1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.492 2007/03/13 11:26:49 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
144 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
145 Exim which did not show up earlier: it was assuming that a call to
146 SSL_CTX_set_info_callback() might give an error value. In fact, there is
147 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
148 was a macro that became an assignment, so it seemed to work. This has
149 changed to a proper function call with a void return, hence the compile
150 error. Exim's code has been fixed.
152 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
155 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
157 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
159 PH/37 If a message is not accepted after it has had an id assigned (e.g.
160 because it turns out to be too big or there is a timeout) there is no
161 "Completed" line in the log. When such a message was selected by exigrep,
162 it was listed as "not completed". Such messages can be identified by the
163 lack of a "<=" line in the log, and they are now listed as "not
170 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
171 fixed by 4.65/MH/01 (is this a record?) are fixed:
173 (i) An empty string was always treated as zero by the numeric comparison
174 operators. This behaviour has been restored.
176 (ii) It is documented that the numeric comparison operators always treat
177 their arguments as decimal numbers. This was broken in that numbers
178 starting with 0 were being interpreted as octal.
180 While fixing these problems I realized that there was another issue that
181 hadn't been noticed. Values of message_size_limit (both the global option
182 and the transport option) were treated as octal if they started with 0.
183 The documentation was vague. These values are now always treated as
184 decimal, and I will make that clear in the documentation.
190 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
191 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
194 MH/01 Don't check that the operands of numeric comparison operators are
195 integers when their expansion is in "skipping" mode (fixes bug
196 introduced by 4.64-PH/07).
198 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
199 child addresses, Exim now panics and dies. Previously, because the count
200 is held in a short int, deliveries were likely to be lost. As such a
201 large number of recipients for a single message is ridiculous
202 (performance will be very, very poor), I have chosen to impose a limit
203 rather than extend the field.
209 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
210 leftover -K file (the existence of which was triggered by #402).
211 While we were at it, introduced process PID as part of the -K
212 filename. This should rule out race conditions when creating
215 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
216 processing considerably. Previous code took too long for large mails,
217 triggering a timeout which in turn triggers #401.
219 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
220 in the DK code in transports.c. sendfile() is not really portable,
221 hence the _LINUX specificness.
223 TF/01 In the add_headers option to the mail command in an Exim filter,
224 there was a bug that Exim would claim a syntax error in any
225 header after the first one which had an odd number of characters
228 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
229 callout verification, Exim cached a "reject" for the entire domain. This
230 is correct for most verifications, but it is not correct for a recipient
231 verification with use_sender or use_postmaster set, because in that case
232 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
233 case of MAIL FROM:<> rejection from other early rejections (e.g.
234 rejection of HELO). When verifying a recipient using a non-null MAIL
235 address, the cache is ignored if it shows MAIL FROM:<> rejection.
236 Whatever the result of the callout, the value of the domain cache is
237 left unchanged (for any other kind of callout, getting as far as trying
238 RCPT means that the domain itself is ok).
240 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
243 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
244 manifest itself as EPIPE rather than ECONNECT. When tidying away a
245 session, the daemon ignores ECONNECT errors and logs others; it now
246 ignores EPIPE as well.
248 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
249 (quoted-printable decoding).
251 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
252 later the small subsequent patch to fix an introduced bug.
254 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
256 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
258 PH/08 An error is now given if message_size_limit is specified negative.
260 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
261 to be given (somewhat) arbitrary names.
263 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
266 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
267 miscellaneous code fixes
269 PH/10 Added the log_reject_target ACL modifier to specify where to log
272 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
273 hostname. This is wrong, because it relates to the incoming message (and
274 probably the interface on which it is arriving) and not to the outgoing
275 callout (which could be using a different interface). This has been
276 changed to use the value of the helo_data option from the smtp transport
277 instead - this is what is used when a message is actually being sent. If
278 there is no remote transport (possible with a router that sets up host
279 addresses), $smtp_active_hostname is used.
281 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
282 tweaks were necessary in order to get it to work (see also 21 below):
283 (a) The code assumed that strncpy() returns a negative number on buffer
284 overflow, which isn't the case. Replaced with Exim's string_format()
286 (b) There were several signed/unsigned issues. I just did the minimum
287 hacking in of casts. There is scope for a larger refactoring.
288 (c) The code used strcasecmp() which is not a standard C function.
289 Replaced with Exim's strcmpic() function.
290 (d) The code set only $1; it now sets $auth1 as well.
291 (e) A simple test gave the error "authentication client didn't specify
292 service in request". It would seem that Dovecot has changed its
293 interface. Fortunately there's a specification; I followed it and
294 changed what the client sends and it appears to be working now.
296 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
299 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
300 address A is aliased to B and C, where B exists and C does not. Without
301 -v the output is "A verified" because verification stops after a
302 successful redirection if more than one address is generated. However,
303 with -v the child addresses are also verified. Exim was outputting "A
304 failed to verify" and then showing the successful verification for C,
305 with its parentage. It now outputs "B failed to verify", showing B's
306 parentage before showing the successful verification of C.
308 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
309 look up a TXT record in a specific list after matching in a combined
312 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
313 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
314 they consult the DNS. I had assumed they would set it the way they
315 wanted; and indeed my experiments on Linux seem to show that in some
316 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
317 To be on the safe side, however, I have now made the interface to
318 host_find_byname() similar to host_find_bydns(), with an argument
319 containing the DNS resolver options. The host_find_byname() function now
320 sets these options at its start, just as host_find_bydns() does. The smtp
321 transport options dns_qualify_single and dns_search_parents are passed to
322 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
323 of host_find_byname() use the default settings of RES_DEFNAMES
324 (qualify_single) but not RES_DNSRCH (search_parents).
326 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
327 spool_read_header() do less string testing, by means of a preliminary
328 switch on the second character of optional "-foo" lines. (This is
329 overdue, caused by the large number of possibilities that now exist.
330 Originally there were few.) While I was there, I also converted the
331 str(n)cmp tests so they don't re-test the leading "-" and the first
332 character, in the hope this might squeeze out yet more improvement.
334 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
335 flag allowing group syntax was set by the header_syntax check but not
336 turned off, possible causing trouble later; (2) The flag was not being
337 set at all for the header_verify test, causing "group"-style headers to
338 be rejected. I have now set it in this case, and also caused header_
339 verify to ignore an empty address taken from a group. While doing this, I
340 came across some other cases where the code for allowing group syntax
341 while scanning a header line wasn't quite right (mostly, not resetting
342 the flag correctly in the right place). These bugs could have caused
343 trouble for malformed header lines. I hope it is now all correct.
345 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
346 with the "reply" argument non-NULL. The code, however (which originally
347 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
348 but it didn't always do it. This confused somebody who was copying the
349 code for some other use. I have removed all the tests.
351 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
352 feature that was used to support insecure browsers during the U.S. crypto
353 embargo. It requires special client support, and Exim is probably the
354 only MTA that supported it -- and would never use it because real RSA is
355 always available. This code has been removed, because it had the bad
356 effect of slowing Exim down by computing (never used) parameters for the
357 RSA_EXPORT functionality.
359 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
360 authenticator to fail if there's a tab character in the incoming data
361 (there should never be unless someone is messing about, as it's supposed
362 to be base64-encoded). Also added, on Timo's advice, the "secured" option
363 if the connection is using TLS or if the remote IP is the same as the
364 local IP, and the "valid-client-cert option" if a client certificate has
367 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
368 authenticators. This can be used for authorization after authentication
369 succeeds. (In the case of plaintext, it servers for both authentication
372 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
373 if any retry times were supplied.
375 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
376 connection, where there is no client IP address to check. In this
377 situation, the verify now always succeeds.
379 PH/25 Applied John Jetmore's -Mset patch.
381 PH/26 Added -bem to be like -Mset, but loading a message from a file.
383 PH/27 In a string expansion for a processed (not raw) header when multiple
384 headers of the same name were present, leading whitespace was being
385 removed from all of them, but trailing whitespace was being removed only
386 from the last one. Now trailing whitespace is removed from each header
387 before concatenation. Completely empty headers in a concatenation (as
390 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
391 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
393 PH/29 [Removed. This was a change that I later backed out, and forgot to
394 correct the ChangeLog entry (that I had efficiently created) before
395 committing the later change.]
397 PH/30 Exim was sometimes attempting to deliver messages that had suffered
398 address errors (4xx response to RCPT) over the same connection as other
399 messages routed to the same hosts. Such deliveries are always "forced",
400 so retry times are not inspected. This resulted in far too many retries
401 for the affected addresses. The effect occurred only when there were more
402 hosts than the hosts_max_try setting in the smtp transport when it had
403 the 4xx errors. Those hosts that it had tried were not added to the list
404 of hosts for which the message was waiting, so if all were tried, there
405 was no problem. Two fixes have been applied:
407 (i) If there are any address or message errors in an SMTP delivery, none
408 of the hosts (tried or untried) are now added to the list of hosts
409 for which the message is waiting, so the message should not be a
410 candidate for sending over the same connection that was used for a
411 successful delivery of some other message. This seems entirely
412 reasonable: after all the message is NOT "waiting for some host".
413 This is so "obvious" that I'm not sure why it wasn't done
414 previously. Hope I haven't missed anything, but it can't do any
415 harm, as the worst effect is to miss an optimization.
417 (ii) If, despite (i), such a delivery is accidentally attempted, the
418 routing retry time is respected, so at least it doesn't keep
419 hammering the server.
421 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
422 in ${readsocket because some servers need this prod.
424 PH/32 Added some extra debug output when updating a wait-xxx database.
426 PH/33 The hint "could be header name not terminated by colon", which has been
427 given for certain expansion errors for a long time, was not being given
428 for the ${if def:h_colon_omitted{... case.
430 PH/34 The spec says: "With one important exception, whenever a domain list is
431 being scanned, $domain contains the subject domain." There was at least
432 one case where this was not true.
434 PH/35 The error "getsockname() failed: connection reset by peer" was being
435 written to the panic log as well as the main log, but it isn't really
436 panic-worthy as it just means the connection died rather early on. I have
437 removed the panic log writing for the ECONNRESET error when getsockname()
440 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
441 runs only) independently of the message's sender address. This meant
442 that, if the 4xx error was in fact related to the sender, a different
443 message to the same recipient with a different sender could confuse
444 things. In particualar, this can happen when sending to a greylisting
445 server, but other circumstances could also provoke similar problems.
446 I have changed the default so that the retry time for these errors is now
447 based a combination of the sender and recipient addresses. This change
448 can be overridden by setting address_retry_include_sender=false in the
451 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
452 remote server are returned as part of bounce messages. This was not
453 happening for LMTP over a pipe (the lmtp transport), but now it is the
454 same for both kinds of LMTP.
456 PH/38 Despite being documented as not happening, Exim was rewriting addresses
457 in header lines that were in fact CNAMEs. This is no longer the case.
459 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
460 and queue runs started by the daemon processed all messages. This has
461 been fixed so that -R and -S can now usefully be given with -q<time>.
463 PH/40 Import PCRE release 6.7 (fixes some bugs).
465 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
467 PH/42 Give an error if -q is specified more than once.
469 PH/43 Renamed the variables $interface_address and $interface_port as
470 $received_ip_address and $received_port, to make it clear that these
471 values apply to message reception, and not to the outgoing interface when
472 a message is delivered. (The old names remain recognized, of course.)
474 PH/44 There was no timeout on the connect() call when using a Unix domain
475 socket in the ${readsocket expansion. There now is.
477 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
478 be meaningful with "accept".
480 SC/01 Eximstats V1.43
481 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
483 SC/02 Eximstats V1.44
484 Use a glob alias rather than an array ref in the generated
485 parser. This improves both readability and performance.
487 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
488 Collect SpamAssassin and rejection statistics.
489 Don't display local sender or destination tables unless
490 there is data to show.
491 Added average volumes into the top table text output.
493 SC/04 Eximstats V1.46
494 Collect data on the number of addresses (recipients)
495 as well as the number of messages.
497 SC/05 Eximstats V1.47
498 Added 'Message too big' to the list of mail rejection
499 reasons (thanks to Marco Gaiarin).
501 SC/06 Eximstats V1.48
502 Mainlog lines which have GMT offsets and are too short to
503 have a flag are now skipped.
505 SC/07 Eximstats V1.49 (Alain Williams)
506 Added the -emptyok flag.
508 SC/08 Eximstats V1.50
509 Fixes for obtaining the IP address from reject messages.
511 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
512 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
513 whitesspace changes from 4.64-PH/27
515 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
518 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
519 are found, allow negative numbers in numeric criteria)
521 JJ/06 exipick.20061117.2, added new $message_body_missing variable
523 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
524 to match changes made in 4.64-PH/43
526 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
528 PH/47 Put in an explicit test for a DNS lookup of an address record where the
529 "domain" is actually an IP address, and force a failure. This locks out
530 those revolvers/nameservers that support "A-for-A" lookups, in
531 contravention of the specifications.
533 PH/48 When a host name was looked up from an IP address, and the subsequent
534 forward lookup of the name timed out, the host name was left in
535 $sender_host_name, contrary to the specification.
537 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
538 restricted to single-key lookups, Exim was not diagnosing an error if
539 * or *@ was used with a query-style lookup.
541 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
543 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
544 long ago, but noone interested enough thought of it. Let's just say that
545 the "1.1" means that there are some new functions that weren't there at
546 some point in the past.
548 PH/51 Error processing for expansion failure of helo_data from an smtp
549 transport during callout processing was broken.
551 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
552 tested/used via the -bh/-bhc/-bs options.
554 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
555 bug, fixed in subsequent PCRE releases).
557 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
558 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
560 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
566 SC/01 Use a glob alias rather than an array ref in eximstats generated
567 parser. This improves both readability and performance.
569 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
570 Don't display local sender or destination tables in eximstats unless
571 there is data to show.
572 Added average volumes into the eximstats top table text output.
574 SC/03 Collect data on the number of addresses (recipients) as well
575 as the number of messages in eximstats.
577 TF/01 Correct an error in the documentation for the redirect router. Exim
578 does (usually) call initgroups() when daemonizing.
580 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
581 with consistent privilege compared to when running as a daemon.
583 TF/03 Note in the spec that $authenticated_id is not set for local
584 submissions from trusted users.
586 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
587 Thanks to Dean Brooks <dean@iglou.com> for the patch.
589 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
590 by adding some example configuration directives to the default
591 configuration file. A little bit of work is required to uncomment the
592 directives and define how usernames and passwords are checked, but
593 there is now a framework to start from.
595 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
596 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
597 without this. I don't know how relevant this is to other LDAP libraries.
599 PH/02 Add the verb name to the "unknown ACL verb" error.
601 PH/03 Magnus Holmgren's patch for filter_prepend_home.
603 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
605 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
606 directory not expanded when it should be if an expanded home directory
607 was set for the address (which is overridden by the transport).
609 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
612 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
613 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
614 because it is too late at that time, and has no effect.
616 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
617 security issue with \' (bugzilla #107). I could not use the
618 PQescapeStringConn() function, because it needs a PGconn value as one of
621 PH/08 When testing addresses using -bt, indicate those final addresses that
622 are duplicates that would not cause an additional delivery. At least one
623 person was confused, thinking that -bt output corresponded to deliveries.
624 (Suppressing duplicates isn't a good idea as you lose the information
625 about possibly different redirections that led to the duplicates.)
627 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
628 systems where poll() doesn't work, in particular OS X.
630 PH/10 Added more information to debugging output for retry time not reached.
632 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
633 operations in malware.c.
635 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
638 PH/13 If write_rejectlog was set false when logging was sent to syslog with
639 syslog_duplication set false, log lines that would normally be written
640 both the the main log and to the reject log were not written to syslog at
643 PH/14 In the default configuration, change the use of "message" in ACL warn
644 statements to "add_header".
646 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
647 not followed by a command (e.g. "seen endif").
649 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
650 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
653 PH/17 Added extra conditions to the default value of delay_warning_condition
657 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
658 { match{$h_precedence:}{(?i)bulk|list|junk} } \
659 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
662 The Auto-Submitted: and various List- headers are standardised, whereas I
663 don't think Precedence: ever was.
665 PH/18 Refactored debugging code in route_finduser() to show more information,
666 in particular, the error code if getpwnam() issues one.
668 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
669 This is apparently needed in addition to the PH/07 change above to avoid
670 any possible encoding problems.
672 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
673 but not after initializing Perl.
675 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
676 output them only if debugging. By default they are written stderr,
677 apparently, which is not desirable.
679 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
682 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
685 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
687 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
688 authenticated or an ident call has been made. Suppress the default
689 values for $authenticated_id and $authenticated_sender (but permit -oMai
690 and -oMas) when testing with -bh.
692 PH/24 Re-jigged the order of the tests in the default configuration so that the
693 tests for valid domains and recipients precede the DNS black list and CSA
694 tests, on the grounds that those ones are more expensive.
696 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
697 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
698 This bug exists in every version of Exim that I still have, right back to
701 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
702 However, an attempt to turn on case-sensitivity in a regex key by
703 including (?-i) didn't work because the subject string was already
704 lowercased, and the effects were non-intuitive. It turns out that a
705 one-line patch can be used to allow (?-i) to work as expected.
711 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
712 other effects) broke the use of negated acl sub-conditions.
714 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
717 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
718 "Deny" causes Exim to reject the incoming connection with a 554 error.
719 Unfortunately, if there is a major crisis, such as a disk failure,
720 tcp-wrappers gives "deny", whereas what one would like would be some
721 kind of temporary error. A kludge has been added to help with this.
722 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
723 554 error is used if errno is still zero or contains ENOENT (which occurs
724 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
727 PH/03 Add -lutil to the default FreeBSD LIBS setting.
729 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
730 errors. Otherwise a message that provokes a temporary error (when other
731 messages do not) can cause a whole host to time out.
733 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
734 addresses were routed directly to files or pipes from a redirect router.
735 File deliveries just didn't batch; pipe deliveries might have suffered
738 PH/06 A failure to get a lock for a hints database would erroneously always say
739 "Failed to get write lock", even when it was really a read lock.
741 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
742 of 0600. This has been changed to use the value of the lockfile_mode
743 option (which defaults to 0600).
745 PH/08 Applied small patch from the Sieve maintainer.
747 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
748 folder from quota calculations, a direct delivery into this folder messed
749 up the contents of the maildirsize file. This was because the regex was
750 used only to exclude .Trash (or whatever) when the size of the mailbox
751 was calculated. There was no check that a delivery was happening into an
752 excluded directory. This bug has been fixed by ignoring all quota
753 processing for deliveries into excluded directories.
755 PH/10 Added the maildirfolder_create_regex option to appendfile.
761 PH/01 The code for finding all the local interface addresses on a FreeBSD
762 system running IPv6 was broken. This may well have applied to all BSD
763 systems, as well as to others that have similar system calls. The broken
764 code found IPv4 interfaces correctly, but gave incorrect values for the
765 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
766 that it would not match correctly against @[] and not recognize the IPv6
769 PH/02 The ipliteral router was not recognizing addresses of the form user@
770 [ipv6:....] because it didn't know about the "ipv6:" prefix.
772 PH/03 Added disable_ipv6.
774 PH/04 Changed $reply_address to use the raw form of the headers instead of the
775 decoded form, because it is most often used to construct To: headers
776 lines in autoreplies, and the decoded form may well be syntactically
777 invalid. However, $reply_address has leading white space removed, and all
778 newlines turned into spaces so that the autoreply transport does not
781 PH/05 If group was specified without a user on a router, and no group or user
782 was specified on a transport, the group from the router was ignored.
784 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
785 for visible compile-time settings that can be used to change these
786 numbers, for those that want even more. Backwards compatibility with old
787 spool files has been maintained. However, going back to a previous Exim
788 release will lost any variables that are in spool files.
790 PH/07 Two small changes when running in the test harness: increase delay when
791 passing a TCP/IP connection to a new process, in case the original
792 process has to generate a bounce, and remove special handling of
793 127.0.0.2 (sic), which is no longer necessary.
795 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
796 be the same on different OS.
798 PH/09 Moved a debug statement in filter processing to avoid a race problem when
801 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
802 whether --show-vars was specified or not
804 JJ/02 exipick: Added support for new ACL variable spool format introduced
807 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
808 syntactically invalid From: or Reply-to: line, and a filter used this to
809 generate an autoreply, and therefore failed to obtain an address for the
810 autoreply, Exim could try to deliver to a non-existent relative file
811 name, causing unrelated and misleading errors. What now happens is that
812 it logs this as a hard delivery error, but does not attempt to create a
815 PH/11 The exinext utility has a -C option for testing purposes, but although
816 the given file was scanned by exinext itself; it wasn't being passed on
817 when Exim was called.
819 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
820 an end-of-file indication when reading a command response.
822 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
823 compiled. In many other places in Exim, IPv6 addresses are always
824 recognized, so I have changed this. It also means that IPv4 domain
825 literals of the form [IPV4:n.n.n.n] are now always recognized.
827 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
828 used if the router is not running as root, for example, when verifying at
829 ACL time, or when using -bh. The debugging output from this situation was
830 non-existent - all you got was a failure to exec. I have made two
833 (a) Failures to set uid/gid, the current directory, or a process leader
834 in a subprocess such as that created by queryprogram now generate
835 suitable debugging ouput when -d is set.
837 (b) The queryprogram router detects when it is not running as root,
838 outputs suitable debugging information if -d is set, and then runs
839 the subprocess without attempting to change uid/gid.
841 PH/15 Minor change to Makefile for building test_host (undocumented testing
844 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
845 additional section of a DNS packet that returns MX or SRV records.
846 Instead, it always explicitly searches for A/AAAA records. This avoids
847 major problems that occur when a DNS server includes only records of one
848 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
849 fixed another bug: if SRV records were looked up and the corresponding
850 address records were *not* found in the additional section, the port
851 values from the SRV records were lost.
853 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
854 using the correct key (the original address) when searching the retry
855 rules in order to find which one to use for generating the retry hint.
857 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
858 adding the default one. Similarly, if it contains a Reply-To: header, the
859 errors_reply_to option, if set, is not used.
861 PH/19 When calculating a retry time, Exim used to measure the "time since
862 failure" by looking at the "first failed" field in the retry record. Now
863 it does not use this if it is later than than the arrival time of the
864 message. Instead it uses the arrival time. This makes for better
865 behaviour in cases where some deliveries succeed, thus re-setting the
866 "first failed" field. An example is a quota failure for a huge message
867 when small messages continue to be delivered. Without this change, the
868 "time since failure" will always be short, possible causing more frequent
869 delivery attempts for the huge message than are intended.
870 [Note: This change was subsequently modified - see PH/04 for 4.62.]
872 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
873 $1, $2, $3) because the numerical variables can be reset during some
874 expansion items (e.g. "match"), thereby losing the authentication data.
876 PH/21 Make -bV show the size of off_t variables so that the test suite can
877 decide whether to run tests for quotas > 2G.
879 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
880 mailbox_size, and mailbox_filecount in the appendfile transport. If a
881 filecount value is greater than 2G or if a quota value is greater than 2G
882 on a system where the size of off_t is not greater than 4, a panic error
885 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
886 never match. The debug and -bh output now contains an explicit error
887 message indicating a malformed IPv4 address or mask.
889 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
890 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
893 PH/25 Do not write to syslog when running in the test harness. The only
894 occasion when this arises is a failure to open the main or panic logs
895 (for which there is an explicit test).
897 PH/26 Added the /no_tell option to "control=freeze".
899 PH/27 If a host name lookup failed very early in a connection, for example, if
900 the IP address matched host_lookup and the reverse lookup yielded a name
901 that did not have a forward lookup, an error message of the form "no IP
902 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
903 could be logged. Now it outputs the IP address instead of "NULL".
905 PH/28 An enabling patch from MH: add new function child_open_exim2() which
906 allows the sender and the authenticated sender to be set when
907 submitting a message from within Exim. Since child_open_exim() is
908 documented for local_scan(), the new function should be too.
910 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
911 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
912 results in an empty string is now treated as unset.
914 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
916 PH/31 Added sender_verify_fail logging option.
918 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
919 needed by RFC 822 but not by RFC 2822 was commented out. I have now
920 tidied the source and removed it altogether.
922 PH/33 When a queue run was abandoned because the load average was too high, a
923 log line was always written; now it is written only if the queue_run log
924 selector is set. In addition, the log line for abandonment now contains
925 information about the queue run such as the pid. This is always present
926 in "start" and "stop" lines but was omitted from the "abandon" line.
928 PH/34 Omit spaces between a header name and the colon in the error message that
929 is given when verify = headers_syntax fails (if there are lots of them,
930 the message gets confusing).
932 PH/35 Change the default for dns_check_names_pattern to allow slashes within
933 names, as there are now some PTR records that contain slashes. This check
934 is only to protect against broken name servers that fall over on strange
935 characters, so the fact that it applies to all lookups doesn't matter.
937 PH/36 Now that the new test suite is complete, we can remove some of the
938 special code in Exim that was needed for the old test suite. For example,
939 sorting DNS records because real resolvers return them in an arbitrary
940 order. The new test suite's fake resolver always returns records in the
943 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
944 bounces) except when queue_only is set, to avoid logging races between
945 the different processes.
947 PH/38 Panic-die if .include specifies a non-absolute path.
949 PH/39 A tweak to the "H" retry rule from its user.
951 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
952 a label. They prevented compilation on older perls.
954 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
955 a warning to be raised on newish perls.
957 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
958 on queue. Changes to match documented behaviour of showing count of
959 messages matching specified criteria.
961 PH/40 Changed the default ident timeout from 30s to 5s.
963 PH/41 Added support for the use of login_cap features, on those BSD systems
964 that have them, for controlling the resources used by pipe deliveries.
966 PH/42 The content-scanning code uses fopen() to create files in which to put
967 message data. Previously it was not paying any attention to the mode of
968 the files. Exim runs with umask(0) because the rest of the code creates
969 files with open(), and sets the required mode explicitly. Thus, these
970 files were ending up world-writeable. This was not a big issue, because,
971 being within the spool directory, they were not world-accessible. I have
972 created a function called modefopen, which takes an additional mode
973 argument. It sets umask(777), creates the file, chmods it to the required
974 mode, then resets the umask. All the relevant calls to fopen() in the
975 content scanning code have been changed to use this function.
977 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
978 to 24 hours. This avoids potential overflow problems when processing G
979 and H retry rules. I suspect nobody ever tinkers with this value.
981 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
983 PH/45 When the plaintext authenticator is running as a client, the server's
984 challenges are checked to ensure they are valid base64 strings. By
985 default, the authentication attempt is cancelled if an invalid string is
986 received. Setting client_ignore_invalid_base64 true ignores these errors.
987 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
988 they are received. Thus, the responses can be made to depend on the
989 challenges. If an invalid string is ignored, an empty string is placed in
992 PH/46 Messages that are created by the autoreply transport now contains a
993 References: header, in accordance with RFCs 2822 and 3834.
995 PH/47 Added authenticated_sender_force to the smtp transport.
997 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
999 PH/49 Installed latest patch from the Sieve maintainer.
1001 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1002 was also set, the appendfile transport was unnecessarily scanning a
1003 directory of message files (e.g. for maildir delivery) to find the count
1004 of files (along with the size), even though it did not need this
1005 information. It now does the scan only if it needs to find either the
1006 size of the count of files.
1008 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1010 PH/52 Two bugs concerned with error handling when the smtp transport is
1013 (i) Exim was not creating retry information for temporary errors given
1014 for individual recipients after the DATA command when the smtp transport
1015 was used in LMTP mode. This meant that they could be retried too
1016 frequently, and not timed out correctly.
1018 (ii) Exim was setting the flag that allows error details to be returned
1019 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1020 recipients that were returned after the DATA command.
1022 PH/53 This is related to PH/52, but is more general: for any failing address,
1023 when detailed error information was permitted to be returned to the
1024 sender, but the error was temporary, then after the final timeout, only
1025 "retry timeout exceeded" was returned. Now it returns the full error as
1026 well as "retry timeout exceeded".
1028 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1029 do this, and (what is worse) MTAs that accept it.
1031 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1032 will now be deprecated.
1034 PH/56 New os.c-cygwin from the Cygwin maintainer.
1036 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1037 formats (previously only available in exim formats via -bpr, -bpru,
1038 and -bpra. Now also available in native and exiqgrep formats)
1040 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1041 with very large, slow to parse queues
1043 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1045 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1047 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1048 responses to authentication challenges, though it was showing the
1049 challenges; (ii) I've removed the CR characters from the debug output for
1052 PH/58 Allow for the insertion of a newline as well as a space when a string
1053 is turned into more than one encoded-word during RFC 2047 encoding. The
1054 Sieve code now uses this.
1056 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1057 data_4xx, lost_connection, tls_required.
1059 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1060 message was being sent as an SMTP response.
1062 PH/61 Add -l and -k options to exicyclog.
1064 PH/62 When verifying, if an address was redirected to one new address, so that
1065 verification continued, and the new address failed or deferred after
1066 having set something in $address_data, the value of $address_data was not
1067 passed back to the ACL. This was different to the case when no
1068 redirection occurred. The value is now passed back in both cases.
1070 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1071 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1072 use login_cap.h, so on its own it isn't the distinguishing feature. The
1073 new name refers directly to the setclassresources() function.
1075 PH/65 Added configuration files for NetBSD3.
1077 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1079 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1080 is preferred over IPv4.
1082 PH/68 The bounce_return_message and bounce_return_body options were not being
1083 honoured for bounces generated during the reception of non-SMTP messages.
1084 In particular, this applied to messages rejected by the ACL. This bug has
1085 been fixed. However, if bounce_return_message is true and bounce_return_
1086 body is false, the headers that are returned for a non-SMTP message
1087 include only those that have been read before the error was detected.
1088 (In the case of an ACL rejection, they have all been read.)
1090 PH/69 The HTML version of the specification is now built in a directory called
1091 spec_html instead of spec.html, because the latter looks like a path with
1092 a MIME-type, and this confuses some software.
1094 PH/70 Catch two compiler warnings in sieve.c.
1096 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1097 function verify_get_ident() calls ip_connect() to connect a socket, but
1098 if the "connect()" function timed out, ip_connect() used to close the
1099 socket. However, verify_get_ident() also closes the socket later, and in
1100 between Exim writes to the log, which may get opened at this point. When
1101 the socket was closed in ip_connect(), the log could get the same file
1102 descriptor number as the socket. This naturally causes chaos. The fix is
1103 not to close the socket in ip_connect(); the socket should be closed by
1104 the function that creates it. There was only one place in the code where
1105 this was missing, in the iplookup router, which I don't think anybody now
1106 uses, but I've fixed it anyway.
1108 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1109 well as to direct DNS lookups. Otherwise the handling of names in host
1110 lists is inconsistent and therefore confusing.
1116 PH/01 Two changes to the default runtime configuration:
1118 (1) Move the checks for relay_from_hosts and authenticated clients from
1119 after to before the (commented out) DNS black list checks.
1121 (2) Add control=submission to the relay_from_hosts and authenticated
1122 clients checks, on the grounds that messages accepted by these
1123 statements are most likely to be submissions.
1125 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1127 (1) Generate an error if the third argument for the ${prvs expansion is
1130 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1133 (3) Reset the variables that are obtained from the first argument of
1134 ${prvscheck and used in the second argument before leaving the code,
1135 because their memory is reclaimed, so using them afterwards may do
1138 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1139 one (it's much easier than Tom thought :-).
1141 (5) Because of (4), we can now allow for the use of $prvscheck_result
1142 inside the third argument.
1144 PH/03 For some reason, the default setting of PATH when running a command from
1145 a pipe transport was just "/usr/bin". I have changed it to
1148 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1149 anything to be listed in the output from -bV.
1151 PH/05 When a filter generated an autoreply, the entire To: header line was
1152 quoted in the delivery log line, like this:
1154 => >A.N.Other <ano@some.domain> <original@ddress> ...
1156 This has been changed so that it extracts the operative address. There
1157 may be more than one such address. If so, they are comma-separated, like
1160 => >ano@some.domain,ona@other.domain <original@ddress> ...
1162 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1163 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1164 not being looked up in the rDNS to get a host name, Exim was showing the
1165 IP address twice in Received: lines, even though the IP addresses were
1166 identical. For example:
1168 Received: from [1.2.3.4] (helo=[1.2.3.4])
1170 However, if the real host name was known, it was omitting the HELO data
1171 if it matched the actual IP address. This has been tidied up so that it
1172 doesn't show the same IP address twice.
1174 PH/07 When both +timestamp and +memory debugging was on, the value given by
1175 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1176 called by the debug printing, thereby overwriting the timestamp buffer.
1177 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1179 PH/08 When the original message was included in an autoreply transport, it
1180 always said "this is a copy of the message, including all the headers",
1181 even if body_only or headers_only was set. It now gives an appropriate
1184 PH/09 Applied a patch from the Sieve maintainer which:
1186 o fixes some comments
1187 o adds the (disabled) notify extension core
1188 o adds some debug output for the result of if/elsif tests
1189 o points to the current vacation draft in the documentation
1190 and documents the missing references header update
1194 o fixes a bug in processing the envelope test (when testing
1195 multiple envelope elements, the last element determinted the
1198 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1199 Electronic Mail") by including:
1201 Auto-submitted: auto-generated
1203 in the messages that it generates (bounce messages and others, such as
1204 warnings). In the case of bounce messages for non-SMTP mesages, there was
1205 also a typo: it was using "Auto_submitted" (underscore instead of
1206 hyphen). Since every message generated by Exim is necessarily in response
1207 to another message, thes have all been changed to:
1209 Auto-Submitted: auto-replied
1211 in accordance with these statements in the RFC:
1213 The auto-replied keyword:
1215 - SHOULD be used on messages sent in direct response to another
1216 message by an automatic process,
1218 - MUST NOT be used on manually-generated messages,
1220 - MAY be used on Delivery Status Notifications (DSNs) and Message
1221 Disposition Notifications (MDNs),
1223 - MUST NOT be used on messages generated by automatic or periodic
1224 processes, except for messages which are automatic responses to
1227 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1228 to the default Received: header definition.
1230 PH/12 Added log selector acl_warn_skipped (default on).
1232 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1233 variables because (a) they are in the wrong storage pool and (b) even if
1234 they were copied, it wouldn't work properly because of the caching.
1236 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1237 checking when decoding. Apparently there are clients that generate
1238 overlong encoded strings. Why am I not surprised?
1240 PH/15 If the first argument of "${if match_address" was not empty, but did not
1241 contain an "@" character, Exim crashed. Now it writes a panic log message
1242 and treats the condition as false.
1244 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1246 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1247 extension "envelope-auth". The code is finished and in agreement with
1248 other implementations, but there is no documentation so far and in fact,
1249 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1250 not changing the active code.
1252 Print executed "if" and "elsif" statements when debugging is used. This
1253 helps a great deal to understand what a filter does.
1255 Document more things not specified clearly in RFC3028. I had all this
1256 sorted out, when out of a sudden new issues came to my mind. Oops."
1258 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1261 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1262 canonical form (as documented). However, after a host name lookup from
1263 the IP address, check_host() was doing a simple string comparison with
1264 addresses acquired from the DNS when checking that the found name did
1265 have the original IP as one of its addresses. Since any found IPv6
1266 addresses are likely to be in abbreviated form, the comparison could
1267 fail. Luckily, there already exists a function for doing the comparison
1268 by converting both addresses to binary, so now that is used instead of
1269 the text comparison.
1271 PH/20 There was another similar case to PH/19, when a complete host name was
1272 given in a host list; looking up its IP address could give an abbreviated
1273 form, whereas the current host's name might or might not be abbreviated.
1274 The same fix has been applied.
1280 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1281 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1284 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1285 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1287 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1289 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1290 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1291 Also, for "Auto-Submitted", treat anything other than "no" as
1292 non-personal, in accordance with RFC 3834. (Previously it treated
1293 anything starting "auto-" as non-personal.)
1295 TF/01 The control=submission/name=... option had a problem with syntax
1296 errors if the name included a slash character. The /name= option
1297 now slurps the rest of the string, so it can include any characters
1298 but it must come last in the list of options (after /sender_retain
1301 PH/05 Some modifications to the interface to the fake nameserver for the new
1309 TK/01 Added the "success_on_redirect" address verification option. See
1310 NewStuff for rationale and an example.
1312 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1314 PH/02 Patch to exigrep to allow it to work on syslog lines.
1316 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1317 fread() to skip over the body file's header line, because in Cygwin the
1318 header line is locked and is inaccessible.
1320 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1321 co-exist for some time) to make it clear that it is the Exim ID that is
1322 referenced, not the Message-ID: header line.
1324 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1325 string_format() function, because snprintf() does not exist on all
1328 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1329 ${stat: expansion item.
1331 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1332 protocol synchronization error", to keep the pedants happy.
1334 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1335 well as for IRIX systems, when gcc is being used. See the host.c source
1338 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1340 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1343 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1344 transport and to the smtp transport in LMTP mode.
1346 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1348 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1349 run a filter in a subprocess. This could lead to confusion in subsequent
1350 lookups in the parent process. There should also be a search_tidyup() at
1351 the end of the subprocess.
1353 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1354 only if the host matched helo_try_verify_hosts, which caused the
1355 verification to occur when the EHLO/HELO command was issued. The ACL just
1356 tested the remembered result. Now, if a previous verification attempt has
1357 not happened, "verify = helo" does it there and then.
1359 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1361 TK/03 Fix log output including CR from clamd.
1363 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1364 exist provoked a memory error which could cause a segfault.
1366 PH/15 Installed PCRE 6.2
1368 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1370 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1371 of the problem. Specifically, suggested +O2 rather than +O1 for the
1374 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1376 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1377 record, the port was not correctly logged when the outgoing_port log
1378 selector was set (it logged the transort's default port).
1380 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1381 fallback_hosts, and "hosts" in the smtp transport.
1383 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1384 host errors such as "Connection refused".
1386 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1387 authentication with radiusclient 0.4.9:
1389 - Error returned from rc_read_config was caught wrongly
1390 - Username/password not passed on to radius server due to wrong length.
1392 The presumption is that some radiusclient API changes for 4.51/PH/17
1393 were not taken care of correctly. The code is still untested by me (my
1394 Linux distribution still has 0.3.2 of radiusclient), but it was
1395 contributed by a Radius user.
1397 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1398 expanding the "port" option of the smtp transport.
1400 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1401 while reading a MIME header. Thanks to Tom Hughes for a patch.
1403 PH/24 Include config.h inside local_scan.h so that configuration settings are
1406 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1407 that in an ACL for RCPT (for example), you can examine exactly what was
1410 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1411 commands, but it was not correctly comparing the address with the actual
1412 client host address. Thus, it would show the EHLO address in Received:
1413 header lines when this was not necessary.
1415 PH/27 Added the % operator to ${eval:}.
1417 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1418 it should be ignoring failures (because with -C, for example, it has lost
1419 privilege). It wasn't ignoring creation failures other than "already
1422 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1425 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1426 because an input line was too long, either on its own, or by virtue of
1427 too many continuations, the temporary file was not being removed, and the
1428 return code was incorrect.
1430 PH/31 Missing "BOOL" in function definition in filtertest.c.
1432 PH/32 Applied Sieve patches from the maintainer.
1434 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1436 PH/33 Added "verify = not_blind".
1438 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1439 Local/Makefile (with some defaults set). These are used in built scripts
1440 such as exicyclog, but they have never been used in the exim_install
1441 script (though there are many overriding facilities there). I have
1442 arranged that the exim_install script now takes note of these two
1445 PH/35 Installed configuration files for Dragonfly.
1447 PH/36 When a locally submitted message by a trusted user did not contain a
1448 From: header, and the sender address was obtained from -f or from an SMTP
1449 MAIL command, and the trusted user did not use -F to supply a sender
1450 name, $originator_name was incorrectly used when constructing a From:
1451 header. Furthermore, $originator_name was used for submission mode
1452 messages from external hosts without From: headers in a similar way,
1453 which is clearly wrong.
1455 PH/37 Added control=suppress_local_fixups.
1457 PH/38 When log_selector = +received_sender was set, and the addition of the
1458 sender made the log line's construction buffer exactly full, or one byte
1459 less than full, an overflow happened when the terminating "\n" was
1462 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1463 when the result of a list match is failure because a DNS lookup failed.
1465 PH/40 RM_COMMAND is now used in the building process.
1467 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1468 the "build-* directories that it finds.
1470 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1471 address in a domain literal was a prefix of an interface address.
1473 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1474 when verifying a sender address, unless rewrite_headers is false.
1476 PH/44 Wrote a long comment about why errors_to addresses are verified as
1477 recipients, not senders.
1479 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1480 the ratelimit ACL was added.
1482 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1484 PH/46 Added extra information about PostgreSQL errors to the error string.
1486 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1487 avoiding the need to install special zones in a real server. This is
1488 backwards compatible; if it can't find the fake resolver, it drops back.
1489 Thus, both old and new test suites can be run.
1491 TF/02 Added util/ratelimit.pl
1493 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1494 clock is set back in time.
1496 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1497 Candler <B.Candler@pobox.com>.
1499 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1500 OK for addresses that are the result of redirections.
1502 PH/48 A number of further additions for the benefit of the new test suite,
1503 including a fake gethostbyname() that interfaces to the fake DNS resolver
1506 TF/06 The fix for widen_domains has also been applied to qualify_single and
1507 search_parents which are the other dnslookup options that can cause
1510 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1513 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1515 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1516 service" if it encountered an MX record with an empty target hostname.
1517 The message is now "an MX or SRV record indicated no SMTP service".
1519 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1520 if the DNS of the sending site is misconfigured. This is quite a
1521 common situation. This change restores the behaviour of treating a
1522 helo verification defer as a failure.
1524 PH/51 If self=fail was set on a router, the bounce message did not include the
1525 actual error message.
1531 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1533 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1534 command itself ended in error, the underlying message about the transport
1535 filter timeout was being overwritten with the pipe command error. Now the
1536 underlying error message should be appended to the second error message.
1538 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1540 PH/02 Reduce the amount of output that "make" produces by default. Full output
1541 can still be requested.
1543 PH/03 The warning log line about a condition test deferring for a "warn" verb
1544 was being output only once per connection, rather than after each
1545 occurrence (because it was using the same function as for successful
1546 "warn" verbs). This seems wrong, so I have changed it.
1548 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1549 it should not have, which might have caused a crash in the right
1550 circumstances, but probably never did.
1552 PH/04 Installed a modified version of Tony Finch's patch to make submission
1553 mode fix the return path as well as the Sender: header line, and to
1554 add a /name= option so that you can make the user's friendly name appear
1557 TF/03 Added the control = fakedefer ACL modifier.
1559 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1560 Mark Lowes for thorough testing.
1562 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1564 TK/03 Merged latest SRS patch from Miles Wilton.
1566 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1567 with the definition in sysexits.h (which is #included earlier).
1568 Fortunately, Exim does not actually use EX_OK. The code used to try to
1569 preserve the sysexits.h value, by assumimg that macro definitions were
1570 scanned for macro replacements. I have been disabused of this notion,
1571 so now the code just undefines EX_OK before #including unistd.h.
1573 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1574 in the smtp transport. When a block could not be written in a single
1575 write() function, the timeout was being re-applied to each part-write.
1576 This seems wrong - if the receiver was accepting one byte at a time it
1577 would take for ever. The timeout is now adjusted when this happens. It
1578 doesn't have to be particularly precise.
1580 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1581 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1583 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1584 without a domain if the check to <postmaster@domain> fails.
1586 SC/01 Eximstats: added -xls and the ability to specify output files
1587 (patch written by Frank Heydlauf).
1589 SC/02 Eximstats: use FileHandles for outputing results.
1591 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1593 SC/04 Eximstats: fixed display of large numbers with -nvr option
1595 SC/05 Eximstats: fixed merging of reports with empty tables.
1597 SC/06 Eximstats: added the -include_original_destination flag
1599 SC/07 Eximstats: removed tabs and trailing whitespace.
1601 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1603 TK/06 MBOX spool code: Add real "From " MBOX separator line
1604 so the .eml file is really in mbox format (even though
1605 most programs do not really care). Patch from Alex Miller.
1607 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1608 The latter is generated from $received_to and is only set if the
1609 message has one envelope recipient. SA can use these headers,
1610 obviously out-of-the-box. Patch from Alex Miller.
1612 PH/08 The ${def test on a variable was returning false if the variable's
1613 value was "0", contrary to what the specification has always said!
1614 The result should be true unless the variable is empty.
1616 PH/09 The syntax error of a character other than { following "${if
1617 def:variable_name" (after optional whitespace) was not being diagnosed.
1618 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1619 accidental colon was present, for example, could give incorrect results.
1621 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1622 result is used (not including appendfile, where other changes are about
1625 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1626 This involved changing a lot of size variables from int to off_t. It
1627 should work with maildirs and everything.
1629 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1630 spamd dying while we are connected to it.
1632 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1635 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1636 function for BDB 4.3.
1638 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1640 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1641 the sources for PCRE have been re-arranged and more files are now
1644 PH/15 The code I had for printing potentially long long variables in PH/11
1645 above was not the best (it lost precision). The length of off_t variables
1646 is now inspected at build time, and an appropriate printing format (%ld
1647 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1648 to be "long long int" or "long int". This is needed for the internal
1649 formatting function string_vformat().
1651 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1652 the configuration file to be ":syslog", then the script "guesses" where
1653 the logs files are, rather than using the compiled in default. In our
1654 case the guess is not the same as the compiled default, so the script
1655 suddenly stopped working when I started to use syslog. The patch checks
1656 to see if log_file_path is "". If so, it attempts to read it from exim
1657 with no configuration file to get the compiled in version, before it
1658 falls back to the previous guessing code."
1660 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1661 implementing BATV in an Exim configuration. See NewStuff for the gory
1664 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1665 Makefile that are specific to HP-UX.
1667 PH/18 If the "use_postmaster" option was set for a recipient callout together
1668 with the "random" option, the postmaster address was used as the MAIL
1669 FROM address for the random test, but not for the subsequent recipient
1670 test. It is now used for both.
1672 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1673 patch removes a few documentation additions to RFC 3028, because the
1674 latest draft now contains them. It adds the new en;ascii-case comparator
1675 and a new error check for 8bit text in MIME parts. Comparator and
1676 require names are now matched exactly. I enabled the subaddress
1677 extension, but it is not well tested yet (read: it works for me)."
1679 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1680 rework some of the code of TK/09 above to avoid the hardwired use of
1681 "%lld" and "long long". Replaced the call to snprintf() with a call to
1684 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1685 records point to non-existent hosts", "retry timeout exceeded", and
1686 "retry time not reached for any host after a long failure period".
1688 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1689 experimental DomainKeys support:
1691 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1692 (2) On an error such as an illegally used "control", the wrong name for
1693 the control was given.
1695 These problems did NOT occur unless DomainKeys support was compiled.
1697 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1699 PH/24 Added ${if match_ip condition.
1701 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1702 databases so that it will be absolutely obvious if a crash occurs in the
1703 DB library. This is a regular occurrence (often caused by mis-matched
1706 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1707 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1708 picked up on a user's system that detects such things. There doesn't seem
1709 to be a gcc warning option for this - only an attribute that has to be
1710 put on the function's prototype. It seems that in Fedora Core 4 they have
1711 set this on a number of new functions. No doubt there will be more in due
1714 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1715 specify a transport. However, if an address that was verified by such a
1716 router was the subject of a callout, Exim crashed because it tried to
1717 read the rcpt_include_affixes from the non-existent transport. Now it
1718 just assumes that the setting of that option is false. This bug was
1719 introduced by 4.51/PH/31.
1721 PH/28 Changed -d+all to exclude +memory, because that information is very
1722 rarely of interest, but it makes the output a lot bigger. People tend to
1723 do -d+all out of habit.
1725 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1726 code in os-type was giving problems when libc.so lives in lib64, like on
1729 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1730 aren't the modern standard, and it seems that some systems' include files
1731 don't always have them. Exim was already checking for some of the newer
1732 ones like T_AAAA, and defining it itself. I've added checks for all the
1733 record types that Exim uses.
1735 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1736 not automatically generating a new one, as it is supposed to. This
1737 prevented TLS from working. If the file did exist, but contained invalid
1738 data, a new version was generated, as expected. It was only the case of a
1739 non-existent file that was broken.
1741 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1742 with a change in libdomainkeys > 0.64.
1744 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1745 from DNS. If the selector record carries the flag, it now has
1746 precedence over the domain-wide flag.
1748 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1750 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1751 the use of a port name that isn't defined in /etc/services) occurred, the
1752 message was deferred as in a normal delivery, and thus remained on the
1753 spool, instead of being failed because of the mua_wrapper setting. This
1754 is now fixed, and I tidied up some of the mua_wrapper messages at the
1757 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1758 the messages in a hash of arrays rather than using individual hashes.
1759 This is a bit cleaner and results in dramatic memory savings, albeit
1760 at a slight CPU cost.
1762 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1763 as requested by Marc Sherman.
1765 SC/10 Eximstats: added histograms for user specified patterns as requested
1768 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1770 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1771 fopen() in the content-scanning modules that did not already have it.
1777 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1778 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1780 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1782 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1784 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1785 to test Sieve filters that use "vacation".
1787 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1788 that changes the way the GnuTLS parameters are stored in the cache file.
1789 The new format can be generated externally. For backward compatibility,
1790 if the data in the cache doesn't make sense, Exim assumes it has read an
1791 old-format file, and it generates new data and writes a new file. This
1792 means that you can't go back to an older release without removing the
1795 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1796 work if there are any delivery delays because "one_time" forces the
1797 parent to be marked "delivered", so its unseen clone is never tried
1798 again. For this reason, Exim now forbids the simultaneous setting of
1801 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1802 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1803 entry if you want to know the details. The fix, however, neglected to
1804 consider the case when local delivery batching is involved. The test for
1805 "previously delivered" was not happening when checking to see if an
1806 address could be batched with a previous (undelivered) one; under
1807 certain circumstances this could lead to multiple deliveries to the same
1810 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1811 in its include files, and this causes problems building Exim.
1813 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1814 header_syntax) but Exim was just ignoring anything given after a slash.
1815 In particular, this caused confusion with an attempt to use "verify =
1816 reverse_host_lookup/defer_ok". An error is now given when options are
1817 supplied for verify items that do not have them. (Maybe reverse_host_
1818 lookup should have a defer_ok option, but that's a different point.)
1820 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1821 defined by RFC 821) to 2048, because there were problems with some AUTH
1822 commands, and RFC 1869 says the size should be increased for extended
1823 SMTP commands that take arguments.
1825 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1828 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1829 "unknown" error; now it says that the functionality isn't in the binary.
1831 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1832 an address' error message when a string expansion fails (syntax or
1833 whatever). Otherwise the password may appear in the log. Following change
1834 PH/42 below, there is no longer a chance of it appearing in a bounce
1837 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1839 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1840 of its addresses was ever tried. (Bugzilla bug #2.)
1842 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1843 the result incorrectly in the debug output. (It correctly added a newline
1844 to what was transported.)
1846 TF/01 Added $received_time.
1848 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1849 commented out examples of how to interface to a virus scanner and to
1850 SpamAssassin. Also added commented examples of av_scanner and
1851 spamd_address settings.
1853 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1854 and controls are allowed in which ACLs. There were a couple of minor
1855 errors. Some of the entries in the conditions table (which is a table of
1856 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1857 negation of where the condition IS allowed.
1859 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1861 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1862 header file does not have a version number, so I've had to invent a new
1863 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1864 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1865 radiusclient), but it was contributed by a Radius user.
1867 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1868 files or directories, for OpenSSL.
1870 PH/19 When an Exim process that is running as root has to create an Exim log
1871 file, it does so in a subprocess that runs as exim:exim so as to get the
1872 ownership right at creation (otherwise, other Exim processes might see
1873 the file with the wrong ownership). There was no test for failure of this
1874 fork() call, which would lead to the process getting stuck as it waited
1875 for a non-existent subprocess. Forks do occasionally fail when resources
1876 run out. I reviewed all the other calls to fork(); they all seem to check
1879 PH/20 When checking for unexpected SMTP input at connect time (before writing
1880 the banner), Exim was not dealing correctly with a non-positive return
1881 from the read() function. If the client had disconnected by this time,
1882 the result was a log entry for a synchronization error with an empty
1883 string after "input=" when read() returned zero. If read() returned -1
1884 (an event I could not check), uninitialized data bytes were printed.
1885 There were reports of junk text (parts of files, etc) appearing after
1888 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1890 PH/22 Added support for macro redefinition, and (re)definition in between
1891 driver and ACL definitions.
1893 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1894 forgetting to use the resulting value; it was using the unexpanded value.
1896 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1897 hadn't been configured. The fix is from Juergen Kreileder, who
1898 understands it better than I do:
1900 "Here's what I see happening with three configured cyrus_sasl
1901 authenticators configured (plain, login, cram-md5):
1903 On startup auth_cyrus_sasl_init() gets called for each of these.
1904 This means three calls to sasl_listmech() without a specified mech_list.
1905 => SASL tests which mechs of all available mechs actually work
1906 => three warnings about OTP not working
1907 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1909 With the patch, sasl_listmech() also gets called three times. But now
1910 SASL's mech_list option is set to the server_mech specified in the the
1911 authenticator. Or in other words, the answer from sasl_listmech()
1912 gets limited to just the mech you're testing for (which is different
1914 => the return list contains just 'plain' or 'login', 'cram-md5' or
1915 nothing depending on the value of ob->server_mech.
1917 I've just tested the patch: Authentication still works fine,
1918 unavailable mechs specified in the exim configuration are still
1919 caught, and the auth.log warnings about OTP are gone."
1921 PH/25 When debugging is enabled, the contents of the command line are added
1922 to the debugging output, even when log_selector=+arguments is not
1925 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1926 answer is "GNU", and only if the return is "GNU/something" is the answer
1929 PH/27 $acl_verify_message is now set immediately after the failure of a
1930 verification in an ACL, and so is available in subsequent modifiers. In
1931 particular, the message can be preserved by coding like this:
1933 warn !verify = sender
1934 set acl_m0 = $acl_verify_message
1936 Previously, $acl_verify_message was set only while expanding "message"
1937 and "log_message" when a very denied access.
1939 PH/28 Modified OS/os.c-Linux with
1941 -#ifndef OS_LOAD_AVERAGE
1942 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1944 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1945 nomenclature these days.)
1947 PH/29 Installed patch from the Sieve maintainer that adds the options
1948 sieve_useraddress and sieve_subaddress to the redirect router.
1950 PH/30 In these circumstances:
1951 . Two addresses routed to the same list of hosts;
1952 . First host does not offer TLS;
1953 . First host accepts first address;
1954 . First host gives temporary error to second address;
1955 . Second host offers TLS and a TLS session is established;
1956 . Second host accepts second address.
1957 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1958 and peerdn, if requested) that were in fact used only for the second
1961 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1962 attention to any local part prefix or suffix that was matched by the
1963 router that accepted the address. It now behaves in the same way as it
1964 does for delivery: the affixes are removed from the local part unless
1965 rcpt_include_affixes is set on the transport.
1967 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1968 timeout during the DATA phase of an incoming message.
1970 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1971 applied a patch sanctioned by the Sieve maintainer.
1973 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1974 the uid or gid is negative. A case of a negative gid caused this to be
1975 noticed. The fix allows for either to be negative.
1977 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1978 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1981 PH/36 The change PH/12 above was broken. Fixed it.
1983 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1984 the grounds that routing the same address twice would always produce the
1985 same answer. This might have been true once, but it is certainly no
1986 longer true now. Routing a child address may depend on the previous
1987 routing that produced that child. Some complicated redirection strategies
1988 went wrong when messages had multiple recipients, and made Exim's
1989 behaviour dependent on the order in which the addresses were given.
1991 I have moved the duplicate checking until after the routing is complete.
1992 Exim scans the addresses that are assigned to local and remote
1993 transports, and removes any duplicates. This means that more work will be
1994 done, as duplicates will always all be routed, but duplicates are
1995 presumably rare, so I don't expect this is of any significance.
1997 For deliveries to pipes, files, and autoreplies, the duplicate checking
1998 still happens during the routing process, since they are not going to be
2001 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2002 It corrects a timeout issue with spamd. This is Ian's comment: "The
2003 background is that sometimes spamd either never reads data from a
2004 connection it has accepted, or it never writes response data. The exiscan
2005 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2006 blindly assumes that writes won't block so it may never time out."
2008 PH/39 Allow G after quota size as well as K and M.
2010 PH/40 The value set for $authenticated_id in an authenticator may not contain
2011 binary zeroes or newlines because the value is written to log lines and
2012 to spool files. There was no check on this. Now the value is run through
2013 the string_printing() function so that such characters are converted to
2014 printable escape sequences.
2016 PH/41 $message_linecount is a new variable that contains the total number of
2017 lines in the message. Compare $body_linecount, which is the count for the
2020 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2021 bounce and delay warning messages, except in certain special cases, which
2024 (a) An SMTP error message from a remote host;
2025 (b) A message specified in a :fail: redirection;
2026 (c) A message specified in a "fail" command in a system filter;
2027 (d) A message specified in a FAIL return from the queryprogram router;
2028 (e) A message specified by the cannot_route_message router option.
2030 In these cases only, Exim does include the error details in bounce and
2031 warning messages. There are also a few cases where bland messages such
2032 as "unrouteable address" or "local delivery error" are given.
2034 PH/43 $value is now also set for the "else" part of a ${run expansion.
2036 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2037 being worked on, but at least Exim now implements the latest version to
2040 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2041 process to complete was treated as a delivery failure, a timeout while
2042 writing the message to the pipe was logged, but erroneously treated as a
2043 successful delivery. Such timeouts include transport filter timeouts. For
2044 consistency with the overall process timeout, these timeouts are now
2045 treated as errors, giving rise to delivery failures by default. However,
2046 there is now a new Boolean option for the pipe transport called
2047 timeout_defer, which, if set TRUE, converts the failures into defers for
2048 both kinds of timeout. A transport filter timeout is now identified in
2051 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2052 systems where "make" and "gmake" are different, calling "gmake" at top
2053 level broke things. I've arranged for the value of $(MAKE) to be passed
2054 from the Makefile to this script so that it can call the same version of
2058 A note about Exim versions 4.44 and 4.50
2059 ----------------------------------------
2061 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2062 changes of various kinds. As a consequence, a big documentation update was
2063 needed. This delayed the release for rather longer than seemed good, especially
2064 in the light of a couple of (minor) security issues. Therefore, the changes
2065 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2066 release. So 4.44 and 4.50 are in effect two different branches that both start
2069 I have left the 4.50 change log unchanged; it contains all the changes since
2070 4.43. The change log for 4.44 is below; many of its items are identical to
2071 those for 4.50. This seems to be the most sensible way to preserve the
2072 historical information.
2078 1. Minor wording change to the doc/README.SIEVE file.
2080 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2081 computation of the current number of files was incorrect.
2083 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2084 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2087 4. Give more explanation in the error message when the command for a transport
2088 filter fails to execute.
2090 5. There are several places where Exim runs a non-Exim command in a
2091 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2092 was being done only for the command run by the queryprogram router. It is
2093 now done for all such subprocesses. The other cases are: ${run, transport
2094 filters, and the commands run by the lmtp and pipe transports.
2096 6. Added CONFIGURE_GROUP build-time option.
2098 7. Some older OS have a limit of 256 on the maximum number of file
2099 descriptors. Exim was using setrlimit() to set 1000 as a large value
2100 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2101 systems. I've change it so that if it can't get 1000, it tries for 256.
2103 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2104 was an oversight, and furthermore, ever since the addition of extra
2105 controls (e.g. 4.43/32), the checks on when to allow different forms of
2106 "control" were broken. There should now be diagnostics for all cases when a
2107 control that does not make sense is encountered.
2109 9. Added the /retain_sender option to "control=submission".
2111 10. $recipients is now available in the predata ACL (oversight).
2113 11. Tidy the search cache before the fork to do a delivery from a message
2114 received from the command line. Otherwise the child will trigger a lookup
2115 failure and thereby defer the delivery if it tries to use (for example) a
2116 cached ldap connection that the parent has called unbind on.
2118 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2119 of $address_data from the recipient verification was clobbered by the
2120 sender verification.
2122 13. The value of address_data from a sender verification is now available in
2123 $sender_address_data in subsequent conditions in the ACL statement.
2125 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2127 15. Added a new option "connect=<time>" to callout options, to set a different
2130 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2131 was its contents. (It was OK if the option was not defined at all.)
2133 17. A "Completed" log line is now written for messages that are removed from
2134 the spool by the -Mrm option.
2136 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2137 information about exactly what failed.
2139 19. Added -dd to debug only the daemon process.
2141 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2142 handles timeouts, both on the server side and network timeouts. Renamed the
2143 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2145 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2146 It is now set to "smtps".
2148 22. $host_address is now set to the target address during the checking of
2149 ignore_target_hosts.
2151 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2152 being passed; this would have caused $sender_host_name to have been used if
2153 matching the list had actually called for a host name (not very likely,
2154 since this list is usually IP addresses). A host name is now passed as
2157 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2158 code that specifies a non-restarting handler (typically sigaction() in
2159 modern systems) in an attempt to fix a rare and obscure crash bug.
2161 25. Narrowed the window for a race in the daemon that could cause it to ignore
2162 SIGCHLD signals. This is not a major problem, because they are used only to
2163 wake it up if nothing else does.
2165 26. A malformed maildirsize file could cause Exim to calculate negative values
2166 for the mailbox size or file count. Odd effects could occur as a result.
2167 The maildirsize information is now recalculated if the size or filecount
2170 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2171 support for a long time. Removed HAVE_SYS_VFS_H.
2173 28. Installed the latest version of exipick from John Jetmore.
2175 29. In an address list, if the pattern was not a regular expression, an empty
2176 subject address (from a bounce message) matched only if the pattern was an
2177 empty string. Non-empty patterns were not even tested. This was the wrong
2178 because it is perfectly reasonable to use an empty address as part of a
2179 database query. An empty address is now tested by patterns that are
2180 lookups. However, all the other forms of pattern expect the subject to
2181 contain a local part and a domain, and therefore, for them, an empty
2182 address still always fails if the pattern is not itself empty.
2184 30. Exim went into a mad DNS loop when attempting to do a callout where the
2185 host was specified on an smtp transport, and looking it up yielded more
2186 than one IP address.
2188 31. Re-factored the code for checking spool and log partition space into a
2189 function that finds that data and another that does the check. The former
2190 is then used to implement four new variables: $spool_space, $log_space,
2191 $spool_inodes, and $log_inodes.
2193 32. The RFC2047 encoding function was originally intended for short strings
2194 such as real names; it was not keeping to the 75-character limit for
2195 encoded words that the RFC imposes. It now respects the limit, and
2196 generates multiple encoded words if necessary. To be on the safe side, I
2197 have increased the buffer size for the ${rfc2047: expansion operator from
2200 33. It is now permitted to omit both strings after an "if" condition; if the
2201 condition is true, the result is "true". As before, when the second string
2202 is omitted, a false condition yields an empty string. This makes it less
2203 cumbersome to write custom ACL and router conditions.
2205 34. Failure to deliver a bounce message always caused it to be frozen, even if
2206 there was an errors_to setting on the router. The errors_to setting is now
2209 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2210 canonical form (fully expanded) before being placed in
2211 $sender_host_address.
2213 36. The table in the code that translates DNS record types into text (T_A to
2214 "A" for instance) was missing entries for NS and CNAME. It is just possible
2215 that this could have caused confusion if both these types were looked up
2216 for the same domain, because the text type is used as part of Exim's
2217 per-process caching. But the chance of anyone hitting this buglet seems
2220 37. The dnsdb lookup has been extended in a number of ways.
2222 (1) There is a new type, "zns", which walks up the domain tree until it
2223 finds some nameserver records. It should be used with care.
2225 (2) There is a new type, "mxh", which is like "mx" except that it returns
2226 just the host names, not the priorities.
2228 (3) It is now possible to give a list of domains (or IP addresses) to be
2229 looked up. The behaviour when one of the lookups defers can be
2230 controlled by a keyword.
2232 (4) It is now possible to specify the separator character for use when
2233 multiple records are returned.
2235 38. The dnslists ACL condition has been extended: it is now possible to supply
2236 a list of IP addresses and/or domains to be looked up in a particular DNS
2239 39. Added log_selector=+queue_time_overall.
2241 40. When running the queue in the test harness, wait just a tad after forking a
2242 delivery process, to get repeatability of debugging output.
2244 41. Include certificate and key file names in error message when GnuTLS fails
2245 to set them up, because the GnuTLS error message doesn't include the name
2246 of the failing file when there is a problem reading it.
2248 42. Allow both -bf and -bF in the same test run.
2250 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2252 44. The "Exiscan patch" is now merged into the mainline Exim source.
2254 45. Sometimes the final signoff response after QUIT could fail to get
2255 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2256 < 0 before doing a fflush(). This bug looks as though it goes back to the
2257 introduction of TLS in release 3.20, but "sometimes" must have been rare
2258 because the tests only now provoked it.
2260 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2261 (this can affect the format of dates).
2263 47. exim_tidydb, when checking for the continued existence of a message for
2264 which it has found a message-specific retry record, was not finding
2265 messages that were in split spool directories. Consequently, it was
2266 deleting retry records that should have stayed in existence.
2268 48. Steve fixed some bugs in eximstats.
2270 49. The SPA authentication driver was not abandoning authentication and moving
2271 on to the next authenticator when an expansion was forced to fail,
2272 contradicting the general specification for all authenticators. Instead it
2273 was generating a temporary error. It now behaves as specified.
2275 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2276 (the order specifies the preference for clients). The order is now AES256,
2277 AES128, 3DES, ARCFOUR128.
2279 51. Small patch to Sieve code - explicitly set From: when generating an
2282 52. Exim crashed if a remote delivery caused a very long error message to be
2283 recorded - for instance if somebody sent an entire SpamAssassin report back
2284 as a large number of 550 error lines. This bug was coincidentally fixed by
2285 increasing the size of one of Exim's internal buffers (big_buffer) that
2286 happened as part of the Exiscan merge. However, to be on the safe side, I
2287 have made the code more robust (and fixed the comments that describe what
2290 53. Now that there can be additional text after "Completed" in log lines (if
2291 the queue_time_overall log selector is set), a one-byte patch to exigrep
2292 was needed to allow it to recognize "Completed" as not the last thing in
2295 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2296 patch that reportedly fixes this has been added. I am not expert enough to
2297 create a test for it. This is what the patch creator wrote:
2299 "I found a little strange behaviour of ldap code when working with
2300 Windows 2003 AD Domain, where users was placed in more than one
2301 Organization Units. When I tried to give exim partial DN, the exit code
2302 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2303 But simultaneously result of request was absolutely normal ldap result,
2304 so I produce this patch..."
2306 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2307 REFERENCE, so I have modified the code to exclude the patch when that macro
2310 55. Some experimental protocols are using DNS PTR records for new purposes. The
2311 keys for these records are domain names, not reversed IP addresses. The
2312 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2313 leaves it alone. Component reversal etc. now happens only for IP addresses.
2316 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2318 57. Double the size of the debug message buffer (to 2048) so that more of very
2319 long debug lines gets shown.
2321 58. The exicyclog utility now does better if the number of log files to keep
2322 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2324 59. Two changes related to the smtp_active_hostname option:
2326 (1) $smtp_active_hostname is now available as a variable.
2327 (2) The default for smtp_banner uses $smtp_active_hostname instead
2328 of $primary_hostname.
2330 60. The host_aton() function is supposed to be passed a string that is known
2331 to be a valid IP address. However, in the case of IPv6 addresses, it was
2332 not checking this. This is a hostage to fortune. Exim now panics and dies
2333 if the condition is not met. A case was found where this could be provoked
2334 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2335 components; fortuitously, this particular loophole had already been fixed
2336 by change 4.50/55 above.
2338 If there are any other similar loopholes, the new check in host_aton()
2339 itself should stop them being exploited. The report I received stated that
2340 data on the command line could provoke the exploit when Exim was running as
2341 exim, but did not say which command line option was involved. All I could
2342 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2343 running as the user.
2346 61. There was a buffer overflow vulnerability in the SPA authentication code
2347 (which came originally from the Samba project). I have added a test to the
2348 spa_base64_to_bits() function which I hope fixes it.
2351 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2352 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2354 63. The daemon start-up calls getloadavg() while still root for those OS that
2355 need the first call to be done as root, but it missed one case: when
2356 deliver_queue_load_max is set with deliver_drop_privilege. This is
2357 necessary for the benefit of the queue runner, because there is no re-exec
2358 when deliver_drop_privilege is set.
2360 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2361 This has been fixed.
2363 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2364 was in use, was not putting the data itself into the right store pool;
2365 consequently, it could be overwritten for a subsequent message in the same
2366 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2369 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2371 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2372 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2373 the function were treating the return as a boolean value, which happened to
2374 work because 0=false and not-0=true, but is not correct code.
2376 68. The host_aton() function was not handling scoped IPv6 addresses (those
2377 with, for example, "%eth0" on the end) correctly.
2379 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2380 negated items (that is, ~something) in unsigned ints. Some compilers
2381 apparently mutter when there is no cast.
2383 70. If an address verification called from an ACL failed, and did not produce a
2384 user-specific message (i.e. there was only a "system" message), nothing was
2385 put in $acl_verify_message. In this situation, it now puts the system
2388 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2389 session; change 4.31/43 added the unwanted input to the log line - except
2390 that it did not do this in the start of session case. It now does.
2392 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2393 This is wrong and can cause the other end to generate a synchronization
2394 error if it is another Exim or anything else that does the synchronization
2395 check. A QUIT command is no longer sent after a timeout.
2397 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2398 during host lookups.
2400 74. The defer_ok option of callout verification was not working if it was used
2401 when verifying addresses in header lines, that is, for this case:
2403 verify = header_sender/callout=defer_ok
2405 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2406 those file descriptors could be used for SMTP connections. If anything
2407 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2408 could be sent to the SMTP client, causing chaos. The daemon now opens
2409 stdin, stdout, and stderr to /dev/null when it puts itself into the
2412 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2413 log by default. The user can override this with suitable Perl magic.
2415 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2416 the log message when discard triggers, was not working for the DATA ACL or
2417 for the non-SMTP ACL.
2419 78. Error message wording change in sieve.c.
2421 79. If smtp_accept_max_per_host was set, the number of connections could be
2422 restricted to fewer than expected, because the daemon was trying to set up
2423 a new connection before checking whether the processes handling previous
2424 connections had finished. The check for completed processes is now done
2425 earlier. On busy systems, this bug wouldn't be noticed because something
2426 else would have woken the daemon, and it would have reaped the completed
2429 80. If a message was submitted locally by a user whose login name contained one
2430 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2431 It caused a spool format error. I have fixed the spool reading code. A
2432 related problem was that the "from" clause in the Received: line became
2433 illegal because of the space(s). It is now covered by ${quote_local_part.
2435 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2438 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2440 83. Patch from Sieve maintainer to fix unterminated string problem in
2441 "vacation" handling.
2443 84. Some minor changes to the Linux configuration files to help with other
2444 OS variants using glibc.
2446 85. One more patch for Sieve to update vacation handling to latest spec.
2449 ----------------------------------------------------
2450 See the note above about the 4.44 and 4.50 releases.
2451 ----------------------------------------------------
2457 1. Change 4.43/35 introduced a bug that caused file counts to be
2458 incorrectly computed when quota_filecount was set in an appendfile
2461 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2462 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2465 3. Give more explanation in the error message when the command for a transport
2466 filter fails to execute.
2468 4. There are several places where Exim runs a non-Exim command in a
2469 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2470 was being done only for the command run by the queryprogram router. It is
2471 now done for all such subprocesses. The other cases are: ${run, transport
2472 filters, and the commands run by the lmtp and pipe transports.
2474 5. Some older OS have a limit of 256 on the maximum number of file
2475 descriptors. Exim was using setrlimit() to set 1000 as a large value
2476 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2477 systems. I've change it so that if it can't get 1000, it tries for 256.
2479 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2480 was an oversight, and furthermore, ever since the addition of extra
2481 controls (e.g. 4.43/32), the checks on when to allow different forms of
2482 "control" were broken. There should now be diagnostics for all cases when a
2483 control that does not make sense is encountered.
2485 7. $recipients is now available in the predata ACL (oversight).
2487 8. Tidy the search cache before the fork to do a delivery from a message
2488 received from the command line. Otherwise the child will trigger a lookup
2489 failure and thereby defer the delivery if it tries to use (for example) a
2490 cached ldap connection that the parent has called unbind on.
2492 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2493 of $address_data from the recipient verification was clobbered by the
2494 sender verification.
2496 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2497 was its contents. (It was OK if the option was not defined at all.)
2499 11. A "Completed" log line is now written for messages that are removed from
2500 the spool by the -Mrm option.
2502 12. $host_address is now set to the target address during the checking of
2503 ignore_target_hosts.
2505 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2506 being passed; this would have caused $sender_host_name to have been used if
2507 matching the list had actually called for a host name (not very likely,
2508 since this list is usually IP addresses). A host name is now passed as
2511 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2512 code that specifies a non-restarting handler (typically sigaction() in
2513 modern systems) in an attempt to fix a rare and obscure crash bug.
2515 15. Narrowed the window for a race in the daemon that could cause it to ignore
2516 SIGCHLD signals. This is not a major problem, because they are used only to
2517 wake it up if nothing else does.
2519 16. A malformed maildirsize file could cause Exim to calculate negative values
2520 for the mailbox size or file count. Odd effects could occur as a result.
2521 The maildirsize information is now recalculated if the size or filecount
2524 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2525 support for a long time. Removed HAVE_SYS_VFS_H.
2527 18. Updated exipick to current release from John Jetmore.
2529 19. Allow an empty sender to be matched against a lookup in an address list.
2530 Previously the only cases considered were a regular expression, or an
2533 20. Exim went into a mad DNS lookup loop when doing a callout where the
2534 host was specified on the transport, if the DNS lookup yielded more than
2537 21. The RFC2047 encoding function was originally intended for short strings
2538 such as real names; it was not keeping to the 75-character limit for
2539 encoded words that the RFC imposes. It now respects the limit, and
2540 generates multiple encoded words if necessary. To be on the safe side, I
2541 have increased the buffer size for the ${rfc2047: expansion operator from
2544 22. Failure to deliver a bounce message always caused it to be frozen, even if
2545 there was an errors_to setting on the router. The errors_to setting is now
2548 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2549 canonical form (fully expanded) before being placed in
2550 $sender_host_address.
2552 24. Updated eximstats to version 1.33
2554 25. Include certificate and key file names in error message when GnuTLS fails
2555 to set them up, because the GnuTLS error message doesn't include the name
2556 of the failing file when there is a problem reading it.
2558 26. Expand error message when OpenSSL has problems setting up cert/key files.
2561 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2562 (this can affect the format of dates).
2564 28. exim_tidydb, when checking for the continued existence of a message for
2565 which it has found a message-specific retry record, was not finding
2566 messages that were in split spool directories. Consequently, it was
2567 deleting retry records that should have stayed in existence.
2569 29. eximstats updated to version 1.35
2570 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2571 1.35 - bugfix such that pie charts by volume are generated correctly
2573 30. The SPA authentication driver was not abandoning authentication and moving
2574 on to the next authenticator when an expansion was forced to fail,
2575 contradicting the general specification for all authenticators. Instead it
2576 was generating a temporary error. It now behaves as specified.
2578 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2579 (the order specifies the preference for clients). The order is now AES256,
2580 AES128, 3DES, ARCFOUR128.
2582 31. Small patch to Sieve code - explicitly set From: when generating an
2585 32. Exim crashed if a remote delivery caused a very long error message to be
2586 recorded - for instance if somebody sent an entire SpamAssassin report back
2587 as a large number of 550 error lines. This bug was coincidentally fixed by
2588 increasing the size of one of Exim's internal buffers (big_buffer) that
2589 happened as part of the Exiscan merge. However, to be on the safe side, I
2590 have made the code more robust (and fixed the comments that describe what
2593 33. Some experimental protocols are using DNS PTR records for new purposes. The
2594 keys for these records are domain names, not reversed IP addresses. The
2595 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2596 leaves it alone. Component reversal etc. now happens only for IP addresses.
2599 34. The host_aton() function is supposed to be passed a string that is known
2600 to be a valid IP address. However, in the case of IPv6 addresses, it was
2601 not checking this. This is a hostage to fortune. Exim now panics and dies
2602 if the condition is not met. A case was found where this could be provoked
2603 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2604 components; fortuitously, this particular loophole had already been fixed
2605 by change 4.50/55 or 4.44/33 above.
2607 If there are any other similar loopholes, the new check in host_aton()
2608 itself should stop them being exploited. The report I received stated that
2609 data on the command line could provoke the exploit when Exim was running as
2610 exim, but did not say which command line option was involved. All I could
2611 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2612 running as the user.
2615 35. There was a buffer overflow vulnerability in the SPA authentication code
2616 (which came originally from the Samba project). I have added a test to the
2617 spa_base64_to_bits() function which I hope fixes it.
2620 36. The daemon start-up calls getloadavg() while still root for those OS that
2621 need the first call to be done as root, but it missed one case: when
2622 deliver_queue_load_max is set with deliver_drop_privilege. This is
2623 necessary for the benefit of the queue runner, because there is no re-exec
2624 when deliver_drop_privilege is set.
2626 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2627 was in use, was not putting the data itself into the right store pool;
2628 consequently, it could be overwritten for a subsequent message in the same
2629 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2632 38. Sometimes the final signoff response after QUIT could fail to get
2633 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2634 < 0 before doing a fflush(). This bug looks as though it goes back to the
2635 introduction of TLS in release 3.20, but "sometimes" must have been rare
2636 because the tests only now provoked it.
2642 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2643 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2644 Now it returns an integer. A number of calls were still expecting a T/F
2645 return. Fortuitously, in all cases, the tests worked in OK situations,
2646 which is the norm. However, things would have gone wrong on any write
2647 failures on the smtp file descriptor. This function is used when sending
2648 messages over SMTP and also when doing verify callouts.
2650 2. When Exim is called to do synchronous delivery of a locally submitted
2651 message (the -odf or -odi options), it no longer closes stderr before doing
2654 3. Implemented the mua_wrapper option.
2656 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2658 5. Implemented the functions header_remove(), header_testname(),
2659 header_add_at_position(), and receive_remove_recipient(), and exported them
2662 6. If an ACL "warn" statement specified the addition of headers, Exim already
2663 inserted X-ACL-Warn: at the start if there was no header name. However, it
2664 was not making this test for the second and subsequent header lines if
2665 there were newlines in the string. This meant that an invalid header could
2666 be inserted if Exim was badly configured.
2668 7. Allow an ACL "warn" statement to add header lines at the start or after all
2669 the Received: headers, as well as at the end.
2671 8. Added the rcpt_4xx retry error code.
2673 9. Added postmaster_mailfrom=xxx to callout verification option.
2675 10. Added mailfrom=xxxx to the callout verification option, for verify=
2678 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2679 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2681 12. Inserted some casts to stop certain compilers warning when using pointer
2682 differences as field lengths or precisions in printf-type calls (mostly
2683 affecting debugging statements).
2685 13. Added optional readline() support for -be (dynamically loaded).
2687 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2688 same clock tick as a message's arrival, so that its received time was the
2689 same as the "first fail" time on the retry record, and that message
2690 remained on the queue past the ultimate address timeout, every queue runner
2691 would try a delivery (because it was past the ultimate address timeout) but
2692 after another failure, the ultimate address timeout, which should have then
2693 bounced the address, did not kick in. This was a "< instead of <=" error;
2694 in most cases the first failure would have been in the next clock tick
2695 after the received time, and all would be well.
2697 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2698 being recognized when the domain list was tested by the match_domain
2699 condition in an expansion string.
2701 16. Added the ${str2b64: operator.
2703 17. Exim was always calling setrlimit() to set a large limit for the number of
2704 processes, without checking whether the existing limit was already
2705 adequate. (It did check for the limit on file descriptors.) Furthermore,
2706 errors from getrlimit() and setrlimit() were being ignored. Now they are
2707 logged to the main and panic logs, but Exim does carry on, to try to do its
2708 job under whatever limits there are.
2710 18. Imported PCRE 5.0.
2712 19. Trivial typo in log message " temporarily refused connection" (the leading
2715 20. If the log selector return_path_on_delivery was set and an address was
2716 redirected to /dev/null, the delivery process crashed because it assumed
2717 that a return path would always be set for a "successful" delivery. In this
2718 case, the whole delivery is bypassed as an optimization, and therefore no
2721 21. Internal re-arrangement: the function for sending a challenge and reading
2722 a response while authentication was assuming a zero-terminated challenge
2723 string. It's now changed to take a pointer and a length, to allow for
2724 binary data in such strings.
2726 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2728 23. Exim was not respecting finduser_retries when seeking the login of the
2729 uid under which it was called; it was always trying 10 times. (The default
2730 setting of finduser_retries is zero.) Also, it was sleeping after the final
2731 failure, which is pointless.
2733 24. Implemented tls_on_connect_ports.
2735 25. Implemented acl_smtp_predata.
2737 26. If the domain in control=submission is set empty, Exim assumes that the
2738 authenticated id is a complete email address when it generates From: or
2739 Sender: header lines.
2741 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2742 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2743 chown and chgrp in /bin and hostname in /usr/bin.
2745 28. Exim was keeping the "process log" file open after each use, just as it
2746 does for the main log. This opens the possibility of it remaining open for
2747 long periods when the USR1 signal hits a daemon. Occasional processlog
2748 errors were reported, that could have been caused by this. Anyway, it seems
2749 much more sensible not to leave this file open at all, so that is what now
2752 29. The long-running daemon process does not normally write to the log once it
2753 has entered its main loop, and it closes the log before doing so. This is
2754 so that log files can straightforwardly be renamed and moved. However,
2755 there are a couple of unusual error situations where the daemon does write
2756 log entries, and I had neglected to close the log afterwards.
2758 30. The text of an SMTP error response that was received during a remote
2759 delivery was being truncated at 512 bytes. This is too short for some of
2760 the long messages that one sometimes sees. I've increased the limit to
2763 31. It is now possible to make retry rules that apply only when a message has a
2764 specific sender, in particular, an empty sender.
2766 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2767 it possible to be selective about when SMTP synchronization is enforced.
2769 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2771 32. Implemented hosts_connection_nolog.
2773 33. Added an ACL for QUIT.
2775 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2778 35. Added mailbox_size and mailbox_filecount to appendfile.
2780 36. Added control = no_multiline_responses to ACLs.
2782 37. There was a bug in the logic of the code that waits for the clock to tick
2783 in the case where the clock went backwards by a substantial amount such
2784 that the microsecond fraction of "now" was more than the microsecond
2785 fraction of "then" (but the whole seconds number was less).
2787 38. Added support for the libradius Radius client library this is found on
2788 FreeBSD (previously only the radiusclient library was supported).
2794 1. When certain lookups returned multiple values in the form name=value, the
2795 quoting of the values was not always being done properly. Specifically:
2796 (a) If the value started with a double quote, but contained no whitespace,
2798 (b) If the value contained whitespace other than a space character (i.e.
2799 tabs or newlines or carriage returns) it was not quoted.
2800 This fix has been applied to the mysql and pgsql lookups by writing a
2801 separate quoting function and calling it from the lookup code. The fix
2802 should probably also be applied to nisplus, ibase and oracle lookups, but
2803 since I cannot test any of those, I have not disturbed their existing code.
2805 2. A hit in the callout cache for a specific address caused a log line with no
2806 reason for rejecting RCPT. Now it says "Previous (cached) callout
2807 verification failure".
2809 3. There was an off-by-one bug in the queryprogram router. An over-long
2810 return line was truncated at 256 instead of 255 characters, thereby
2811 overflowing its buffer with the terminating zero. As well as fixing this, I
2812 have increased the buffer size to 1024 (and made a note to document this).
2814 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2815 when Exim is waiting for an SMTP response from a remote server, Exim
2816 restarts its select() call on the socket, thereby resetting its timeout.
2817 This is not a problem when such interrupts are rare. Somebody set up a cron
2818 job to run exiwhat every 2 minutes, which is less than the normal select()
2819 timeout (5 or 10 minutes). This meant that the select() timeout never
2820 kicked in because it was always reset. I have fixed this by comparing the
2821 time when an interrupt arrives with the time at the start of the first call
2822 to select(). If more time than the timeout has elapsed, the interrupt is
2823 treated as a timeout.
2825 5. Some internal re-factoring in preparation for the addition of Sieve
2826 extensions (by MH). In particular, the "personal" test is moved to a
2827 separate function, and given an option for scanning Cc: and Bcc: (which is
2828 not set for Exim filters).
2830 6. When Exim created an email address using the login of the caller as the
2831 local part (e.g. when creating a From: or Sender: header line), it was not
2832 quoting the local part when it contained special characters such as @.
2834 7. Installed new OpenBSD configuration files.
2836 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2837 try to make them clearer.
2839 9. Callout options, other than the timeout value, were being ignored when
2840 verifying sender addresses in header lines. For example, when using
2842 verify = header_sender/callout=no_cache
2844 the cache was (incorrectly) being used.
2846 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2847 only the Cygwin environment.
2849 11. When return_path_on_delivery was set as a log selector, if different remote
2850 addresses in the same message used different return paths and parallel
2851 remote delivery occurred, the wrong values would sometimes be logged.
2852 (Whenever a remote delivery process finished, the return path value from
2853 the most recently started remote delivery process was logged.)
2855 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2856 lines when AUTH and/or TLS are in use. This is the "received protocol"
2857 field. Exim used to use "asmtp" for authenticated SMTP, without any
2858 indication (in the protocol name) for TLS use. Now it follows the RFC and
2859 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2860 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2861 names appear in log lines as well as in Received: header lines.
2863 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2864 extensions, and comparison tests, and to fix some bugs.
2866 14. Changes to the "personal" filter test:
2868 (1) The test was buggy in that it was just doing the equivalent of
2869 "contains" tests on header lines. For example, if a user's address was
2870 anne@some.where, the "personal" test would incorrectly be true for
2872 To: susanne@some.where
2874 This test is now done by extracting each address from the header in turn,
2875 and checking the entire address. Other tests that are part of "personal"
2876 are now done using regular expressions (for example, to check local parts
2877 of addresses in From: header lines).
2879 (2) The list of non-personal local parts in From: addresses has been
2880 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2881 taken from the Sieve specification recommendations.
2883 (3) If the message contains any header line starting with "List-" it is
2884 treated as non-personal.
2886 (4) The test for "circular" in the Subject: header line has been removed
2887 because it now seems ill-conceived.
2889 15. Minor typos in src/EDITME comments corrected.
2891 16. Installed latest exipick from John Jetmore.
2893 17. If headers_add on a router specified a text string that was too long for
2894 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2895 of string_sprintf() is now avoided.
2897 18. $message_body_size was not set (it was always zero) when running the DATA
2898 ACL and the local_scan() function.
2900 19. For the "mail" command in an Exim filter, no default was being set for
2901 the once_repeat time, causing a random time value to be used if "once" was
2902 specified. (If the value happened to be <= 0, no repeat happened.) The
2903 default is now 0s, meaning "never repeat". The "vacation" command was OK
2904 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2905 (I found it when inspecting the code).
2907 20. There is now an overall timeout for performing a callout verification. It
2908 defaults to 4 times the callout timeout, which applies to individual SMTP
2909 commands during the callout. The overall timeout applies when there is more
2910 than one host that can be tried. The timeout is checked before trying the
2911 next host. This prevents very long delays if there are a large number of
2912 hosts and all are timing out (e.g. when the network connections are timing
2913 out). The value of the overall timeout can be changed by specifying an
2914 additional sub-option for "callout", called "maxwait". For example:
2916 verify = sender/callout=5s,maxwait=20s
2918 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2919 to the end before writing, but this should make it even safer).
2921 22. Exim was forgetting that it had advertised PIPELINING for the second and
2922 subsequent messages on an SMTP connection. It was also not resetting its
2923 memory on STARTTLS and an internal HELO.
2925 23. When Exim logs an SMTP synchronization error within a session, it now
2926 records whether PIPELINING has been advertised or not.
2928 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2929 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2930 rather than long int.
2932 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2934 26. Added the never_mail option to autoreply.
2940 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2941 crash if the getsockname() call failed; this can happen if a connection is
2942 closed very soon after it is established. The problem was simply in the
2943 order in which certain operations were done, causing Exim to try to write
2944 to the SMTP stream before it had set up the file descriptor. The bug has
2945 been fixed by making things happen in the correct order.
2951 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2952 before the connection was closed, thus losing the rejection response.
2954 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2955 some early Solaris releases, but causes trouble in current releases where
2956 socklen_t is defined.
2958 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2961 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2964 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2965 that releasing the top of it at the end releases what was used for sub-
2966 expansions (unless the block got too big). However, discard this block if
2967 the first thing is a variable or header, so that we can use its block when
2968 it is dynamic (useful for very large $message_headers, for example).
2970 6. Lookups now cache *every* query, not just the most recent. A new, separate
2971 store pool is used for this. It can be recovered when all lookup caches are
2972 flushed. Lookups now release memory at the end of their result strings.
2973 This has involved some general refactoring of the lookup sources.
2975 7. Some code has been added to the store_xxx() functions to reduce the amount
2976 of flapping under certain conditions.
2978 8. log_incoming_interface used to affect only the <= reception log lines. Now
2979 it causes the local interface and port to be added to several more SMTP log
2980 lines, for example "SMTP connection from", and rejection lines.
2982 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2984 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2986 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2987 could be overwritten at the end of the current message (or the start of a
2988 new message if it was set in a HELO ACL). The value is now preserved for
2989 the duration of the SMTP connection.
2991 12. If a transport had a headers_rewrite setting, and a matching header line
2992 contained an unqualified address, that address was qualified, even if it
2993 did not match any rewriting rules. The underlying bug was that the values
2994 of the flags that permit the existence of unqualified sender and recipient
2995 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2996 non-local messages, and by -bnq for local messages) were not being
2997 preserved with the message after it was received.
2999 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3000 "next input=" as part of the text comprising the host identity instead of
3001 the correct text. The code was using the same buffer for two different
3002 strings. However, depending on which order the printing function evaluated
3003 its arguments, the bug did not always show up. Under Linux, for example, my
3004 test suite worked just fine.
3006 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3007 doesn't work with some older versions of Perl. It has been changed to "my",
3008 which in any case is probably the better facility to use.
3010 15. A really picky compiler found some instances of statements for creating
3011 error messages that either had too many or two few arguments for the format
3014 16. The size of the buffer for calls to the DNS resolver has been increased
3015 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3016 for addresses that have a lot of PTR records. This alleviates a problem; it
3017 does not fully solve it.
3019 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3020 buffer now truncates the list and logs the incident, which is the same
3021 action as happens when Exim is looking up a host name and its aliases.
3022 Previously in this situation something unpredictable would happen;
3023 sometimes it was "internal error: store_reset failed".
3025 18. If a server dropped the connection unexpectedly when an Exim client was
3026 using GnuTLS and trying to read a response, the client delivery process
3027 crashed while trying to generate an error log message.
3029 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3030 string, for example:
3032 warn message = H1: something\nH2: something
3034 the text was added as a single header line from Exim's point of view
3035 though it ended up OK in the delivered message. However, searching for the
3036 second and subsequent header lines using $h_h2: did not work. This has been
3037 fixed. Similarly, if a system filter added multiple headers in this way,
3038 the routers could not see them.
3040 20. Expanded the error message when iplsearch is called with an invalid key to
3041 suggest using net-iplsearch in a host list.
3043 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3044 ACLs are no longer actually imposed (and a message to that effect is
3047 22. If a "gecos" field in a passwd entry contained escaped characters, in
3048 particular, if it contained a \" sequence, Exim got it wrong when building
3049 a From: or a Sender: header from that name. A second bug also caused
3050 incorrect handling when an unquoted " was present following a character
3051 that needed quoting.
3053 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3054 was not being matched caselessly.
3056 24. Arranged for all hyphens in the exim.8 source to be escaped with
3059 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3060 properly. Recipient callout cache records were still being keyed to include
3061 the sender, even when use_sender was set false. This led to far more
3062 callouts that were necessary. The sender is no longer included in the key
3063 when use_sender is false.
3065 26. Added "control = submission" modifier to ACLs.
3067 27. Added the ${base62d: operator to decode base 62 numbers.
3069 28. dnsdb lookups can now access SRV records.
3071 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3072 the configuration file.
3074 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3075 (-v) mode. This makes the output for a verbose queue run more intelligible.
3077 31. Added a use_postmaster feature to recipient callouts.
3079 32. Added the $body_zerocount variable, containing the number of binary zero
3080 bytes in the message body.
3082 33. The time of last modification of the "new" subdirectory is now used as the
3083 "mailbox time last read" when there is a quota error for a maildir
3086 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3088 35. Added +ignore_unknown as a special item in host lists.
3090 36. Code for decoding IPv6 addresses in host lists is now included, even if
3091 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3092 address was recognized as an IP address, but was then not correctly decoded
3093 into binary, causing unexpected and incorrect effects when compared with
3100 1. Very minor rewording of debugging text in manualroute to say "list of
3101 hosts" instead of "hostlist".
3103 2. If verify=header_syntax was set, and a header line with an unqualified
3104 address (no domain) and a large number of spaces between the end of the
3105 name and the colon was received, the reception process suffered a buffer
3106 overflow, and (when I tested it) crashed. This was caused by some obsolete
3107 code that should have been removed. The fix is to remove it!
3109 3. When running in the test harness, delay a bit after writing a bounce
3110 message to get a bit more predictability in the log output.
3112 4. Added a call to search_tidyup() just before forking a reception process. In
3113 theory, someone could use a lookup in the expansion of smtp_accept_max_
3114 per_host which, without the tidyup, could leave open a database connection.
3116 5. Added the variables $recipient_data and $sender_data which get set from a
3117 lookup success in an ACL "recipients" or "senders" condition, or a router
3118 "senders" option, similar to $domain_data and $local_part_data.
3120 6. Moved the writing of debug_print from before to after the "senders" test
3123 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3124 problems for message scanning, either using a data ACL, or using
3125 local_scan() because the Received: header was not generated till after they
3126 were called (in order to set the time as the time of reception completion).
3127 I have revised the way this works. The header is now generated after the
3128 body is received, but before the ACL or local_scan() are called. After they
3129 are run, the timestamp in the header is updated.
3135 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3136 before starting a queue runner without re-exec. This happened only when
3137 deliver_drop_privilege was set or when the Exim user was set to root. The
3138 effect of the bug was that timeouts during subsequent deliveries caused
3139 crashes instead of being properly handled. The handler is now left at its
3140 default (and expected) setting.
3142 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3143 message, again when deliver_drop_privilege is set or Exim is run as root.
3144 The bug described in (1) was not present in this case, but the tidying up
3145 of the other signals was missing. I have made the two cases consistent.
3147 3. The ignore_target_hosts setting on a manualroute router was being ignored
3148 for hosts that were looked up using the /MX notation.
3150 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3153 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3154 operated on the sender address. After changing the $sender_address to <>
3155 for the sender address verify, Exim was re-instated it as the original
3156 (before rewriting) address, but remembering that it had rewritten it, so it
3157 wasn't rewriting it again. This bug also had the effect of breaking the
3158 sender address verification caching when the sender address was rewritten.
3160 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3161 This has been changed so that if the ip literal address matches
3162 ignore_target_hosts, the router declines.
3164 7. Added expansion conditions match_domain, match_address, and match_local_
3165 part (NOT match_host).
3167 8. The placeholder for the Received: header didn't have a length field set.
3169 9. Added code to Exim itself and to exim_lock to test for a specific race
3170 condition that could lead to file corruption when using MBX delivery. The
3171 issue is with the lockfile that is created in /tmp. If this file is removed
3172 after a process has opened it but before that process has acquired a lock,
3173 there is the potential for a second process to recreate the file and also
3174 acquire a lock. This could lead to two Exim processes writing to the file
3175 at the same time. The added code performs the same test as UW imapd; it
3176 checks after acquiring the lock that its file descriptor still refers to
3177 the same named file.
3179 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3180 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3183 11. Added the smtp_active_hostname option. If used, this will typically be made
3184 to depend on the incoming interface address. Because $interface_address is
3185 not set up until the daemon has forked a reception process, error responses
3186 that can happen earlier (such as "too many connections") no longer contain
3189 12. If an expansion in a condition on a "warn" statement fails because a lookup
3190 defers, the "warn" statement is abandoned, and the next ACL statement is
3191 processed. Previously this caused the whole ACL to be aborted.
3193 13. Added the iplsearch lookup type.
3195 14. Added ident_timeout as a log selector.
3197 15. Added tls_certificate_verified as a log selector.
3199 16. Added a global option tls_require_ciphers (compare the smtp transport
3200 option of the same name). This controls incoming TLS connections.
3202 17. I finally figured out how to make tls_require_ciphers do a similar thing
3203 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3204 before starting the TLS session.
3206 18. Tabs are now shown as \t in -bP output.
3208 19. If the log selector return_path_on_delivery was set, Exim crashed when
3209 bouncing a message because it had too many Received: header lines.
3211 20. If two routers both had headers_remove settings, and the first one included
3212 a superfluous trailing colon, the final name in the first list and the
3213 first name in the second list were incorrectly joined into one item (with a
3214 colon in the middle).
3220 1. Added -C and -D options to the exinext utility, mainly to make it easier
3221 to include in the automated testing, but these could be helpful when
3222 multiple configurations are in use.
3224 2. The exinext utility was not formatting the output nicely when there was
3225 an alternate port involved in the retry record key, nor when there was a
3226 message id as well (for retries that were specific to a specific message
3227 and a specific host). It was also confused by IPv6 addresses, because of
3228 the additional colons they contain. I have fixed the IPv4 problem, and
3229 patched it up to do a reasonable job for IPv6.
3231 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3232 delivery, the log line now contains "pipelined" if PIPELINING was used.
3234 4. An SMTP transport process used to panic and die if the bind() call to set
3235 an explicit outgoing interface failed. This has been changed; it is now
3236 treated in the same way as a connect() failure.
3238 5. A reference to $sender_host_name in the part of a conditional expansion
3239 that was being skipped was still causing a DNS lookup. This no longer
3242 6. The def: expansion condition was not recognizing references to header lines
3243 that used bh_ and bheader_.
3245 7. Added the _cache feature to named lists.
3247 8. The code for checking quota_filecount in the appendfile transport was
3248 allowing one more file than it should have been.
3250 9. For compatibility with Sendmail, the command line option
3258 and sets the incoming protocol and host name (for trusted callers). The
3259 host name and its colon can be omitted when only the protocol is to be set.
3260 Note the Exim already has two private options, -pd and -ps, that refer to
3261 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3262 "s", but I don't think that's a major issue.
3264 10. A number of refactoring changes to the code, none of which should affect
3267 (a) The number of logging options was getting close to filling up the
3268 32-bit word that was used as a bit map. I have split them into two classes:
3269 those that are passed in the argument to log_write(), and those that are
3270 only ever tested independently outside of that function. These are now in
3271 separate 32-bit words, so there is plenty of room for expansion again.
3272 There is no change in the user interface or the logging behaviour.
3274 (b) When building, for example, log lines, the code previously used a
3275 macro that called string_cat() twice, in order to add two strings. This is
3276 not really sufficiently general. Furthermore, there was one instance where
3277 it was actually wrong because one of the argument was used twice, and in
3278 one call a function was used. (As it happened, calling the function twice
3279 did not affect the overall behaviour.) The macro has been replaced by a
3280 function that can join an arbitrary number of extra strings onto a growing
3283 (c) The code for expansion conditions now uses a table and a binary chop
3284 instead of a serial search (which was left over from when there were very
3285 few conditions). Also, it now recognizes conditions like "pam" even when
3286 the relevant support is not compiled in: a suitably worded error message is
3287 given if an attempt is made to use such a condition.
3289 11. Added ${time_interval:xxxxx}.
3291 12. A bug was causing one of the ddress fields not to be passed back correctly
3292 from remote delivery subprocesses. The field in question was not being
3293 subsequently used, so this caused to problems in practice.
3295 13. Added new log selectors queue_time and deliver_time.
3297 14. Might have fixed a bug in maildirsizefile handling that threw up
3298 "unexpected character" debug warnings, and recalculated the data
3299 unnecessarily. In any case, I expanded the warning message to give more
3302 15. Added the message "Restricted characters in address" to the statements in
3303 the default ACL that block characters like @ and % in local parts.
3305 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3306 Three changes have been made:
3308 (a) There was a serious bug; a negative response to MAIL caused the whole
3309 recipient domain to be cached as invalid, thereby blocking all messages
3310 to all local parts at the same domain, from all senders. This bug has
3311 been fixed. The domain is no longer cached after a negative response to
3312 MAIL if the sender used is not empty.
3314 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3317 (c) A new callout option, "use_sender" has been added for people who want
3318 the modified behaviour.
3324 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3327 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3328 indeed breaks things for older releases.
3330 3. Added additional logging to the case where there is a problem reading data
3331 from a filter that is running in a subprocess using a pipe, in order to
3332 try to track down a specific problem.
3334 4. Testing facility fudge: when running in the test harness and attempting
3335 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3336 getting "No route to host". Convert this to a timeout.
3338 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3341 6. Some OS don't have socklen_t but use size_t instead. This affects the
3342 fifth argument of getsockopt() amongst other things. This is now
3343 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3344 can be set for individual OS. I have set it for SunOS5, OSF1, and
3345 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3346 some earlier ones do not.
3348 7. Change 4.30/15 was not doing the test caselessly.
3350 8. The standard form for an IPv6 address literal was being rejected by address
3351 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3352 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3353 this, as well as the form without the "IPv6" on the front (but only when
3354 address literals are enabled, of course).
3356 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3358 10. Exim crashed if a message with an empty sender address specified by -f
3359 encountered a router with an errors_to setting. This could be provoked only
3360 by a command such as
3364 where an empty string was supplied; "<>" did not hit this bug.
3366 11. Installed PCRE release 4.5.
3368 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3369 remained set. It is now erased.
3371 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3372 times from message ids (which are base 36 rather than the normal 62).
3374 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3375 were being counted as actual protocol errors, and logged if the log
3376 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3377 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3378 and DATA following a set of rejected RCPTs do not count as protocol errors.
3379 In other words, Exim assumes they were pipelined, though this may not
3380 actually be the case. Of course, in all cases the client gets an
3381 appropriate error code.
3383 15. If a lookup fails in an ACL condition, a message about the failure may
3384 be available; it is used if testing the ACL cannot continue, because most
3385 such messages specify what the cause of the deferral is. However, some
3386 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3387 that caused an old message to be retained and used if a later statement
3388 caused a defer, replacing the real cause of the deferral.
3390 16. If an IP address had so many PTR records that the DNS lookup buffer
3391 was not large enough to hold them, Exim could crash while trying to process
3392 the truncated data. It now detects and logs this case.
3394 17. Further to 4.21/58, another change has been made: if (and only if) the
3395 first line of a message (the first header line) ends with CRLF, a bare LF
3396 in a subsequent header line has a space inserted after it, so as not to
3397 terminate the header.
3399 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3400 unnecessarily, used atoi() instead of strtol(), and didn't check the
3401 termination when getting file sizes from file names by regex.
3403 19. Completely re-implemented the support for maildirsize files, in the light
3404 of a number of problems with the previous contributed implementation
3405 (4.30/29). In particular:
3407 . If the quota is zero, the maildirsize file is maintained, but no quota is
3410 . If the maildir directory does not exist, it is created before any attempt
3411 to write a maildirsize file.
3413 . The quota value in the file is just a cache; if the quota is changed in
3414 the transport, the new value overrides.
3416 . A regular expression is available for excluding directories from the
3419 20. The autoreply transport checks the characters in options that define the
3420 message's headers; it allows continued headers, but it was checking with
3421 isspace() after an embedded newline instead of explicitly looking for a
3424 21. If all the "regular" hosts to which an address was routed had passed their
3425 expiry times, and had not reached their retry times, the address was
3426 bounced, even if fallback hosts were defined. Now Exim should go on to try
3429 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3430 equivalent code in the SMTP transport. Some hosts send humungous responses
3431 to HELO/EHLO, more than 1024 it seems.
3433 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3434 gives compiler warnings in some environments. I've now done it "properly",
3437 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3438 (because of problems with the built-in one) was declared to return uschar *
3439 instead of char *, causing compiler failure.
3441 25. Fixed a file descriptor leak when processing alias/forward files.
3443 26. Fixed a minor format string issue in dbfn.c.
3445 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3447 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3448 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3451 29. When a sender address is verified, it is cached, to save repeating the test
3452 when there is more than one recipient in a message. However, when the
3453 verification involves a callout, it is possible for different callout
3454 options to be set for different recipients. It is too complicated to keep
3455 track of this in the cache, so now Exim always runs a verification when a
3456 callout is required, relying on the callout cache for the optimization.
3457 The overhead is duplication of the address routing, but this should not be
3460 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3461 to be verified with callout=postmaster, and the main callout worked but the
3462 postmaster check failed, the verification correctly failed. However, if a
3463 subsequent RCPT command asked for sender verification *without* the
3464 postmaster check, incorrect caching caused this verification also to fail,
3467 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3468 it was not caching the DNS options (qualify_single, search_parents) that
3469 were used when the lookup failed. A subsequent lookup with different
3470 options therefore always gave the same answer, though there were cases
3471 where it should not have. (Example: a "domains = !$mx_any" option on a
3472 dnslookup router: the "domains" option is always processed without any
3473 widening, but the router might have qualify_single set.) Now Exim uses the
3474 cached value only when the same options are set.
3476 32. Added John Jetmore's "exipick" utility to the distribution.
3478 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3479 than a timeout (e.g. a certificate is required, and is not provided), an
3480 Exim server now closes the connection immediately. Previously it waited for
3481 the client to close - but if the client is SSL, it seems that they each
3482 wait for each other, leading to a delay before one of them times out.
3484 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3485 maintained 0.8.x compatibility because I don't think many are using it, and
3486 it is clearly obsolete.
3488 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3491 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3492 client certificate was expired. A simple patch fixes this, though I don't
3493 understand the full logic of why the verify callback is called multiple
3496 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3497 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3498 which causes problems with some clients (such as the Certicom SSL Plus
3499 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3500 disables the coutermeasure allowing Eudora to connect."
3502 38. Exim was not checking that a write() to a log file succeeded. This could
3503 lead to Bad Things if a log got too big, in particular if it hit a file
3504 size limit. Exim now panics and dies if it cannot write to a log file, just
3505 as it does if it cannot open a log file.
3507 39. Modified OS/Makefile-Linux so that it now contains
3509 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3511 The two -D definitions ensure that Exim is compiled with large file
3512 support, which makes it possible to handle log files that are bigger than
3515 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3516 instance) a domain was checked against a named list that involved a lookup,
3517 causing $domain_data to be set, then another domain was checked against the
3518 same list, then the first domain was re-checked, the value of $domain_data
3519 after the final check could be wrong. In particular, if the second check
3520 failed, it could be set empty. This bug probably also applied to
3523 41. The strip_trailing_dot option was not being applied to the address given
3524 with the -f command-line option.
3526 42. The code for reading a message's header from the spool was incrementing
3527 $received_count, but never initializing it. This meant that the value was
3528 incorrect (doubled) while delivering a message in the same process in which
3529 it was received. In the most common configuration of Exim, this never
3530 happens - a fresh exec is done - but it can happen when
3531 deliver_drop_privilege is set.
3533 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3534 it now includes up to 150 characters of the unexpected data in the log
3537 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3538 and building data strings. The size of both of these buffers was 10 000
3539 bytes - far larger than anybody would *ever* want, thought I. Needless to
3540 say, somebody hit the limit. I have increased the maximum line length to
3541 20 000 and the maximum data length of concatenated lines to 100 000. I have
3542 also fixed two bugs, because there was no checking on these buffers. Tsk,
3543 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3544 buffer is too small.
3546 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3547 lsearch lookups. Now it does.
3549 46. When parsing a route_list item in a manualroute router, a fixed-length
3550 buffer was used for the list of hosts. I made this 1024 bytes long,
3551 thinking that nobody would ever have a list of hosts that long. Wrong.
3552 Somebody had a whole pile of complicated expansion conditions, and the
3553 string was silently truncated, leading to an expansion error. It turns out
3554 that it is easier to change to an unlimited length (owing to other changes
3555 that have happened since this code was originally written) than to build
3556 structure for giving a limitation error. The length of the item that
3557 expands into the list of hosts is now unlimited.
3559 47. The lsearch lookup could not handle data where the length of text line was
3560 more than 4095 characters. Such lines were truncated, leading to shortened
3561 data being returned. It should now handle lines of any length.
3563 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3564 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3567 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3568 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3569 no longer happens in comment lines. A list of replacements is now placed
3570 at the head of all of the source files, except those whose only change is
3571 to replace PERL_COMMAND in the very first #! line.
3573 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3574 messages on the queue, with a bottom-up merge sort, using code contributed
3575 by Michael Haardt. This should make operations like -bp somewhat faster on
3576 large queues. It won't affect queue runners, except when queue_run_in_order
3579 51. Installed eximstats 1.31 in the distribution.
3581 52. Added support for SRV lookups to the dnslookup router.
3583 53. If an ACL referred to $message_body or $message_body_end, the value was not
3584 reset for any messages that followed in the same SMTP session.
3586 54. The store-handling optimization for building very long strings was not
3587 differentiating between the different store pools. I don't think this
3588 actually made any difference in practice, but I've tidied it.
3590 55. While running the routers to verify a sender address, $sender_address
3591 was still set to the sender address. This is wrong, because when routing to
3592 send a bounce to the sender, it would be empty. Therefore, I have changed
3593 it so that, while verifying a sender address, $sender_address is set to <>.
3594 (There is no change to what happens when verifying a recipient address.)
3596 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3597 target A or AAAA records (if not already returned) without resetting the
3598 qualify_single or search_parents options of the DNS resolver. These are
3599 inappropriate in this case because the targets of MX and SRV records must
3600 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3601 target that, when qualified, matched something in the local domain. These
3602 two options are now turned off when doing these lookups.
3604 57. It seems that at least some releases of Reiserfs (which does not have the
3605 concept of a fixed number of inodes) returns zero and not -1 for the
3606 number of available inodes. This interacted badly with check_spool_inodes,
3607 which assumed that -1 was the "no such thing" setting. What I have done is
3608 to check that the total number of inodes is greater than zero before doing
3609 the test of how many are available.
3611 58. When a "warn" ACL statement has a log_message modifier, the message is
3612 remembered, and not repeated. This is to avoid a lot of repetition when a
3613 message has many recipients that cause the same warning to be written.
3614 Howewer, Exim was preserving the list of already written lines for an
3615 entire SMTP session, which doesn't seem right. The memory is now reset if a
3616 new message is started.
3618 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3619 debugging output unless log_rewrite was also set.
3621 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3622 of (int)(handle) when we know that handle contains (void *)(-1).
3624 61. The Exim daemon panic-logs an error return when it closes the incoming
3625 connection. However "connection reset by peer" seems to be common, and
3626 isn't really an error worthy of noting specially, so that particular error
3629 62. When Exim is trying to find all the local interfaces, it used to panic and
3630 die if the ioctl to get the interface flags failed. However, it seems that
3631 on at least one OS (Solaris 9) it is possible to have an interface that is
3632 included in the list of interfaces, but for which you get a failure error
3633 for this call. This happens when the interface is not "plumbed" into a
3634 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3635 failure of the "get flags" call assumes that the interface is down.
3637 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3638 makes life easier for people who are doing arithmetic on fields extracted
3639 from dates, where you often get leading zeros that should not be
3640 interpreted as octal.
3642 64. Added qualify_domain to the redirect router, to override the global
3645 65. If a pathologically long header line contained very many addresses (the
3646 report of this problem mentioned 10 000) and each of them was rewritten,
3647 Exim could use up a very large amount of memory. (It kept on making new
3648 copies of the header line as it rewrote, and never released the old ones.)
3649 At the expense of a bit more processing, the header rewriting function has
3650 been changed so that it no longer eats memory in this way.
3652 66. The generation of the Received: header has been moved from the time that a
3653 message starts to be received, to the time that it finishes. The timestamp
3654 in the Received: header should now be very close to that of the <= log
3655 line. There are two side-effects of this change:
3657 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3658 logged header lines no longer include the local Received: line, because
3659 it has not yet been created. The same applies to a copy of the message
3660 that is returned to a non-SMTP sender when a message is rejected.
3662 (b) When a filter file is tested using -bf, no additional Received: header
3663 is added to the test message. After some thought, I decided that this
3666 This change does not affect the value of $received_for. It is still set
3667 after address rewriting, but before local_scan() is called.
3669 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3671 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3672 gave an unhelpful panic error message, and a defer error. I have managed to
3673 change this behaviour so that it now rejects any supplied certificate,
3674 which seems right, as the list of acceptable certificates is empty.
3676 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3677 gave an unhelpful defer error. I have not managed to make this reject any
3678 supplied certificates, but the error message it gives is "no certificate
3679 supplied", which is not helpful.
3681 70. exigrep's output now also includes lines that are not associated with any
3682 message, but which match the given pattern. Implemented by a patch from
3683 Martin Sluka, which also tidied up the Perl a bit.
3685 71. Recipient callout verification, like sender verification, was using <> in
3686 the MAIL FROM command. This isn't really the right thing, since the actual
3687 sender may affect whether the remote host accepts the recipient or not. I
3688 have changed it to use the actual sender in the callout; this means that
3689 the cache record is now keyed on a recipient/sender pair, not just the
3690 recipient address. There doesn't seem to be a real danger of callout loops,
3691 since a callout by the remote host to check the sender would use <>.
3692 [SEE ABOVE: changed after hitting problems.]
3694 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3695 temporary errors. However, in the case of such a code being given after
3696 the end of a data transmission (i.e. after ".") Exim was failing to write
3697 a retry record for the message. (Yes, there was some broken host that was
3698 actually sending 8xx at this point.)
3700 73. An unknown lookup type in a host list could cause Exim to panic-die when
3701 the list was checked. (An example that provoked this was putting <; in the
3702 middle of a list instead of at the start.) If this happened during a DATA
3703 ACL check, a -D file could be left lying around. This kind of configuration
3704 error no longer causes Exim to die; instead it causes a defer errror. The
3705 incident is still logged to the main and panic logs.
3707 74. Buglet left over from Exim 3 conversion. The message "too many messages
3708 in one connection" was written to the rejectlog but not the mainlog, except
3709 when address rewriting (yes!) was being logged.
3711 75. Added write_rejectlog option.
3713 76. When a system filter was run not as root (that is, when system_filter_user
3714 was set), the values of the $n variables were not being returned to the
3715 main process; thus, they were not subsequently available in the $sn
3718 77. Added +return_path_on_delivery log selector.
3720 78. A connection timeout was being treated differently from recipients deferred
3721 when testing hosts_max_try with a message that was older than the host's
3722 retry timeout. (The host should not be counted, thus allowing all hosts to
3723 be tried at least once before bouncing.) This may have been the cause of an
3724 occasionally reported bug whereby a message would remain on the queue
3725 longer than the retry timeout, but would be bounced if a delivery was
3726 forced. I say "may" because I never totally pinned down the problem;
3727 setting up timeout/retry tests is difficult. See also the next item.
3729 79. The ultimate address timeout was not being applied to errors that involved
3730 a combination of host plus message (for example, a timeout on a MAIL
3731 command). When an address resolved to a number of possible hosts, and they
3732 were not all tried for each delivery (e.g. because of hosts_max_try), a
3733 message could remain on the queue longer than the retry timeout.
3735 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3738 81. Fixed an obscure SMTP outgoing bug which required at least the following
3739 conditions: (a) there was another message waiting for the same server;
3740 (b) the server returned 5xx to all RCPT commands in the first message so
3741 that the message was not completed; (c) the server dropped the connection
3742 or gave a negative response to the RSET that Exim sends to abort the
3743 transaction. The observed case was a dropped connection after DATA that had
3744 been sent in pipelining mode. That is, the server had advertised PIPELINING
3745 but was not implementing it correctly. The effect of the bug was incorrect
3746 behaviour, such as trying another host, and this could lead to a crash.
3752 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3753 and daemon.c were passed as pointers to ints; they should have been
3754 pointers to socklen_t variables (which are typically unsigned ints).
3756 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3759 3. Fixed a really odd bug that affected only the testing scheme; patching a
3760 certain fixed string in the binary changed the value of another string that
3761 happened to be identical to the end of the original first string.
3763 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3764 name", it returns that address as the IP address. On some operating
3765 systems (e.g. Solaris), it also passes back the IP address string as the
3766 "host name". However, on others (e.g. Linux), it passes back an empty
3767 string. Exim wasn't checking for this, and was changing the host name to an
3768 empty string, assuming it had been canonicized.
3770 5. Although rare, it is permitted to have more than one PTR record for a given
3771 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3772 all the names associated with an address, because they do in Solaris.
3773 However, it seems that they do not in Linux for data that comes from the
3774 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3775 I found this out when I moved to a new Linux workstation and tried to run
3776 the Exim test suite.
3778 To get round this problem I have changed the code so that it now does its
3779 own call to the DNS to look up PTR records when searching for a host name.
3780 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3781 addresses that are only in /etc/hosts are still found.
3783 This behaviour is, however, controlled by an option called host_lookup_
3784 order, which defaults to "bydns:byaddr". If people want to use the other
3785 order, or indeed, just use one or the other means of lookup, they can
3786 specify it in this variable.
3788 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3789 some operating systems, this comes back from gethostbyaddr() as an empty
3790 string, and this is what Exim used to test for. However, it seems that in
3791 other systems, "." is yielded. Exim now tests for this case too.
3793 7. The values of check_spool_space and check_log_space are now held internally
3794 as a number of kilobytes instead of an absolute number of bytes. If a
3795 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3796 kilobyte. This means that much larger values can be stored.
3798 8. Exim monitor: an attempt to get the action menu when not actually pointing
3799 at a message produces an empty menu entitled "No message selected". This
3800 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3801 no entries in it ("Shell widget menu has zero width and/or height"). So I
3802 have added a single, blank menu entry in this case.
3804 9. Added ${quote_local_part.
3806 10. MIME decoding is now applied to the contents of Subject: header lines when
3809 11. Now that a reference to $sender_host_address automatically causes a reverse
3810 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3811 host lookup before query-style lookups in lists that might use this
3812 variable. This has therefore been abolished, and the "net-" prefix is no
3813 longer necessary for query-style lookups.
3815 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3816 have been a typo for LFLAGS, so it has been changed.
3818 13. The install script calls Exim with "-C /dev/null" in order to find the
3819 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3820 to be output. Howeve, since Exim outputs its version number before the
3821 error, it didn't break the script. It just looked ugly. I fixed this by
3822 always allowing "-C /dev/null" if the caller is root.
3824 14. Ignore overlarge ACL variable number when reading spool file - insurance
3825 against a later release with more variables having written the file.
3827 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3828 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3829 this, as well as the form without the "IPv6" on the front.
3831 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3832 OS/Makefile-Darwin file.
3834 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3835 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3837 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3838 variable. However, some OS use ints for inodes. I've added cast to long int
3839 to get rid of the compiler warning.
3841 19. I had forgotten to lock out "/../" in configuration file names when
3842 ALT_CONFIG_PREFIX was set.
3844 20. Routers used for verification do not need to specify transports. However,
3845 if such a router generated a host list, and callout was configured, Exim
3846 crashed, because it could not find a port number from the (non-existent)
3847 transport. It now assumes port 25 in this circumstance.
3849 21. Added the -t option to exigrep.
3851 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3852 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3853 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3854 them is set, LOOKUP_LSEARCH is forced.
3856 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3858 24. Added sender and host information to the "rejected by local_scan()" log
3859 line; previously there was no indication of these.
3861 25. Added .include_if_exists.
3863 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3864 Linux. It turns out that not all file systems support this. Apparently some
3865 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3866 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3867 not supported on the file descriptor, is now ignored when Exim is trying to
3868 sync a directory. This applies only to Linux.
3870 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3872 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3873 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3876 29. Added support for maildirsize files from supplied patch (modified a bit).
3878 30. The use of :fail: followed by an empty string could lead Exim to respond to
3879 sender verification failures with (e.g.):
3881 550 Verification failed for <xxx>
3882 550 Sender verify failed
3884 where the first response line was missing the '-' that indicates it is not
3885 the final line of the response.
3887 31. The loop for finding the name of the user that called Exim had a hardwired
3888 limit of 10; it now uses the value of finduser_retries, which is used for
3889 all other user lookups.
3891 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3894 33. Exim was neglecting to zero errno before one call of strtol() when
3895 expanding a string and expecting an integer value. On some systems this
3896 resulted in spurious "integer overflow" errors. Also, it was casting the
3897 result into an int without checking.
3899 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3900 did not work. The code looks as if it has *never* worked, though it appears
3901 to have been documented since at least releast 1.62. I have made it work.
3903 35. The "timeout_DNS" error in retry rules, also documented since at least
3904 1.62, also never worked. As it isn't clear exactly what this means, and
3905 clearly it isn't a major issue, I have abolished the feature by treating it
3906 as "timeout", and writing a warning to the main and panic logs.
3908 36. The display of retry rules for -brt wasn't always showing the error code
3911 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3912 timeout_connect_A, timeout_connect_MX.
3914 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3915 to the empty sender.
3917 39. The daemon was not analysing the content of -oX till after it had closed
3918 stderr and disconnected from the controlling terminal. This meant that any
3919 syntax errors were only noted on the panic log, and the return code from
3920 the command was 0. By re-arranging the code a little, I've made the
3921 decoding happen first, so such errors now appear on stderr, and the return
3922 code is 1. However, the actual setting up of the sockets still happens in
3923 the disconnected process, so errors there are still only recorded on the
3926 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3927 connections (as happens on some IP stacks) was logged at start up time as
3928 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3929 it from "IPv6 and IPv4", which means that two separate sockets are being
3932 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3933 says whether AF_INET or AF_INET6 was passed as an argument.
3935 42. Exiwhat output was messed up when time zones were included in log
3938 43. Exiwhat now gives more information about the daemon's listening ports,
3939 and whether -tls-on-connect was used.
3941 44. The "port" option of the smtp transport is now expanded.
3943 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3944 silently ignored. Now an error message is written to the main and panic
3947 46. There's a new ACL modifier called "logwrite" which writes to a log file
3948 as soon as it is encountered.
3950 47. Added $local_user_uid and $local_user_gid at routing time.
3952 48. Exim crashed when trying to verify a sender address that was being
3955 49. Exim was recognizing only a space character after ".include". It now also
3956 recognizes a tab character.
3958 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3959 extracting the relevant information from the specification. The man page no
3960 longer contains scrambled data for the -d option, and I've added a section
3961 at the front about calling Exim under different names.
3963 51. Added "extra_headers" argument to the "mail" command in filter files.
3965 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3968 53. Installed eximstats 1.29.
3970 54. Added transport_filter_timeout as a generic transport option.
3972 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3973 Cc: header lines. This was required by RFC 822, but it not required by RFC
3976 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3977 incoming messages that did not have them. Now it does so only if the
3978 message originates locally, that is, if there is no associated remote host
3979 address. When Resent- header lines are present, this applies to the Resent-
3980 lines rather than the non-Resent- lines.
3982 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3983 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3985 58. Messages for configuration errors now include the name of the main
3986 configuration file - useful now that there may be more than one file in a
3987 list (.included file names were always shown).
3989 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3990 for those rare installations that do not start the daemon as root or run it
3991 setuid root. I've cut out the call to initgroups() if the daemon is not
3994 60. The Exim user and group can now be bound into the binary as text strings
3995 that are looked up at the start of Exim's processing.
3997 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3999 62. Added $mailstore_basename variable.
4001 63. Installed patch to sieve.c from Michael Haardt.
4003 64. When Exim failed to open the panic log after failing to open the main log,
4004 the original message it was trying to log was written to stderr and debug
4005 output, but if they were not available (the usual case in production), it
4006 was lost. Now it is written to syslog before the two lines that record the
4007 failures to open the logs.
4009 65. Users' Exim filters run in subprocesses under the user's uid. It is
4010 possible for a "deliver" command or an alias in a "personal" command to
4011 provoke an address rewrite. If logging of address rewriting is configured,
4012 this fails because the process is not running as root or exim. There may be
4013 a better way of dealing with this, but for the moment (because 4.30 needs
4014 to be released), I have disabled address rewrite logging when running a
4015 filter in a non-root, non-exim process.
4021 1. The buildconfig auxiliary program wasn't quoting the value set for
4022 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4023 not defined. This bug was masked in 4.22 by the effect that was fixed in
4026 2. Some messages that were rejected after a message id was allocated were
4027 shown as "incomplete" by exigrep. It no longer does this for messages that
4028 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4030 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4031 have allow_domain_literals set, the ID did not get logged in the <= line.
4032 Domain literals are now always recognized in Message-ID: header lines.
4034 4. The first argument for a ${extract expansion item is the key name or field
4035 number. Leading and trailing spaces in this item were not being ignored,
4036 causing some misleading effects.
4038 5. When deliver_drop_privilege was set, single queue runner processes started
4039 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4040 same command in the process it spins off) were not dropping privilege.
4042 6. When the daemon running as "exim" started a queue runner, it always
4043 re-executed Exim in the spun-off process. This is a waste of effort when
4044 deliver_drop_privilege is set. The new process now just calls the
4045 queue-runner function directly.
4051 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4054 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4055 ignored. Though the use of -oP was forcing the writing of a pid file, it
4056 was always written to the default place.
4058 3. If the message "no IP address found for host xxxx" is generated during
4059 incoming verification, it is now followed by identification of the incoming
4060 connection (so you can more easily find what provoked it).
4062 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4064 5. Added some features to "harden" Exim a bit more against certain attacks:
4066 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4067 be put in Local/Makefile. This is like the never_users runtime option,
4068 but it cannot be overridden. The default setting is "root".
4070 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4071 prefix string with which any file named in a -C command line option
4074 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4075 is retained for -C and -D only if the caller of Exim is root. Without
4076 it, the exim user may also use -C and -D and retain privilege.
4078 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4079 command line option is disabled.
4081 6. Macro names set by the -D option must start with an upper case letter, just
4082 like macro names defined in the configuration file.
4084 7. Added "dereference=" facility to LDAP.
4086 8. Two instances of the typo "uknown" in the source files are fixed.
4088 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4089 the Configure-Makefile script screwed up while processing it.
4091 10. Incorporated PCRE 4.4.
4093 11. The SMTP synchronization check was not operating right at the start of an
4094 SMTP session. For example, it could not catch a HELO sent before the client
4095 waited for the greeting. There is now a check for outstanding input at the
4096 point when the greeting is written. Because of the duplex, asynchronous
4097 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4098 way, but not yet received, when the check is performed.
4100 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4101 on TCP/IP sockets, because this apparently causes some broken clients to
4104 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4105 unchanged) from the Cygwin maintainer.
4107 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4108 was supported instead of testing for mbx. Effectively a typo.
4110 15. The spa authenticator server code was not checking that the input it
4111 received was valid base64.
4113 16. The debug output line for the "set" modifier in ACLs was not showing the
4114 name of the variable that was being set.
4116 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4118 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4119 lookup on its own. Something else was needed to trigger the lookup. For
4120 example, a match in host_lookup or the need for a host name in a host list.
4121 Now, if $sender_host_name is referenced and the host name has not yet been
4122 looked up, a lookup is performed. If the lookup fails, the variable remains
4123 empty, and $host_lookup_failed is set to "1".
4125 19. Added "eqi" as a case-independent comparison operator.
4127 20. The saslauthd authentication condition could segfault if neither service
4128 nor realm was specified.
4130 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4131 error message that was logged was misleading, and incoming SMTP
4132 connections were dropped. The message is now more accurate, and temporary
4133 errors are given to SMTP connections.
4135 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4136 (except RSET) with a 421 error, until QUIT is received. However, it was
4137 failing to send a response to QUIT.
4139 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4140 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4141 ACL is now run after the helo_try_verify_hosts code.
4143 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4144 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4145 case-independent, so other case variants are also recognized). Apparently
4146 some systems use these upper case variants.
4148 25. If more than two messages were waiting for the same host, and a transport
4149 filter was specified for the transport, Exim sent two messages over the
4150 same TCP/IP connection, and then failed with "socket operation on non-
4151 socket" when it tried to send the third.
4153 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4155 27. The extern definition of crypt16() in expand.c was not being excluded when
4156 the OS had its own crypt16() function.
4158 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4159 as a preferred synonym for return_size_limit, both as an option and as an
4162 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4164 30. Changed the default configuration ACL to relax the local part checking rule
4165 for addresses that are not in any local domains. For these addresses,
4166 slashes and pipe symbols are allowed within local parts, but the sequence
4167 /../ is explicitly forbidden.
4169 31. SPA server authentication was not clearing the challenge buffer before
4172 32. log_message in a "warn" ACL statement was writing to the reject log as
4173 well as to the main log, which contradicts the documentation and doesn't
4174 seem right (because no rejection is happening). So I have stopped it.
4176 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4177 However, I am unable to do any testing of this.
4179 34. Fixed an infelicity in the appendfile transport. When checking directories
4180 for a mailbox, to see if any needed to be created, it was accidentally
4181 using path names with one or more superfluous leading slashes; tracing
4182 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4184 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4185 contents are added to the log line that is written for every discarded
4186 recipient. (Previously a log_message setting was ignored.)
4188 36. The ${quote: operator now quotes the string if it is empty.
4190 37. The install script runs exim in order to find its version number. If for
4191 some reason other than non-existence or emptiness, which it checks, it
4192 could not run './exim', it was installing it with an empty version number,
4193 i.e. as "exim-". This error state is now caught, and the installation is
4196 38. An argument was missing from the function that creates an error message
4197 when Exim fails to connect to the socket for saslauthd authentication.
4198 This could cause Exim to crash, or give a corrupted message.
4200 39. Added isip, isip4, and isip6 to ${if conditions.
4202 40. The ACL variables $acl_xx are now saved with the message, and can be
4203 accessed later in routers, transports, and filters.
4205 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4206 strings in the file are not string-expanded.
4208 42. If a MAIL command specified a SIZE value that was too large to fit into an
4209 int variable, the check against message_size_limit failed. Such values are
4210 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4211 day this will have to be increased, but I don't think I want to be around
4212 when emails are that large.
4219 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4220 iconv() is not standard in FreeBSD.
4222 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4223 IPv6 enabled. The observed symptom was a segmentation fault on return from
4224 the function os_common_find_running_interfaces() in src/os.c.
4226 3. In the check_special_case() function in daemon.c I had used "errno" as an
4227 argument name, which causes warnings on some systems. This was basically a
4228 typo, since it was named "eno" in the comments!
4230 4. The code that waits for the clock to tick (at a resolution of some fraction
4231 of a second) so as to ensure message-id uniqueness was always waiting for
4232 at least one whole tick, when it could have waited for less. [This is
4233 almost certainly not relevant at current processor speeds, where it is
4234 unlikely to ever wait at all. But we try to future-proof.]
4236 5. The function that sleeps for a time interval that includes fractions of a
4237 second contained a race. It did not block SIGALRM between setting the
4238 timer, and suspending (a couple of lines later). If the interval was short
4239 and the sigsuspend() was delayed until after it had expired, the suspension
4240 never ended. On busy systems this could lead to processes getting stuck for
4243 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4244 process, before it forks any delivery processes. The open lookup caching
4245 mechanism meant that the open file or database connection was passed into
4246 the delivery process. The problem was that delivery processes always tidy
4247 up cached lookup data. This could cause a problem for the next delivery
4248 process started by the queue runner, because the external queue runner
4249 process does not know about the closure. So the next delivery process
4250 still has data in the lookup cache. In the case of a file lookup, there was
4251 no problem because closing a file descriptor in a subprocess doesn't affect
4252 the parent. However, if the lookup was caching a connection to a database,
4253 the connection was closed, and the second delivery process was likely to
4254 see errors such as "PGSQL: query failed: server closed the connection
4255 unexpectedly". The problem has been fixed by closing all cached lookups
4256 in a queue runner before running a delivery process.
4258 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4259 seem to have the "const" qualifier which it has on other OS. I've
4262 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4263 *of the same type* (client or server) with the same public name that an
4264 error should be diagnosed.
4266 9. When Exim looked up a host name for an IP address, but failed to find the
4267 original IP address when looking up the host name (a safety check), it
4268 output the message "<ip address> does not match any IP for NULL", which was
4269 confusing, to say the least. The bug was that the host name should have
4270 appeared instead of "NULL".
4272 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4273 user that is built into the binary, and the -C or -D options is used, root
4274 privilege is dropped before the configuration file is read. In addition,
4275 logging is switched to stderr instead of the normal log files. If the
4276 configuration then re-defines the Exim user, the unprivileged environment
4277 is probably not what is expected, so Exim logs a panic warning message (but
4280 However, if deliver_drop_privilege is set, the unprivileged state may well
4281 be exactly what is intended, so the warning has been cut out in that case,
4282 and Exim is allowed to try to write to its normal log files.
4288 1. smtp_return_error_details was not giving details for temporary sender
4289 or receiver verification errors.
4291 2. Diagnose a configuration error if two authenticators have the same public
4294 3. Exim used not to create the message log file for a message until the first
4295 delivery attempt. This could be confusing when incoming messages were held
4296 for policy or load reasons. The message log file is now created at the time
4297 the message is received, and an initial "Received" line is written to it.
4299 4. The automatically generated man page for command line options had a minor
4300 bug that caused no ill effects; however, a more serious problem was that
4301 the procedure for building the man page automatically didn't always
4302 operate. Consequently, release 4.20 contains an out-of-date version. This
4303 shouldn't happen again.
4305 5. When building Exim with embedded Perl support, the script that builds the
4306 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4307 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4309 6. The freeze_tell option was not being used for messages that were frozen on
4310 arrival, either by an ACL or by local_scan().
4312 7. Added the smtp_incomplete_transaction log selector.
4314 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4315 was accepting AUTH without a new EHLO.
4317 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4318 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4321 10. Logging of TCP/IP connections (when configured) now happens in the main
4322 daemon process instead of the child process, so that the TCP/IP connection
4323 count is more accurate (but it can never be perfect).
4325 11. The use of "drop" in a nested ACL was not being handled correctly in the
4326 outer ACL. Now, if condition failure induced by the nested "drop" causes
4327 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4328 or "require"), the connection is dropped.
4330 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4331 that yield "discard" can now be used with an "accept" or a "discard" verb,
4332 but an error is generated for any others (because I can't see a useful way
4333 to define what should happen).
4335 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4336 are now processed in the same way as lines in the Exim configuration file.
4337 In particular, continuation lines are supported.
4339 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4341 15. Added -ti meaning -t -i.
4343 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4344 domains, and warn by logging if others are found.
4346 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4347 returned from GIFCONF when Exim is trying to find the list of interfaces on
4348 a host. The code in os.c has been modified to copy each ifreq to an aligned
4349 structure in all cases.
4351 Also, in some cases, the returned ifreq's were being copied to a 'struct
4352 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4353 means the last couple of bytes of an IPv6 address could be chopped if the
4354 ifreq contained only a normal sockaddr (14 bytes storage).
4356 18. Named domain lists were not supported in the hosts_treat_as_local option.
4357 An entry such as +xxxx was not recognized, and was treated as a literal
4360 19. Ensure that header lines added by a DATA ACL are included in the reject log
4361 if the ACL subsequently rejects the message.
4363 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4364 MD5 (which is deprecated).
4366 21. When testing a filter file using -bf, Exim was writing a message when it
4367 took the sender from a "From " line in the message, but it was not doing so
4368 when it took $return_path from a Return-Path: header line. It now does.
4370 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4371 with a valid header line field name (a series of printing characters
4372 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4374 23. Changed "disc" in the source to "disk" to conform to the documentation and
4375 the book and for uniformity.
4377 24. Ignore Sendmail's -Ooption=value command line item.
4379 25. When execve() failed while trying to run a command in a pipe transport,
4380 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4381 could be confused with a return value of 69 from the command itself. This
4382 has been changed to 127, the value the shell returns if it is asked to run
4383 a non-existent command. The wording for the related log line suggests a
4384 non-existent command as the problem.
4386 26. If received_header_text expands to an empty string, do not add a Received:
4387 header line to the message. (Well, it adds a token one on the spool, but
4388 marks it "old" so that it doesn't get used or transmitted.)
4390 27. Installed eximstats 1.28 (addition of -nt option).
4392 28. There was no check for failure on the call to getsockname() in the daemon
4393 code. This can fail if there is a shortage of resources on the system, with
4394 ENOMEM, for example. A temporary error is now given on failure.
4396 29. Contrary to the C standard, it seems that in some environments, the
4397 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4398 program. Exim now does this explicitly; it affects the formatting of
4399 timestamps using strftime().
4401 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4402 complaints. I've now initialized all the variables, to avoid this.
4404 32. Header lines added by a system filter were not being "seen" during
4405 transport-time rewrites.
4407 33. The info_callback() function passed to OpenSSL is set up with type void
4408 (*)(SSL *, int, int), as described somewhere. However, when calling the
4409 function (actually a macro) that sets it up, the type void(*)() is
4410 expected. I've put in a cast to prevent warnings from picky compilers.
4412 34. If a DNS black list lookup found a CNAME record, but there were no A
4413 records associated with the domain it pointed at, Exim crashed.
4415 35. If a DNS black list lookup returned more than one A record, Exim ignored
4416 all but the first. It now scans all returned addresses if a particular IP
4417 value is being sought. In this situation, the contents of the
4418 $dnslist_value variable are a list of all the addresses, separated by a
4421 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4422 to accept a host name and all its aliases if the forward lookup for any of
4423 them yielded the IP address of the incoming connection. Now it accepts only
4424 those names whose forward lookup yields the correct IP address. Any other
4425 names are discarded. This closes a loophole whereby a rogue DNS
4426 administrator could create reverse DNS records to break through a
4427 wildcarded host restriction in an ACL.
4429 37. If a user filter or a system filter that ran in a subprocess used any of
4430 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4431 the wrong values were passed to the pipe command ($thisaddress had the
4432 value of $0, $0 had the value of $1, etc). This bug was introduced by
4433 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4435 38. Improved the line breaking for long SMTP error messages from ACLs.
4436 Previously, if there was no break point between 40 and 75 characters, Exim
4437 left the rest of the message alone. Two changes have been made: (a) I've
4438 reduced the minimum length to 35 characters; (b) if it can't find a break
4439 point between 35 and 75 characters, it looks ahead and uses the first one
4440 that it finds. This may give the occasional overlong line, but at least the
4441 remaining text gets split now.
4443 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4444 file descriptors might be low, and that setting 1000 would always raise it.
4445 It turns out that in some environments, the limit is already over 1000 and
4446 that lowering it causes trouble. So now Exim takes care not to decrease it.
4448 40. When delivering a message, the value of $return_path is set to $sender_
4449 address at the start of routing (routers may change the value). By an
4450 oversight, this default was not being set up when an address was tested by
4451 -bt or -bv, which affected the outcome if any router or filter referred to
4454 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4455 log only when "message" or "log_message" are set. However, if one of the
4456 conditions was an address verification, or a call to a nested ACL, the
4457 messages generated by the underlying test were being passed through. This
4458 no longer happens. The underlying message is available in $acl_verify_
4459 message for both "message" and "log_message" expansions, so it can be
4460 passed through if needed.
4462 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4463 new expansion $bh_ to give the encoded byte string without charset
4464 translation. Translation happens only if iconv() is available; HAVE_ICONV
4465 indicates this at build time. HEADERS_CHARSET gives the charset to
4466 translate to; headers_charset can change it in the configuration, and
4467 "headers charset" can change it in an individual filter file.
4469 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4470 that creates RFC 2047 encoded "words" labels them as that charset instead
4471 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4472 expansion operator; (ii) when Exim creates a From: line for a local
4473 message; (iii) when a header line is rewritten to include a "phrase" part.
4475 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4476 buggy, causing it to skip the first lines of messages whose message ID
4477 ended in 'D'. This would not have bitten before Exim release 4.14, because
4478 message IDs were unlikely to end in 'D' before then. The effect was to have
4479 incorrect size information for certain domains.
4481 45. #include "config.h" was missing at the start of the crypt16.c module. This
4482 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4485 46. If there was a timeout during a "random" callout check, Exim treated it as
4486 a failure of the random address, and carried on sending RSET and the real
4487 address. If the delay was just some slowness somewhere, the response to the
4488 original RCPT would be taken as a response to RSET and so on, causing
4489 mayhem of various kinds.
4491 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4492 when I implemented it. It didn't allow for the fact that some option values
4493 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4494 the right test for positive values.
4496 48. Domain names in DNS records are case-independent. Exim always looks them up
4497 in lower case. Some resolvers return domain names in exactly the case they
4498 appear in the zone file, that is, they may contain uppercase letters. Not
4499 all resolvers do this - some return always lower case. Exim was treating a
4500 change of case by a resolver as a change of domain, similar to a widening
4501 of a domain abbreviation. This triggered its re-routing code and so it was
4502 trying to route what was effectively the same domain again. This normally
4503 caused routing to fail (because the router wouldn't handle the domain
4504 twice). Now Exim checks for this case specially, and just changes the
4505 casing of the domain that it ultimately uses when it transmits the message
4508 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4511 50. If a filter generated a file delivery with a non-absolute name (possible if
4512 no home directory exists for the router), the forbid_file option was not
4515 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4516 the existing equality matching.
4518 52. Exim was using ints instead of ino_t variables in some places where it was
4519 dealing with inode numbers.
4521 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4522 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4523 called TMPDIR, and if it finds it is different, it changes its value.
4525 54. The smtp_printf() function is now made available to local_scan() so
4526 additional output lines can be written before returning. There is also an
4527 smtp_fflush() function to enable the detection of a dropped connection.
4528 The variables smtp_input and smtp_batched_input are exported to
4531 55. Changed the default runtime configuration: the message "Unknown user"
4532 has been removed from the ACL, and instead placed on the localuser router,
4533 using the cannot_route_message feature. This means that any verification
4534 failures that generate their own messages won't get overridden. Similarly,
4535 the "Unrouteable address" message that was in the ACL for unverifiable
4536 relay addresses has also been removed.
4538 56. Added hosts_avoid_esmtp to the smtp transport.
4540 57. The exicyclog script was not checking for the esoteric option
4541 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4542 will work only if exicyclog is run under the appropriate euid.
4544 58. Following a discussion on the list, the rules by which Exim recognises line
4545 endings on incoming messages have been changed. The -dropcr and drop_cr
4546 options are now no-ops, retained only for backwards compatibility. The
4547 following line terminators are recognized: LF CRLF CR. However, special
4548 processing applies to CR:
4550 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4551 nor a local message in the state where . is a terminator.
4553 (ii) If a bare CR is encountered in a header line, an extra space is added
4554 after the line terminator so as not to end the header. The reasoning
4555 behind this is that bare CRs in header lines are most likely either
4556 to be mistakes, or people trying to play silly games.
4558 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4559 was being incorrectly given as 18 bytes larger than it should have been.
4560 This is a VOB (very old bug).
4562 60. This may never have affected anything current, but just in case it has:
4563 When the local host is found other than at the start of a list of hosts,
4564 the local host, those with the same MX, and any that follow, are discarded.
4565 When the list in question was part of a longer list of hosts, the following
4566 hosts (not currently being processed) were also being discarded. This no
4567 longer happens. I'm not sure if this situation could ever has previously
4570 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4573 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4574 header. This is something that is recommended in a new Internet Draft, and
4575 is something that is documented as being done by Sendmail. There are two
4576 possible values. For messages generated by the autoreply transport, Exim
4579 Auto-Submitted: auto-replied
4581 whereas for all other generated messages (e.g. bounces) it adds
4583 Auto-Submitted: auto-generated
4585 63. The "personal" condition in filters now includes a test for the
4586 Auto-Submitted: header. If it contains the string "auto-" the message it
4587 not considered personal.
4589 64. Added rcpt_include_affixes as a generic transport option.
4591 65. Added queue_only_override (default true).
4593 66. Added the syslog_duplication option.
4595 67. If what should have been the first header line of a message consisted of
4596 a space followed by a colon, Exim was mis-interpreting it as a header line.
4597 It isn't of course - it is syntactically invalid and should therefore be
4598 treated as the start of the message body. The misbehaviour could have
4599 caused a number of strange effects, including loss of data in subsequent
4600 header lines, and spool format errors.
4602 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4603 client host had authenticated. This control can now be exercised by an ACL
4604 for more flexibility.
4606 69. By default, callouts do not happen when testing with -bh. There is now a
4607 variant, -bhc, which does actually run the callout code, including
4608 consulting and updating the callout cache.
4610 70. Added support for saslauthd authentication, courtesy of Alexander
4613 71. If statvfs() failed on the spool or log directories while checking their
4614 size for availability, Exim confusingly gave the error "space shortage".
4615 Furthermore, in debugging mode it crashed with a floating point exception.
4616 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4617 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4618 serious problem, Exim now writes to the main and panic logs when this
4619 happens, with details of the failure. It then refuses to accept the
4620 incoming message, giving the message "spool directory problem" or "log
4621 directory problem" with a 421 code for SMTP messages.
4623 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4624 0, 1, and 2 exist, because some OS complain for execs without them (see
4625 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4626 descriptors. However, the code omitted to check that the open succeeded,
4627 causing mysterious errors if for some reason the permissions on /dev/null
4628 got screwed. Now Exim writes a message to the main and panic logs, and
4629 bombs out if it can't open /dev/null.
4631 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4632 interact so that it is all more flexible. It is supposed to remain
4633 backwards compatible. Also added extra_local_interfaces.
4635 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4636 to bomb out with an assertion failure - to the client this appears as a
4637 connection drop. This problem occurs in the part of the code that was taken
4638 from the Samba project. Fortunately, the assertion is in a very simple
4639 function, so I have fixed this by reproducing the function inline in the
4640 one place where it is called, and arranging for authentication to fail
4641 instead of killing the process with assert().
4643 75. The SPA client code was not working when the server requested OEM rather
4644 than Unicode encoding.
4646 76. Added code to make require_files with a specific uid setting more usable in
4647 the case where statting the file as root fails - usually a non-root-mounted
4648 NFS file system. When this happens and the failure is EACCES, Exim now
4649 forks a subprocess and does the per-uid checking as the relevant uid.
4651 77. Added process_log_path.
4653 78. If log_file_path was not explicitly set, a setting of check_log_space or
4654 check_log_inodes was ignored.
4656 79. If a space check for the spool or log partitions fails, the incident is now
4657 logged. Of course, in the latter case the data may get lost...
4659 80. Added the %p formatting code to string_format() so that it can be used to
4660 print addresses in debug_print(). Adjusted all the address printing in the
4661 debugging in store.c to use %p rather than %d.
4663 81. There was a concern that a line of code in smtp_in.c could overflow a
4664 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4665 initially expressed, the concern was not well-founded, because trailing
4666 spaces are removed early. However, if the trailing spaces were followed by
4667 a NULL, they did not get removed, so the overflow was possible. Two fixes
4670 (a) I re-wrote the offending code in a cleaner fashion.
4671 (b) If an incoming SMTP command contains a NULL character, it is rejected
4674 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4675 runs initgroups(), so that if the Exim user is in any additional groups,
4676 they will be used during message reception.
4682 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob