received, the reception process suffered a buffer
3084 overflow, and (when I tested it) crashed. This was caused by some obsolete
3085 code that should have been removed. The fix is to remove it!
3087 3. When running in the test harness, delay a bit after writing a bounce
3088 message to get a bit more predictability in the log output.
3090 4. Added a call to search_tidyup() just before forking a reception process. In
3091 theory, someone could use a lookup in the expansion of smtp_accept_max_
3092 per_host which, without the tidyup, could leave open a database connection.
3094 5. Added the variables $recipient_data and $sender_data which get set from a
3095 lookup success in an ACL "recipients" or "senders" condition, or a router
3096 "senders" option, similar to $domain_data and $local_part_data.
3098 6. Moved the writing of debug_print from before to after the "senders" test
3101 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3102 problems for message scanning, either using a data ACL, or using
3103 local_scan() because the Received: header was not generated till after they
3104 were called (in order to set the time as the time of reception completion).
3105 I have revised the way this works. The header is now generated after the
3106 body is received, but before the ACL or local_scan() are called. After they
3107 are run, the timestamp in the header is updated.
3113 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3114 before starting a queue runner without re-exec. This happened only when
3115 deliver_drop_privilege was set or when the Exim user was set to root. The
3116 effect of the bug was that timeouts during subsequent deliveries caused
3117 crashes instead of being properly handled. The handler is now left at its
3118 default (and expected) setting.
3120 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3121 message, again when deliver_drop_privilege is set or Exim is run as root.
3122 The bug described in (1) was not present in this case, but the tidying up
3123 of the other signals was missing. I have made the two cases consistent.
3125 3. The ignore_target_hosts setting on a manualroute router was being ignored
3126 for hosts that were looked up using the /MX notation.
3128 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3131 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3132 operated on the sender address. After changing the $sender_address to <>
3133 for the sender address verify, Exim was re-instated it as the original
3134 (before rewriting) address, but remembering that it had rewritten it, so it
3135 wasn't rewriting it again. This bug also had the effect of breaking the
3136 sender address verification caching when the sender address was rewritten.
3138 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3139 This has been changed so that if the ip literal address matches
3140 ignore_target_hosts, the router declines.
3142 7. Added expansion conditions match_domain, match_address, and match_local_
3143 part (NOT match_host).
3145 8. The placeholder for the Received: header didn't have a length field set.
3147 9. Added code to Exim itself and to exim_lock to test for a specific race
3148 condition that could lead to file corruption when using MBX delivery. The
3149 issue is with the lockfile that is created in /tmp. If this file is removed
3150 after a process has opened it but before that process has acquired a lock,
3151 there is the potential for a second process to recreate the file and also
3152 acquire a lock. This could lead to two Exim processes writing to the file
3153 at the same time. The added code performs the same test as UW imapd; it
3154 checks after acquiring the lock that its file descriptor still refers to
3155 the same named file.
3157 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3158 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3161 11. Added the smtp_active_hostname option. If used, this will typically be made
3162 to depend on the incoming interface address. Because $interface_address is
3163 not set up until the daemon has forked a reception process, error responses
3164 that can happen earlier (such as "too many connections") no longer contain
3167 12. If an expansion in a condition on a "warn" statement fails because a lookup
3168 defers, the "warn" statement is abandoned, and the next ACL statement is
3169 processed. Previously this caused the whole ACL to be aborted.
3171 13. Added the iplsearch lookup type.
3173 14. Added ident_timeout as a log selector.
3175 15. Added tls_certificate_verified as a log selector.
3177 16. Added a global option tls_require_ciphers (compare the smtp transport
3178 option of the same name). This controls incoming TLS connections.
3180 17. I finally figured out how to make tls_require_ciphers do a similar thing
3181 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3182 before starting the TLS session.
3184 18. Tabs are now shown as \t in -bP output.
3186 19. If the log selector return_path_on_delivery was set, Exim crashed when
3187 bouncing a message because it had too many Received: header lines.
3189 20. If two routers both had headers_remove settings, and the first one included
3190 a superfluous trailing colon, the final name in the first list and the
3191 first name in the second list were incorrectly joined into one item (with a
3192 colon in the middle).
3198 1. Added -C and -D options to the exinext utility, mainly to make it easier
3199 to include in the automated testing, but these could be helpful when
3200 multiple configurations are in use.
3202 2. The exinext utility was not formatting the output nicely when there was
3203 an alternate port involved in the retry record key, nor when there was a
3204 message id as well (for retries that were specific to a specific message
3205 and a specific host). It was also confused by IPv6 addresses, because of
3206 the additional colons they contain. I have fixed the IPv4 problem, and
3207 patched it up to do a reasonable job for IPv6.
3209 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3210 delivery, the log line now contains "pipelined" if PIPELINING was used.
3212 4. An SMTP transport process used to panic and die if the bind() call to set
3213 an explicit outgoing interface failed. This has been changed; it is now
3214 treated in the same way as a connect() failure.
3216 5. A reference to $sender_host_name in the part of a conditional expansion
3217 that was being skipped was still causing a DNS lookup. This no longer
3220 6. The def: expansion condition was not recognizing references to header lines
3221 that used bh_ and bheader_.
3223 7. Added the _cache feature to named lists.
3225 8. The code for checking quota_filecount in the appendfile transport was
3226 allowing one more file than it should have been.
3228 9. For compatibility with Sendmail, the command line option
3236 and sets the incoming protocol and host name (for trusted callers). The
3237 host name and its colon can be omitted when only the protocol is to be set.
3238 Note the Exim already has two private options, -pd and -ps, that refer to
3239 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3240 "s", but I don't think that's a major issue.
3242 10. A number of refactoring changes to the code, none of which should affect
3245 (a) The number of logging options was getting close to filling up the
3246 32-bit word that was used as a bit map. I have split them into two classes:
3247 those that are passed in the argument to log_write(), and those that are
3248 only ever tested independently outside of that function. These are now in
3249 separate 32-bit words, so there is plenty of room for expansion again.
3250 There is no change in the user interface or the logging behaviour.
3252 (b) When building, for example, log lines, the code previously used a
3253 macro that called string_cat() twice, in order to add two strings. This is
3254 not really sufficiently general. Furthermore, there was one instance where
3255 it was actually wrong because one of the argument was used twice, and in
3256 one call a function was used. (As it happened, calling the function twice
3257 did not affect the overall behaviour.) The macro has been replaced by a
3258 function that can join an arbitrary number of extra strings onto a growing
3261 (c) The code for expansion conditions now uses a table and a binary chop
3262 instead of a serial search (which was left over from when there were very
3263 few conditions). Also, it now recognizes conditions like "pam" even when
3264 the relevant support is not compiled in: a suitably worded error message is
3265 given if an attempt is made to use such a condition.
3267 11. Added ${time_interval:xxxxx}.
3269 12. A bug was causing one of the ddress fields not to be passed back correctly
3270 from remote delivery subprocesses. The field in question was not being
3271 subsequently used, so this caused to problems in practice.
3273 13. Added new log selectors queue_time and deliver_time.
3275 14. Might have fixed a bug in maildirsizefile handling that threw up
3276 "unexpected character" debug warnings, and recalculated the data
3277 unnecessarily. In any case, I expanded the warning message to give more
3280 15. Added the message "Restricted characters in address" to the statements in
3281 the default ACL that block characters like @ and % in local parts.
3283 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3284 Three changes have been made:
3286 (a) There was a serious bug; a negative response to MAIL caused the whole
3287 recipient domain to be cached as invalid, thereby blocking all messages
3288 to all local parts at the same domain, from all senders. This bug has
3289 been fixed. The domain is no longer cached after a negative response to
3290 MAIL if the sender used is not empty.
3292 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3295 (c) A new callout option, "use_sender" has been added for people who want
3296 the modified behaviour.
3302 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3305 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3306 indeed breaks things for older releases.
3308 3. Added additional logging to the case where there is a problem reading data
3309 from a filter that is running in a subprocess using a pipe, in order to
3310 try to track down a specific problem.
3312 4. Testing facility fudge: when running in the test harness and attempting
3313 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3314 getting "No route to host". Convert this to a timeout.
3316 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3319 6. Some OS don't have socklen_t but use size_t instead. This affects the
3320 fifth argument of getsockopt() amongst other things. This is now
3321 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3322 can be set for individual OS. I have set it for SunOS5, OSF1, and
3323 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3324 some earlier ones do not.
3326 7. Change 4.30/15 was not doing the test caselessly.
3328 8. The standard form for an IPv6 address literal was being rejected by address
3329 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3330 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3331 this, as well as the form without the "IPv6" on the front (but only when
3332 address literals are enabled, of course).
3334 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3336 10. Exim crashed if a message with an empty sender address specified by -f
3337 encountered a router with an errors_to setting. This could be provoked only
3338 by a command such as
3342 where an empty string was supplied; "<>" did not hit this bug.
3344 11. Installed PCRE release 4.5.
3346 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3347 remained set. It is now erased.
3349 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3350 times from message ids (which are base 36 rather than the normal 62).
3352 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3353 were being counted as actual protocol errors, and logged if the log
3354 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3355 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3356 and DATA following a set of rejected RCPTs do not count as protocol errors.
3357 In other words, Exim assumes they were pipelined, though this may not
3358 actually be the case. Of course, in all cases the client gets an
3359 appropriate error code.
3361 15. If a lookup fails in an ACL condition, a message about the failure may
3362 be available; it is used if testing the ACL cannot continue, because most
3363 such messages specify what the cause of the deferral is. However, some
3364 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3365 that caused an old message to be retained and used if a later statement
3366 caused a defer, replacing the real cause of the deferral.
3368 16. If an IP address had so many PTR records that the DNS lookup buffer
3369 was not large enough to hold them, Exim could crash while trying to process
3370 the truncated data. It now detects and logs this case.
3372 17. Further to 4.21/58, another change has been made: if (and only if) the
3373 first line of a message (the first header line) ends with CRLF, a bare LF
3374 in a subsequent header line has a space inserted after it, so as not to
3375 terminate the header.
3377 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3378 unnecessarily, used atoi() instead of strtol(), and didn't check the
3379 termination when getting file sizes from file names by regex.
3381 19. Completely re-implemented the support for maildirsize files, in the light
3382 of a number of problems with the previous contributed implementation
3383 (4.30/29). In particular:
3385 . If the quota is zero, the maildirsize file is maintained, but no quota is
3388 . If the maildir directory does not exist, it is created before any attempt
3389 to write a maildirsize file.
3391 . The quota value in the file is just a cache; if the quota is changed in
3392 the transport, the new value overrides.
3394 . A regular expression is available for excluding directories from the
3397 20. The autoreply transport checks the characters in options that define the
3398 message's headers; it allows continued headers, but it was checking with
3399 isspace() after an embedded newline instead of explicitly looking for a
3402 21. If all the "regular" hosts to which an address was routed had passed their
3403 expiry times, and had not reached their retry times, the address was
3404 bounced, even if fallback hosts were defined. Now Exim should go on to try
3407 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3408 equivalent code in the SMTP transport. Some hosts send humungous responses
3409 to HELO/EHLO, more than 1024 it seems.
3411 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3412 gives compiler warnings in some environments. I've now done it "properly",
3415 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3416 (because of problems with the built-in one) was declared to return uschar *
3417 instead of char *, causing compiler failure.
3419 25. Fixed a file descriptor leak when processing alias/forward files.
3421 26. Fixed a minor format string issue in dbfn.c.
3423 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3425 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3426 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3429 29. When a sender address is verified, it is cached, to save repeating the test
3430 when there is more than one recipient in a message. However, when the
3431 verification involves a callout, it is possible for different callout
3432 options to be set for different recipients. It is too complicated to keep
3433 track of this in the cache, so now Exim always runs a verification when a
3434 callout is required, relying on the callout cache for the optimization.
3435 The overhead is duplication of the address routing, but this should not be
3438 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3439 to be verified with callout=postmaster, and the main callout worked but the
3440 postmaster check failed, the verification correctly failed. However, if a
3441 subsequent RCPT command asked for sender verification *without* the
3442 postmaster check, incorrect caching caused this verification also to fail,
3445 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3446 it was not caching the DNS options (qualify_single, search_parents) that
3447 were used when the lookup failed. A subsequent lookup with different
3448 options therefore always gave the same answer, though there were cases
3449 where it should not have. (Example: a "domains = !$mx_any" option on a
3450 dnslookup router: the "domains" option is always processed without any
3451 widening, but the router might have qualify_single set.) Now Exim uses the
3452 cached value only when the same options are set.
3454 32. Added John Jetmore's "exipick" utility to the distribution.
3456 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3457 than a timeout (e.g. a certificate is required, and is not provided), an
3458 Exim server now closes the connection immediately. Previously it waited for
3459 the client to close - but if the client is SSL, it seems that they each
3460 wait for each other, leading to a delay before one of them times out.
3462 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3463 maintained 0.8.x compatibility because I don't think many are using it, and
3464 it is clearly obsolete.
3466 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3469 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3470 client certificate was expired. A simple patch fixes this, though I don't
3471 understand the full logic of why the verify callback is called multiple
3474 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3475 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3476 which causes problems with some clients (such as the Certicom SSL Plus
3477 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3478 disables the coutermeasure allowing Eudora to connect."
3480 38. Exim was not checking that a write() to a log file succeeded. This could
3481 lead to Bad Things if a log got too big, in particular if it hit a file
3482 size limit. Exim now panics and dies if it cannot write to a log file, just
3483 as it does if it cannot open a log file.
3485 39. Modified OS/Makefile-Linux so that it now contains
3487 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3489 The two -D definitions ensure that Exim is compiled with large file
3490 support, which makes it possible to handle log files that are bigger than
3493 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3494 instance) a domain was checked against a named list that involved a lookup,
3495 causing $domain_data to be set, then another domain was checked against the
3496 same list, then the first domain was re-checked, the value of $domain_data
3497 after the final check could be wrong. In particular, if the second check
3498 failed, it could be set empty. This bug probably also applied to
3501 41. The strip_trailing_dot option was not being applied to the address given
3502 with the -f command-line option.
3504 42. The code for reading a message's header from the spool was incrementing
3505 $received_count, but never initializing it. This meant that the value was
3506 incorrect (doubled) while delivering a message in the same process in which
3507 it was received. In the most common configuration of Exim, this never
3508 happens - a fresh exec is done - but it can happen when
3509 deliver_drop_privilege is set.
3511 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3512 it now includes up to 150 characters of the unexpected data in the log
3515 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3516 and building data strings. The size of both of these buffers was 10 000
3517 bytes - far larger than anybody would *ever* want, thought I. Needless to
3518 say, somebody hit the limit. I have increased the maximum line length to
3519 20 000 and the maximum data length of concatenated lines to 100 000. I have
3520 also fixed two bugs, because there was no checking on these buffers. Tsk,
3521 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3522 buffer is too small.
3524 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3525 lsearch lookups. Now it does.
3527 46. When parsing a route_list item in a manualroute router, a fixed-length
3528 buffer was used for the list of hosts. I made this 1024 bytes long,
3529 thinking that nobody would ever have a list of hosts that long. Wrong.
3530 Somebody had a whole pile of complicated expansion conditions, and the
3531 string was silently truncated, leading to an expansion error. It turns out
3532 that it is easier to change to an unlimited length (owing to other changes
3533 that have happened since this code was originally written) than to build
3534 structure for giving a limitation error. The length of the item that
3535 expands into the list of hosts is now unlimited.
3537 47. The lsearch lookup could not handle data where the length of text line was
3538 more than 4095 characters. Such lines were truncated, leading to shortened
3539 data being returned. It should now handle lines of any length.
3541 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3542 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3545 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3546 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3547 no longer happens in comment lines. A list of replacements is now placed
3548 at the head of all of the source files, except those whose only change is
3549 to replace PERL_COMMAND in the very first #! line.
3551 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3552 messages on the queue, with a bottom-up merge sort, using code contributed
3553 by Michael Haardt. This should make operations like -bp somewhat faster on
3554 large queues. It won't affect queue runners, except when queue_run_in_order
3557 51. Installed eximstats 1.31 in the distribution.
3559 52. Added support for SRV lookups to the dnslookup router.
3561 53. If an ACL referred to $message_body or $message_body_end, the value was not
3562 reset for any messages that followed in the same SMTP session.
3564 54. The store-handling optimization for building very long strings was not
3565 differentiating between the different store pools. I don't think this
3566 actually made any difference in practice, but I've tidied it.
3568 55. While running the routers to verify a sender address, $sender_address
3569 was still set to the sender address. This is wrong, because when routing to
3570 send a bounce to the sender, it would be empty. Therefore, I have changed
3571 it so that, while verifying a sender address, $sender_address is set to <>.
3572 (There is no change to what happens when verifying a recipient address.)
3574 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3575 target A or AAAA records (if not already returned) without resetting the
3576 qualify_single or search_parents options of the DNS resolver. These are
3577 inappropriate in this case because the targets of MX and SRV records must
3578 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3579 target that, when qualified, matched something in the local domain. These
3580 two options are now turned off when doing these lookups.
3582 57. It seems that at least some releases of Reiserfs (which does not have the
3583 concept of a fixed number of inodes) returns zero and not -1 for the
3584 number of available inodes. This interacted badly with check_spool_inodes,
3585 which assumed that -1 was the "no such thing" setting. What I have done is
3586 to check that the total number of inodes is greater than zero before doing
3587 the test of how many are available.
3589 58. When a "warn" ACL statement has a log_message modifier, the message is
3590 remembered, and not repeated. This is to avoid a lot of repetition when a
3591 message has many recipients that cause the same warning to be written.
3592 Howewer, Exim was preserving the list of already written lines for an
3593 entire SMTP session, which doesn't seem right. The memory is now reset if a
3594 new message is started.
3596 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3597 debugging output unless log_rewrite was also set.
3599 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3600 of (int)(handle) when we know that handle contains (void *)(-1).
3602 61. The Exim daemon panic-logs an error return when it closes the incoming
3603 connection. However "connection reset by peer" seems to be common, and
3604 isn't really an error worthy of noting specially, so that particular error
3607 62. When Exim is trying to find all the local interfaces, it used to panic and
3608 die if the ioctl to get the interface flags failed. However, it seems that
3609 on at least one OS (Solaris 9) it is possible to have an interface that is
3610 included in the list of interfaces, but for which you get a failure error
3611 for this call. This happens when the interface is not "plumbed" into a
3612 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3613 failure of the "get flags" call assumes that the interface is down.
3615 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3616 makes life easier for people who are doing arithmetic on fields extracted
3617 from dates, where you often get leading zeros that should not be
3618 interpreted as octal.
3620 64. Added qualify_domain to the redirect router, to override the global
3623 65. If a pathologically long header line contained very many addresses (the
3624 report of this problem mentioned 10 000) and each of them was rewritten,
3625 Exim could use up a very large amount of memory. (It kept on making new
3626 copies of the header line as it rewrote, and never released the old ones.)
3627 At the expense of a bit more processing, the header rewriting function has
3628 been changed so that it no longer eats memory in this way.
3630 66. The generation of the Received: header has been moved from the time that a
3631 message starts to be received, to the time that it finishes. The timestamp
3632 in the Received: header should now be very close to that of the <= log
3633 line. There are two side-effects of this change:
3635 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3636 logged header lines no longer include the local Received: line, because
3637 it has not yet been created. The same applies to a copy of the message
3638 that is returned to a non-SMTP sender when a message is rejected.
3640 (b) When a filter file is tested using -bf, no additional Received: header
3641 is added to the test message. After some thought, I decided that this
3644 This change does not affect the value of $received_for. It is still set
3645 after address rewriting, but before local_scan() is called.
3647 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3649 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3650 gave an unhelpful panic error message, and a defer error. I have managed to
3651 change this behaviour so that it now rejects any supplied certificate,
3652 which seems right, as the list of acceptable certificates is empty.
3654 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3655 gave an unhelpful defer error. I have not managed to make this reject any
3656 supplied certificates, but the error message it gives is "no certificate
3657 supplied", which is not helpful.
3659 70. exigrep's output now also includes lines that are not associated with any
3660 message, but which match the given pattern. Implemented by a patch from
3661 Martin Sluka, which also tidied up the Perl a bit.
3663 71. Recipient callout verification, like sender verification, was using <> in
3664 the MAIL FROM command. This isn't really the right thing, since the actual
3665 sender may affect whether the remote host accepts the recipient or not. I
3666 have changed it to use the actual sender in the callout; this means that
3667 the cache record is now keyed on a recipient/sender pair, not just the
3668 recipient address. There doesn't seem to be a real danger of callout loops,
3669 since a callout by the remote host to check the sender would use <>.
3670 [SEE ABOVE: changed after hitting problems.]
3672 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3673 temporary errors. However, in the case of such a code being given after
3674 the end of a data transmission (i.e. after ".") Exim was failing to write
3675 a retry record for the message. (Yes, there was some broken host that was
3676 actually sending 8xx at this point.)
3678 73. An unknown lookup type in a host list could cause Exim to panic-die when
3679 the list was checked. (An example that provoked this was putting <; in the
3680 middle of a list instead of at the start.) If this happened during a DATA
3681 ACL check, a -D file could be left lying around. This kind of configuration
3682 error no longer causes Exim to die; instead it causes a defer errror. The
3683 incident is still logged to the main and panic logs.
3685 74. Buglet left over from Exim 3 conversion. The message "too many messages
3686 in one connection" was written to the rejectlog but not the mainlog, except
3687 when address rewriting (yes!) was being logged.
3689 75. Added write_rejectlog option.
3691 76. When a system filter was run not as root (that is, when system_filter_user
3692 was set), the values of the $n variables were not being returned to the
3693 main process; thus, they were not subsequently available in the $sn
3696 77. Added +return_path_on_delivery log selector.
3698 78. A connection timeout was being treated differently from recipients deferred
3699 when testing hosts_max_try with a message that was older than the host's
3700 retry timeout. (The host should not be counted, thus allowing all hosts to
3701 be tried at least once before bouncing.) This may have been the cause of an
3702 occasionally reported bug whereby a message would remain on the queue
3703 longer than the retry timeout, but would be bounced if a delivery was
3704 forced. I say "may" because I never totally pinned down the problem;
3705 setting up timeout/retry tests is difficult. See also the next item.
3707 79. The ultimate address timeout was not being applied to errors that involved
3708 a combination of host plus message (for example, a timeout on a MAIL
3709 command). When an address resolved to a number of possible hosts, and they
3710 were not all tried for each delivery (e.g. because of hosts_max_try), a
3711 message could remain on the queue longer than the retry timeout.
3713 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3716 81. Fixed an obscure SMTP outgoing bug which required at least the following
3717 conditions: (a) there was another message waiting for the same server;
3718 (b) the server returned 5xx to all RCPT commands in the first message so
3719 that the message was not completed; (c) the server dropped the connection
3720 or gave a negative response to the RSET that Exim sends to abort the
3721 transaction. The observed case was a dropped connection after DATA that had
3722 been sent in pipelining mode. That is, the server had advertised PIPELINING
3723 but was not implementing it correctly. The effect of the bug was incorrect
3724 behaviour, such as trying another host, and this could lead to a crash.
3730 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3731 and daemon.c were passed as pointers to ints; they should have been
3732 pointers to socklen_t variables (which are typically unsigned ints).
3734 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3737 3. Fixed a really odd bug that affected only the testing scheme; patching a
3738 certain fixed string in the binary changed the value of another string that
3739 happened to be identical to the end of the original first string.
3741 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3742 name", it returns that address as the IP address. On some operating
3743 systems (e.g. Solaris), it also passes back the IP address string as the
3744 "host name". However, on others (e.g. Linux), it passes back an empty
3745 string. Exim wasn't checking for this, and was changing the host name to an
3746 empty string, assuming it had been canonicized.
3748 5. Although rare, it is permitted to have more than one PTR record for a given
3749 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3750 all the names associated with an address, because they do in Solaris.
3751 However, it seems that they do not in Linux for data that comes from the
3752 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3753 I found this out when I moved to a new Linux workstation and tried to run
3754 the Exim test suite.
3756 To get round this problem I have changed the code so that it now does its
3757 own call to the DNS to look up PTR records when searching for a host name.
3758 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3759 addresses that are only in /etc/hosts are still found.
3761 This behaviour is, however, controlled by an option called host_lookup_
3762 order, which defaults to "bydns:byaddr". If people want to use the other
3763 order, or indeed, just use one or the other means of lookup, they can
3764 specify it in this variable.
3766 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3767 some operating systems, this comes back from gethostbyaddr() as an empty
3768 string, and this is what Exim used to test for. However, it seems that in
3769 other systems, "." is yielded. Exim now tests for this case too.
3771 7. The values of check_spool_space and check_log_space are now held internally
3772 as a number of kilobytes instead of an absolute number of bytes. If a
3773 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3774 kilobyte. This means that much larger values can be stored.
3776 8. Exim monitor: an attempt to get the action menu when not actually pointing
3777 at a message produces an empty menu entitled "No message selected". This
3778 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3779 no entries in it ("Shell widget menu has zero width and/or height"). So I
3780 have added a single, blank menu entry in this case.
3782 9. Added ${quote_local_part.
3784 10. MIME decoding is now applied to the contents of Subject: header lines when
3787 11. Now that a reference to $sender_host_address automatically causes a reverse
3788 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3789 host lookup before query-style lookups in lists that might use this
3790 variable. This has therefore been abolished, and the "net-" prefix is no
3791 longer necessary for query-style lookups.
3793 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3794 have been a typo for LFLAGS, so it has been changed.
3796 13. The install script calls Exim with "-C /dev/null" in order to find the
3797 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3798 to be output. Howeve, since Exim outputs its version number before the
3799 error, it didn't break the script. It just looked ugly. I fixed this by
3800 always allowing "-C /dev/null" if the caller is root.
3802 14. Ignore overlarge ACL variable number when reading spool file - insurance
3803 against a later release with more variables having written the file.
3805 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3806 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3807 this, as well as the form without the "IPv6" on the front.
3809 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3810 OS/Makefile-Darwin file.
3812 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3813 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3815 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3816 variable. However, some OS use ints for inodes. I've added cast to long int
3817 to get rid of the compiler warning.
3819 19. I had forgotten to lock out "/../" in configuration file names when
3820 ALT_CONFIG_PREFIX was set.
3822 20. Routers used for verification do not need to specify transports. However,
3823 if such a router generated a host list, and callout was configured, Exim
3824 crashed, because it could not find a port number from the (non-existent)
3825 transport. It now assumes port 25 in this circumstance.
3827 21. Added the -t option to exigrep.
3829 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3830 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3831 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3832 them is set, LOOKUP_LSEARCH is forced.
3834 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3836 24. Added sender and host information to the "rejected by local_scan()" log
3837 line; previously there was no indication of these.
3839 25. Added .include_if_exists.
3841 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3842 Linux. It turns out that not all file systems support this. Apparently some
3843 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3844 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3845 not supported on the file descriptor, is now ignored when Exim is trying to
3846 sync a directory. This applies only to Linux.
3848 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3850 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3851 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3854 29. Added support for maildirsize files from supplied patch (modified a bit).
3856 30. The use of :fail: followed by an empty string could lead Exim to respond to
3857 sender verification failures with (e.g.):
3859 550 Verification failed for <xxx>
3860 550 Sender verify failed
3862 where the first response line was missing the '-' that indicates it is not
3863 the final line of the response.
3865 31. The loop for finding the name of the user that called Exim had a hardwired
3866 limit of 10; it now uses the value of finduser_retries, which is used for
3867 all other user lookups.
3869 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3872 33. Exim was neglecting to zero errno before one call of strtol() when
3873 expanding a string and expecting an integer value. On some systems this
3874 resulted in spurious "integer overflow" errors. Also, it was casting the
3875 result into an int without checking.
3877 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3878 did not work. The code looks as if it has *never* worked, though it appears
3879 to have been documented since at least releast 1.62. I have made it work.
3881 35. The "timeout_DNS" error in retry rules, also documented since at least
3882 1.62, also never worked. As it isn't clear exactly what this means, and
3883 clearly it isn't a major issue, I have abolished the feature by treating it
3884 as "timeout", and writing a warning to the main and panic logs.
3886 36. The display of retry rules for -brt wasn't always showing the error code
3889 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3890 timeout_connect_A, timeout_connect_MX.
3892 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3893 to the empty sender.
3895 39. The daemon was not analysing the content of -oX till after it had closed
3896 stderr and disconnected from the controlling terminal. This meant that any
3897 syntax errors were only noted on the panic log, and the return code from
3898 the command was 0. By re-arranging the code a little, I've made the
3899 decoding happen first, so such errors now appear on stderr, and the return
3900 code is 1. However, the actual setting up of the sockets still happens in
3901 the disconnected process, so errors there are still only recorded on the
3904 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3905 connections (as happens on some IP stacks) was logged at start up time as
3906 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3907 it from "IPv6 and IPv4", which means that two separate sockets are being
3910 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3911 says whether AF_INET or AF_INET6 was passed as an argument.
3913 42. Exiwhat output was messed up when time zones were included in log
3916 43. Exiwhat now gives more information about the daemon's listening ports,
3917 and whether -tls-on-connect was used.
3919 44. The "port" option of the smtp transport is now expanded.
3921 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3922 silently ignored. Now an error message is written to the main and panic
3925 46. There's a new ACL modifier called "logwrite" which writes to a log file
3926 as soon as it is encountered.
3928 47. Added $local_user_uid and $local_user_gid at routing time.
3930 48. Exim crashed when trying to verify a sender address that was being
3933 49. Exim was recognizing only a space character after ".include". It now also
3934 recognizes a tab character.
3936 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3937 extracting the relevant information from the specification. The man page no
3938 longer contains scrambled data for the -d option, and I've added a section
3939 at the front about calling Exim under different names.
3941 51. Added "extra_headers" argument to the "mail" command in filter files.
3943 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3946 53. Installed eximstats 1.29.
3948 54. Added transport_filter_timeout as a generic transport option.
3950 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3951 Cc: header lines. This was required by RFC 822, but it not required by RFC
3954 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3955 incoming messages that did not have them. Now it does so only if the
3956 message originates locally, that is, if there is no associated remote host
3957 address. When Resent- header lines are present, this applies to the Resent-
3958 lines rather than the non-Resent- lines.
3960 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3961 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3963 58. Messages for configuration errors now include the name of the main
3964 configuration file - useful now that there may be more than one file in a
3965 list (.included file names were always shown).
3967 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3968 for those rare installations that do not start the daemon as root or run it
3969 setuid root. I've cut out the call to initgroups() if the daemon is not
3972 60. The Exim user and group can now be bound into the binary as text strings
3973 that are looked up at the start of Exim's processing.
3975 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3977 62. Added $mailstore_basename variable.
3979 63. Installed patch to sieve.c from Michael Haardt.
3981 64. When Exim failed to open the panic log after failing to open the main log,
3982 the original message it was trying to log was written to stderr and debug
3983 output, but if they were not available (the usual case in production), it
3984 was lost. Now it is written to syslog before the two lines that record the
3985 failures to open the logs.
3987 65. Users' Exim filters run in subprocesses under the user's uid. It is
3988 possible for a "deliver" command or an alias in a "personal" command to
3989 provoke an address rewrite. If logging of address rewriting is configured,
3990 this fails because the process is not running as root or exim. There may be
3991 a better way of dealing with this, but for the moment (because 4.30 needs
3992 to be released), I have disabled address rewrite logging when running a
3993 filter in a non-root, non-exim process.
3999 1. The buildconfig auxiliary program wasn't quoting the value set for
4000 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4001 not defined. This bug was masked in 4.22 by the effect that was fixed in
4004 2. Some messages that were rejected after a message id was allocated were
4005 shown as "incomplete" by exigrep. It no longer does this for messages that
4006 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4008 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4009 have allow_domain_literals set, the ID did not get logged in the <= line.
4010 Domain literals are now always recognized in Message-ID: header lines.
4012 4. The first argument for a ${extract expansion item is the key name or field
4013 number. Leading and trailing spaces in this item were not being ignored,
4014 causing some misleading effects.
4016 5. When deliver_drop_privilege was set, single queue runner processes started
4017 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4018 same command in the process it spins off) were not dropping privilege.
4020 6. When the daemon running as "exim" started a queue runner, it always
4021 re-executed Exim in the spun-off process. This is a waste of effort when
4022 deliver_drop_privilege is set. The new process now just calls the
4023 queue-runner function directly.
4029 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4032 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4033 ignored. Though the use of -oP was forcing the writing of a pid file, it
4034 was always written to the default place.
4036 3. If the message "no IP address found for host xxxx" is generated during
4037 incoming verification, it is now followed by identification of the incoming
4038 connection (so you can more easily find what provoked it).
4040 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4042 5. Added some features to "harden" Exim a bit more against certain attacks:
4044 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4045 be put in Local/Makefile. This is like the never_users runtime option,
4046 but it cannot be overridden. The default setting is "root".
4048 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4049 prefix string with which any file named in a -C command line option
4052 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4053 is retained for -C and -D only if the caller of Exim is root. Without
4054 it, the exim user may also use -C and -D and retain privilege.
4056 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4057 command line option is disabled.
4059 6. Macro names set by the -D option must start with an upper case letter, just
4060 like macro names defined in the configuration file.
4062 7. Added "dereference=" facility to LDAP.
4064 8. Two instances of the typo "uknown" in the source files are fixed.
4066 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4067 the Configure-Makefile script screwed up while processing it.
4069 10. Incorporated PCRE 4.4.
4071 11. The SMTP synchronization check was not operating right at the start of an
4072 SMTP session. For example, it could not catch a HELO sent before the client
4073 waited for the greeting. There is now a check for outstanding input at the
4074 point when the greeting is written. Because of the duplex, asynchronous
4075 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4076 way, but not yet received, when the check is performed.
4078 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4079 on TCP/IP sockets, because this apparently causes some broken clients to
4082 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4083 unchanged) from the Cygwin maintainer.
4085 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4086 was supported instead of testing for mbx. Effectively a typo.
4088 15. The spa authenticator server code was not checking that the input it
4089 received was valid base64.
4091 16. The debug output line for the "set" modifier in ACLs was not showing the
4092 name of the variable that was being set.
4094 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4096 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4097 lookup on its own. Something else was needed to trigger the lookup. For
4098 example, a match in host_lookup or the need for a host name in a host list.
4099 Now, if $sender_host_name is referenced and the host name has not yet been
4100 looked up, a lookup is performed. If the lookup fails, the variable remains
4101 empty, and $host_lookup_failed is set to "1".
4103 19. Added "eqi" as a case-independent comparison operator.
4105 20. The saslauthd authentication condition could segfault if neither service
4106 nor realm was specified.
4108 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4109 error message that was logged was misleading, and incoming SMTP
4110 connections were dropped. The message is now more accurate, and temporary
4111 errors are given to SMTP connections.
4113 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4114 (except RSET) with a 421 error, until QUIT is received. However, it was
4115 failing to send a response to QUIT.
4117 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4118 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4119 ACL is now run after the helo_try_verify_hosts code.
4121 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4122 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4123 case-independent, so other case variants are also recognized). Apparently
4124 some systems use these upper case variants.
4126 25. If more than two messages were waiting for the same host, and a transport
4127 filter was specified for the transport, Exim sent two messages over the
4128 same TCP/IP connection, and then failed with "socket operation on non-
4129 socket" when it tried to send the third.
4131 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4133 27. The extern definition of crypt16() in expand.c was not being excluded when
4134 the OS had its own crypt16() function.
4136 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4137 as a preferred synonym for return_size_limit, both as an option and as an
4140 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4142 30. Changed the default configuration ACL to relax the local part checking rule
4143 for addresses that are not in any local domains. For these addresses,
4144 slashes and pipe symbols are allowed within local parts, but the sequence
4145 /../ is explicitly forbidden.
4147 31. SPA server authentication was not clearing the challenge buffer before
4150 32. log_message in a "warn" ACL statement was writing to the reject log as
4151 well as to the main log, which contradicts the documentation and doesn't
4152 seem right (because no rejection is happening). So I have stopped it.
4154 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4155 However, I am unable to do any testing of this.
4157 34. Fixed an infelicity in the appendfile transport. When checking directories
4158 for a mailbox, to see if any needed to be created, it was accidentally
4159 using path names with one or more superfluous leading slashes; tracing
4160 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4162 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4163 contents are added to the log line that is written for every discarded
4164 recipient. (Previously a log_message setting was ignored.)
4166 36. The ${quote: operator now quotes the string if it is empty.
4168 37. The install script runs exim in order to find its version number. If for
4169 some reason other than non-existence or emptiness, which it checks, it
4170 could not run './exim', it was installing it with an empty version number,
4171 i.e. as "exim-". This error state is now caught, and the installation is
4174 38. An argument was missing from the function that creates an error message
4175 when Exim fails to connect to the socket for saslauthd authentication.
4176 This could cause Exim to crash, or give a corrupted message.
4178 39. Added isip, isip4, and isip6 to ${if conditions.
4180 40. The ACL variables $acl_xx are now saved with the message, and can be
4181 accessed later in routers, transports, and filters.
4183 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4184 strings in the file are not string-expanded.
4186 42. If a MAIL command specified a SIZE value that was too large to fit into an
4187 int variable, the check against message_size_limit failed. Such values are
4188 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4189 day this will have to be increased, but I don't think I want to be around
4190 when emails are that large.
4197 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4198 iconv() is not standard in FreeBSD.
4200 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4201 IPv6 enabled. The observed symptom was a segmentation fault on return from
4202 the function os_common_find_running_interfaces() in src/os.c.
4204 3. In the check_special_case() function in daemon.c I had used "errno" as an
4205 argument name, which causes warnings on some systems. This was basically a
4206 typo, since it was named "eno" in the comments!
4208 4. The code that waits for the clock to tick (at a resolution of some fraction
4209 of a second) so as to ensure message-id uniqueness was always waiting for
4210 at least one whole tick, when it could have waited for less. [This is
4211 almost certainly not relevant at current processor speeds, where it is
4212 unlikely to ever wait at all. But we try to future-proof.]
4214 5. The function that sleeps for a time interval that includes fractions of a
4215 second contained a race. It did not block SIGALRM between setting the
4216 timer, and suspending (a couple of lines later). If the interval was short
4217 and the sigsuspend() was delayed until after it had expired, the suspension
4218 never ended. On busy systems this could lead to processes getting stuck for
4221 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4222 process, before it forks any delivery processes. The open lookup caching
4223 mechanism meant that the open file or database connection was passed into
4224 the delivery process. The problem was that delivery processes always tidy
4225 up cached lookup data. This could cause a problem for the next delivery
4226 process started by the queue runner, because the external queue runner
4227 process does not know about the closure. So the next delivery process
4228 still has data in the lookup cache. In the case of a file lookup, there was
4229 no problem because closing a file descriptor in a subprocess doesn't affect
4230 the parent. However, if the lookup was caching a connection to a database,
4231 the connection was closed, and the second delivery process was likely to
4232 see errors such as "PGSQL: query failed: server closed the connection
4233 unexpectedly". The problem has been fixed by closing all cached lookups
4234 in a queue runner before running a delivery process.
4236 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4237 seem to have the "const" qualifier which it has on other OS. I've
4240 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4241 *of the same type* (client or server) with the same public name that an
4242 error should be diagnosed.
4244 9. When Exim looked up a host name for an IP address, but failed to find the
4245 original IP address when looking up the host name (a safety check), it
4246 output the message "<ip address> does not match any IP for NULL", which was
4247 confusing, to say the least. The bug was that the host name should have
4248 appeared instead of "NULL".
4250 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4251 user that is built into the binary, and the -C or -D options is used, root
4252 privilege is dropped before the configuration file is read. In addition,
4253 logging is switched to stderr instead of the normal log files. If the
4254 configuration then re-defines the Exim user, the unprivileged environment
4255 is probably not what is expected, so Exim logs a panic warning message (but
4258 However, if deliver_drop_privilege is set, the unprivileged state may well
4259 be exactly what is intended, so the warning has been cut out in that case,
4260 and Exim is allowed to try to write to its normal log files.
4266 1. smtp_return_error_details was not giving details for temporary sender
4267 or receiver verification errors.
4269 2. Diagnose a configuration error if two authenticators have the same public
4272 3. Exim used not to create the message log file for a message until the first
4273 delivery attempt. This could be confusing when incoming messages were held
4274 for policy or load reasons. The message log file is now created at the time
4275 the message is received, and an initial "Received" line is written to it.
4277 4. The automatically generated man page for command line options had a minor
4278 bug that caused no ill effects; however, a more serious problem was that
4279 the procedure for building the man page automatically didn't always
4280 operate. Consequently, release 4.20 contains an out-of-date version. This
4281 shouldn't happen again.
4283 5. When building Exim with embedded Perl support, the script that builds the
4284 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4285 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4287 6. The freeze_tell option was not being used for messages that were frozen on
4288 arrival, either by an ACL or by local_scan().
4290 7. Added the smtp_incomplete_transaction log selector.
4292 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4293 was accepting AUTH without a new EHLO.
4295 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4296 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4299 10. Logging of TCP/IP connections (when configured) now happens in the main
4300 daemon process instead of the child process, so that the TCP/IP connection
4301 count is more accurate (but it can never be perfect).
4303 11. The use of "drop" in a nested ACL was not being handled correctly in the
4304 outer ACL. Now, if condition failure induced by the nested "drop" causes
4305 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4306 or "require"), the connection is dropped.
4308 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4309 that yield "discard" can now be used with an "accept" or a "discard" verb,
4310 but an error is generated for any others (because I can't see a useful way
4311 to define what should happen).
4313 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4314 are now processed in the same way as lines in the Exim configuration file.
4315 In particular, continuation lines are supported.
4317 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4319 15. Added -ti meaning -t -i.
4321 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4322 domains, and warn by logging if others are found.
4324 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4325 returned from GIFCONF when Exim is trying to find the list of interfaces on
4326 a host. The code in os.c has been modified to copy each ifreq to an aligned
4327 structure in all cases.
4329 Also, in some cases, the returned ifreq's were being copied to a 'struct
4330 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4331 means the last couple of bytes of an IPv6 address could be chopped if the
4332 ifreq contained only a normal sockaddr (14 bytes storage).
4334 18. Named domain lists were not supported in the hosts_treat_as_local option.
4335 An entry such as +xxxx was not recognized, and was treated as a literal
4338 19. Ensure that header lines added by a DATA ACL are included in the reject log
4339 if the ACL subsequently rejects the message.
4341 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4342 MD5 (which is deprecated).
4344 21. When testing a filter file using -bf, Exim was writing a message when it
4345 took the sender from a "From " line in the message, but it was not doing so
4346 when it took $return_path from a Return-Path: header line. It now does.
4348 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4349 with a valid header line field name (a series of printing characters
4350 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4352 23. Changed "disc" in the source to "disk" to conform to the documentation and
4353 the book and for uniformity.
4355 24. Ignore Sendmail's -Ooption=value command line item.
4357 25. When execve() failed while trying to run a command in a pipe transport,
4358 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4359 could be confused with a return value of 69 from the command itself. This
4360 has been changed to 127, the value the shell returns if it is asked to run
4361 a non-existent command. The wording for the related log line suggests a
4362 non-existent command as the problem.
4364 26. If received_header_text expands to an empty string, do not add a Received:
4365 header line to the message. (Well, it adds a token one on the spool, but
4366 marks it "old" so that it doesn't get used or transmitted.)
4368 27. Installed eximstats 1.28 (addition of -nt option).
4370 28. There was no check for failure on the call to getsockname() in the daemon
4371 code. This can fail if there is a shortage of resources on the system, with
4372 ENOMEM, for example. A temporary error is now given on failure.
4374 29. Contrary to the C standard, it seems that in some environments, the
4375 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4376 program. Exim now does this explicitly; it affects the formatting of
4377 timestamps using strftime().
4379 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4380 complaints. I've now initialized all the variables, to avoid this.
4382 32. Header lines added by a system filter were not being "seen" during
4383 transport-time rewrites.
4385 33. The info_callback() function passed to OpenSSL is set up with type void
4386 (*)(SSL *, int, int), as described somewhere. However, when calling the
4387 function (actually a macro) that sets it up, the type void(*)() is
4388 expected. I've put in a cast to prevent warnings from picky compilers.
4390 34. If a DNS black list lookup found a CNAME record, but there were no A
4391 records associated with the domain it pointed at, Exim crashed.
4393 35. If a DNS black list lookup returned more than one A record, Exim ignored
4394 all but the first. It now scans all returned addresses if a particular IP
4395 value is being sought. In this situation, the contents of the
4396 $dnslist_value variable are a list of all the addresses, separated by a
4399 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4400 to accept a host name and all its aliases if the forward lookup for any of
4401 them yielded the IP address of the incoming connection. Now it accepts only
4402 those names whose forward lookup yields the correct IP address. Any other
4403 names are discarded. This closes a loophole whereby a rogue DNS
4404 administrator could create reverse DNS records to break through a
4405 wildcarded host restriction in an ACL.
4407 37. If a user filter or a system filter that ran in a subprocess used any of
4408 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4409 the wrong values were passed to the pipe command ($thisaddress had the
4410 value of $0, $0 had the value of $1, etc). This bug was introduced by
4411 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4413 38. Improved the line breaking for long SMTP error messages from ACLs.
4414 Previously, if there was no break point between 40 and 75 characters, Exim
4415 left the rest of the message alone. Two changes have been made: (a) I've
4416 reduced the minimum length to 35 characters; (b) if it can't find a break
4417 point between 35 and 75 characters, it looks ahead and uses the first one
4418 that it finds. This may give the occasional overlong line, but at least the
4419 remaining text gets split now.
4421 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4422 file descriptors might be low, and that setting 1000 would always raise it.
4423 It turns out that in some environments, the limit is already over 1000 and
4424 that lowering it causes trouble. So now Exim takes care not to decrease it.
4426 40. When delivering a message, the value of $return_path is set to $sender_
4427 address at the start of routing (routers may change the value). By an
4428 oversight, this default was not being set up when an address was tested by
4429 -bt or -bv, which affected the outcome if any router or filter referred to
4432 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4433 log only when "message" or "log_message" are set. However, if one of the
4434 conditions was an address verification, or a call to a nested ACL, the
4435 messages generated by the underlying test were being passed through. This
4436 no longer happens. The underlying message is available in $acl_verify_
4437 message for both "message" and "log_message" expansions, so it can be
4438 passed through if needed.
4440 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4441 new expansion $bh_ to give the encoded byte string without charset
4442 translation. Translation happens only if iconv() is available; HAVE_ICONV
4443 indicates this at build time. HEADERS_CHARSET gives the charset to
4444 translate to; headers_charset can change it in the configuration, and
4445 "headers charset" can change it in an individual filter file.
4447 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4448 that creates RFC 2047 encoded "words" labels them as that charset instead
4449 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4450 expansion operator; (ii) when Exim creates a From: line for a local
4451 message; (iii) when a header line is rewritten to include a "phrase" part.
4453 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4454 buggy, causing it to skip the first lines of messages whose message ID
4455 ended in 'D'. This would not have bitten before Exim release 4.14, because
4456 message IDs were unlikely to end in 'D' before then. The effect was to have
4457 incorrect size information for certain domains.
4459 45. #include "config.h" was missing at the start of the crypt16.c module. This
4460 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4463 46. If there was a timeout during a "random" callout check, Exim treated it as
4464 a failure of the random address, and carried on sending RSET and the real
4465 address. If the delay was just some slowness somewhere, the response to the
4466 original RCPT would be taken as a response to RSET and so on, causing
4467 mayhem of various kinds.
4469 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4470 when I implemented it. It didn't allow for the fact that some option values
4471 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4472 the right test for positive values.
4474 48. Domain names in DNS records are case-independent. Exim always looks them up
4475 in lower case. Some resolvers return domain names in exactly the case they
4476 appear in the zone file, that is, they may contain uppercase letters. Not
4477 all resolvers do this - some return always lower case. Exim was treating a
4478 change of case by a resolver as a change of domain, similar to a widening
4479 of a domain abbreviation. This triggered its re-routing code and so it was
4480 trying to route what was effectively the same domain again. This normally
4481 caused routing to fail (because the router wouldn't handle the domain
4482 twice). Now Exim checks for this case specially, and just changes the
4483 casing of the domain that it ultimately uses when it transmits the message
4486 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4489 50. If a filter generated a file delivery with a non-absolute name (possible if
4490 no home directory exists for the router), the forbid_file option was not
4493 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4494 the existing equality matching.
4496 52. Exim was using ints instead of ino_t variables in some places where it was
4497 dealing with inode numbers.
4499 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4500 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4501 called TMPDIR, and if it finds it is different, it changes its value.
4503 54. The smtp_printf() function is now made available to local_scan() so
4504 additional output lines can be written before returning. There is also an
4505 smtp_fflush() function to enable the detection of a dropped connection.
4506 The variables smtp_input and smtp_batched_input are exported to
4509 55. Changed the default runtime configuration: the message "Unknown user"
4510 has been removed from the ACL, and instead placed on the localuser router,
4511 using the cannot_route_message feature. This means that any verification
4512 failures that generate their own messages won't get overridden. Similarly,
4513 the "Unrouteable address" message that was in the ACL for unverifiable
4514 relay addresses has also been removed.
4516 56. Added hosts_avoid_esmtp to the smtp transport.
4518 57. The exicyclog script was not checking for the esoteric option
4519 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4520 will work only if exicyclog is run under the appropriate euid.
4522 58. Following a discussion on the list, the rules by which Exim recognises line
4523 endings on incoming messages have been changed. The -dropcr and drop_cr
4524 options are now no-ops, retained only for backwards compatibility. The
4525 following line terminators are recognized: LF CRLF CR. However, special
4526 processing applies to CR:
4528 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4529 nor a local message in the state where . is a terminator.
4531 (ii) If a bare CR is encountered in a header line, an extra space is added
4532 after the line terminator so as not to end the header. The reasoning
4533 behind this is that bare CRs in header lines are most likely either
4534 to be mistakes, or people trying to play silly games.
4536 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4537 was being incorrectly given as 18 bytes larger than it should have been.
4538 This is a VOB (very old bug).
4540 60. This may never have affected anything current, but just in case it has:
4541 When the local host is found other than at the start of a list of hosts,
4542 the local host, those with the same MX, and any that follow, are discarded.
4543 When the list in question was part of a longer list of hosts, the following
4544 hosts (not currently being processed) were also being discarded. This no
4545 longer happens. I'm not sure if this situation could ever has previously
4548 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4551 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4552 header. This is something that is recommended in a new Internet Draft, and
4553 is something that is documented as being done by Sendmail. There are two
4554 possible values. For messages generated by the autoreply transport, Exim
4557 Auto-Submitted: auto-replied
4559 whereas for all other generated messages (e.g. bounces) it adds
4561 Auto-Submitted: auto-generated
4563 63. The "personal" condition in filters now includes a test for the
4564 Auto-Submitted: header. If it contains the string "auto-" the message it
4565 not considered personal.
4567 64. Added rcpt_include_affixes as a generic transport option.
4569 65. Added queue_only_override (default true).
4571 66. Added the syslog_duplication option.
4573 67. If what should have been the first header line of a message consisted of
4574 a space followed by a colon, Exim was mis-interpreting it as a header line.
4575 It isn't of course - it is syntactically invalid and should therefore be
4576 treated as the start of the message body. The misbehaviour could have
4577 caused a number of strange effects, including loss of data in subsequent
4578 header lines, and spool format errors.
4580 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4581 client host had authenticated. This control can now be exercised by an ACL
4582 for more flexibility.
4584 69. By default, callouts do not happen when testing with -bh. There is now a
4585 variant, -bhc, which does actually run the callout code, including
4586 consulting and updating the callout cache.
4588 70. Added support for saslauthd authentication, courtesy of Alexander
4591 71. If statvfs() failed on the spool or log directories while checking their
4592 size for availability, Exim confusingly gave the error "space shortage".
4593 Furthermore, in debugging mode it crashed with a floating point exception.
4594 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4595 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4596 serious problem, Exim now writes to the main and panic logs when this
4597 happens, with details of the failure. It then refuses to accept the
4598 incoming message, giving the message "spool directory problem" or "log
4599 directory problem" with a 421 code for SMTP messages.
4601 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4602 0, 1, and 2 exist, because some OS complain for execs without them (see
4603 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4604 descriptors. However, the code omitted to check that the open succeeded,
4605 causing mysterious errors if for some reason the permissions on /dev/null
4606 got screwed. Now Exim writes a message to the main and panic logs, and
4607 bombs out if it can't open /dev/null.
4609 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4610 interact so that it is all more flexible. It is supposed to remain
4611 backwards compatible. Also added extra_local_interfaces.
4613 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4614 to bomb out with an assertion failure - to the client this appears as a
4615 connection drop. This problem occurs in the part of the code that was taken
4616 from the Samba project. Fortunately, the assertion is in a very simple
4617 function, so I have fixed this by reproducing the function inline in the
4618 one place where it is called, and arranging for authentication to fail
4619 instead of killing the process with assert().
4621 75. The SPA client code was not working when the server requested OEM rather
4622 than Unicode encoding.
4624 76. Added code to make require_files with a specific uid setting more usable in
4625 the case where statting the file as root fails - usually a non-root-mounted
4626 NFS file system. When this happens and the failure is EACCES, Exim now
4627 forks a subprocess and does the per-uid checking as the relevant uid.
4629 77. Added process_log_path.
4631 78. If log_file_path was not explicitly set, a setting of check_log_space or
4632 check_log_inodes was ignored.
4634 79. If a space check for the spool or log partitions fails, the incident is now
4635 logged. Of course, in the latter case the data may get lost...
4637 80. Added the %p formatting code to string_format() so that it can be used to
4638 print addresses in debug_print(). Adjusted all the address printing in the
4639 debugging in store.c to use %p rather than %d.
4641 81. There was a concern that a line of code in smtp_in.c could overflow a
4642 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4643 initially expressed, the concern was not well-founded, because trailing
4644 spaces are removed early. However, if the trailing spaces were followed by
4645 a NULL, they did not get removed, so the overflow was possible. Two fixes
4648 (a) I re-wrote the offending code in a cleaner fashion.
4649 (b) If an incoming SMTP command contains a NULL character, it is rejected
4652 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4653 runs initgroups(), so that if the Exim user is in any additional groups,
4654 they will be used during message reception.
4660 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob