1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.522 2007/08/17 11:16:45 ph10 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 PH/01 Another patch from the Sieve maintainer.
11 PH/02 When an IPv6 address is converted to a string for single-key lookup
12 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
13 dots are used instead of colons so that keys in lsearch files need not
14 contain colons. This was done some time before quoting was made available
15 in lsearch files. However, iplsearch files do require colons in IPv6 keys
16 (notated using the quote facility) so as to distinguish them from IPv4
17 keys. This meant that lookups for IP addresses in host lists did not work
18 for iplsearch lookups.
20 This has been fixed by arranging for IPv6 addresses to be expressed with
21 colons if the lookup type is iplsearch. This is not incompatible, because
22 previously such lookups could never work.
24 The situation is now rather anomolous, since one *can* have colons in
25 ordinary lsearch keys. However, making the change in all cases is
26 incompatible and would probably break a number of configurations.
28 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
31 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
32 conversion specification without a maximum field width, thereby enabling
33 a rogue spamd server to cause a buffer overflow. While nobody in their
34 right mind would setup Exim to query an untrusted spamd server, an
35 attacker that gains access to a server running spamd could potentially
36 exploit this vulnerability to run arbitrary code as the Exim user.
38 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
39 $primary_hostname instead of what libspf2 thinks the hosts name is.
41 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
42 a directory entry by the name of the lookup key. Previously, if a
43 symlink pointed to a non-existing file or a file in a directory that
44 Exim lacked permissions to read, a lookup for a key matching that
45 symlink would fail. Now it is enough that a matching directory entry
46 exists, symlink or not. (Bugzilla 503.)
48 PH/03 The body_linecount and body_zerocount variables are now exported in the
51 PH/04 Added the $dnslist_matched variable.
53 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
54 This means they are set thereafter only if the connection becomes
57 PH/06 Added the client_condition to authenticators so that some can be skipped
58 by clients under certain conditions.
60 PH/07 The error message for a badly-placed control=no_multiline_responses left
61 "_responses" off the end of the name.
63 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
65 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
66 (without spaces) instead of just copying the configuration text.
68 PH/10 Added the /noupdate option to the ratelimit ACL condition.
70 PH/11 Added $max_received_linelength.
72 PH/12 Added +ignore_defer and +include_defer to host lists.
74 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
75 way in which PCRE > 7.0 is built.
77 PH/14 Implemented queue_only_load_latch.
79 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
80 MAIL command. The effect was to mangle the value on 64-bit systems.
82 PH/16 Another patch from the Sieve maintainer.
89 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
90 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
93 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
96 PH/02 In an ACL statement such as
98 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
100 if a client was not listed at all, or was listed with a value other than
101 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
102 the condition was not true (as it should be), so access was not denied.
103 The bug was that the ! inversion was incorrectly passed on to the second
104 item. This has been fixed.
106 PH/03 Added additional dnslists conditions == and =& which are different from
107 = and & when the dns lookup returns more than one IP address.
109 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
110 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
112 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
113 FSYNC, which compiles an option called disable_fsync that allows for
114 bypassing fsync(). The documentation is heavily laced with warnings.
116 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
118 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
119 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
120 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
121 including adding "make clean"; (3) Added -fPIC when compiling the test
122 dynamically loaded module, to get rid of a warning.
124 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
125 message fails, move_frozen_messages = true and ignore_bounce_errors_after
126 = 0s. The bug is otherwise harmless.
128 PH/07 There was a bug in the dovecot authenticator such that the value of
129 $auth1 could be overwritten, and so not correctly preserved, after a
130 successful authentication. This usually meant that the value preserved by
131 the server_setid option was incorrect.
133 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
135 PH/09 Installed PCRE release 7.0.
137 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
138 run for batched SMTP input. It is now run at the start of every message
139 in the batch. While fixing this I discovered that the process information
140 (output by running exiwhat) was not always getting set for -bs and -bS
141 input. This is fixed, and it now also says "batched" for BSMTP.
143 PH/11 Added control=no_pipelining.
145 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
146 patch, slightly modified), and move the expansion of helo_data till after
147 the connection is made in the smtp transport (so it can use these
150 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
152 PH/14 Added log_selector = +pid.
154 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
156 PH/16 Add ${if forany and ${if forall.
158 PH/17 Added dsn_from option to vary the From: line in DSNs.
160 PH/18 Flush SMTP output before performing a callout, unless control =
161 no_callout_flush is set.
163 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
164 was true (the default) a successful delivery failed to delete the retry
165 item, thus causing premature timeout of the address. The bug is now
168 PH/20 Added hosts_avoid_pipelining to the smtp transport.
170 PH/21 Long custom messages for fakedefer and fakereject are now split up
171 into multiline reponses in the same way that messages for "deny" and
172 other ACL rejections are.
174 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
175 with slight modification.
177 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
178 draft, changing the syntax and factoring some duplicate code".
180 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
181 for deliveries of the second and subsequent messages over the same SMTP
184 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
185 ${reduce, with only minor "tidies".
187 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
189 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
190 expansion side effects.
192 PH/27 When a message times out after an over-quota error from an Exim-imposed
193 quota, the bounce message says "mailbox is full". This message was not
194 being given when it was a system quota that was exceeded. It now should
197 MH/03 Made $recipients available in local_scan(). local_scan() already has
198 better access to the recipient list through recipients_list[], but
199 $recipients can be useful in postmaster-provided expansion strings.
201 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
202 in the case of a MAIL command with additional options following the
203 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
204 were accidentally chopped off.
206 PH/29 SMTP synchronization checks are implemented when a command is read -
207 there is a check that no more input is waiting when there shouldn't be
208 any. However, for some commands, a delay in an ACL can mean that it is
209 some time before the response is written. In this time, more input might
210 arrive, invalidly. So now there are extra checks after an ACL has run for
211 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
212 pipelining has not been advertised.
214 PH/30 MH's patch to allow iscntrl() characters to be list separators.
216 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
217 returned in the SMTP response when smtp_return_error_details was false.
220 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
221 instead of the C I/O that was originally supplied, because problems were
224 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
225 Exim which did not show up earlier: it was assuming that a call to
226 SSL_CTX_set_info_callback() might give an error value. In fact, there is
227 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
228 was a macro that became an assignment, so it seemed to work. This has
229 changed to a proper function call with a void return, hence the compile
230 error. Exim's code has been fixed.
232 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
235 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
237 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
239 PH/37 If a message is not accepted after it has had an id assigned (e.g.
240 because it turns out to be too big or there is a timeout) there is no
241 "Completed" line in the log. When some messages of this type were
242 selected by exigrep, they were listed as "not completed". Others were
243 picked up by some special patterns. I have improved the selection
244 criteria to be more general.
246 PH/38 The host_find_failed option in the manualroute router can now be set
247 to "ignore", to completely ignore a host whose IP address cannot be
248 found. If all hosts are ignored, the behaviour is controlled by the new
249 host_all_ignored option.
251 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
252 homing or because of multiple MX records with /mx) generated more than
253 one IP address, and the following item turned out to be the local host,
254 all the secondary addresses of the first item were incorrectly removed
255 from the list, along with the local host and any following hosts (which
256 is what is supposed to happen).
258 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
259 whoever called Exim into the -H file. In the case of the daemon it was
260 behaving confusingly. When first started, it used values for whoever
261 started the daemon, but after a SIGHUP it used the Exim user (because it
262 calls itself on a restart). I have changed the code so that it now always
265 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
266 message are rejected with the same error (e.g. no authentication or bad
267 sender address), and a DATA command is nevertheless sent (as can happen
268 with PIPELINING or a stupid MUA), the error message that was given to the
269 RCPT commands is included in the rejection of the DATA command. This is
270 intended to be helpful for MUAs that show only the final error to their
273 PH/42 Another patch from the Sieve maintainer.
275 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
276 Eximstats - Fixed some broken HTML links and added missing column headers
278 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
279 columns for Rejects, Temp Rejects, Ham, and Spam rows.
281 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
283 PH/43 Yet another patch from the Sieve maintainer.
285 PH/44 I found a way to check for a TCP/IP connection going away before sending
286 the response to the final '.' that terminates a message, but only in the
287 case where the client has not sent further data following the '.'
288 (unfortunately, this is allowed). However, in many cases there won't be
289 any further data because there won't be any more messages to send. A call
290 to select() can be used: if it shows that the input is "ready", there is
291 either input waiting, or the socket has been closed. An attempt to read
292 the next input character can distinguish the two cases. Previously, Exim
293 would have sent an OK response which the client would never have see.
294 This could lead to message repetition. This fix should cure that, at
295 least in a lot of common cases.
297 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
298 advertised in response to EHLO.
304 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
305 fixed by 4.65/MH/01 (is this a record?) are fixed:
307 (i) An empty string was always treated as zero by the numeric comparison
308 operators. This behaviour has been restored.
310 (ii) It is documented that the numeric comparison operators always treat
311 their arguments as decimal numbers. This was broken in that numbers
312 starting with 0 were being interpreted as octal.
314 While fixing these problems I realized that there was another issue that
315 hadn't been noticed. Values of message_size_limit (both the global option
316 and the transport option) were treated as octal if they started with 0.
317 The documentation was vague. These values are now always treated as
318 decimal, and I will make that clear in the documentation.
324 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
325 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
328 MH/01 Don't check that the operands of numeric comparison operators are
329 integers when their expansion is in "skipping" mode (fixes bug
330 introduced by 4.64-PH/07).
332 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
333 child addresses, Exim now panics and dies. Previously, because the count
334 is held in a short int, deliveries were likely to be lost. As such a
335 large number of recipients for a single message is ridiculous
336 (performance will be very, very poor), I have chosen to impose a limit
337 rather than extend the field.
343 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
344 leftover -K file (the existence of which was triggered by #402).
345 While we were at it, introduced process PID as part of the -K
346 filename. This should rule out race conditions when creating
349 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
350 processing considerably. Previous code took too long for large mails,
351 triggering a timeout which in turn triggers #401.
353 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
354 in the DK code in transports.c. sendfile() is not really portable,
355 hence the _LINUX specificness.
357 TF/01 In the add_headers option to the mail command in an Exim filter,
358 there was a bug that Exim would claim a syntax error in any
359 header after the first one which had an odd number of characters
362 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
363 callout verification, Exim cached a "reject" for the entire domain. This
364 is correct for most verifications, but it is not correct for a recipient
365 verification with use_sender or use_postmaster set, because in that case
366 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
367 case of MAIL FROM:<> rejection from other early rejections (e.g.
368 rejection of HELO). When verifying a recipient using a non-null MAIL
369 address, the cache is ignored if it shows MAIL FROM:<> rejection.
370 Whatever the result of the callout, the value of the domain cache is
371 left unchanged (for any other kind of callout, getting as far as trying
372 RCPT means that the domain itself is ok).
374 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
377 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
378 manifest itself as EPIPE rather than ECONNECT. When tidying away a
379 session, the daemon ignores ECONNECT errors and logs others; it now
380 ignores EPIPE as well.
382 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
383 (quoted-printable decoding).
385 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
386 later the small subsequent patch to fix an introduced bug.
388 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
390 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
392 PH/08 An error is now given if message_size_limit is specified negative.
394 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
395 to be given (somewhat) arbitrary names.
397 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
400 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
401 miscellaneous code fixes
403 PH/10 Added the log_reject_target ACL modifier to specify where to log
406 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
407 hostname. This is wrong, because it relates to the incoming message (and
408 probably the interface on which it is arriving) and not to the outgoing
409 callout (which could be using a different interface). This has been
410 changed to use the value of the helo_data option from the smtp transport
411 instead - this is what is used when a message is actually being sent. If
412 there is no remote transport (possible with a router that sets up host
413 addresses), $smtp_active_hostname is used.
415 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
416 tweaks were necessary in order to get it to work (see also 21 below):
417 (a) The code assumed that strncpy() returns a negative number on buffer
418 overflow, which isn't the case. Replaced with Exim's string_format()
420 (b) There were several signed/unsigned issues. I just did the minimum
421 hacking in of casts. There is scope for a larger refactoring.
422 (c) The code used strcasecmp() which is not a standard C function.
423 Replaced with Exim's strcmpic() function.
424 (d) The code set only $1; it now sets $auth1 as well.
425 (e) A simple test gave the error "authentication client didn't specify
426 service in request". It would seem that Dovecot has changed its
427 interface. Fortunately there's a specification; I followed it and
428 changed what the client sends and it appears to be working now.
430 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
433 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
434 address A is aliased to B and C, where B exists and C does not. Without
435 -v the output is "A verified" because verification stops after a
436 successful redirection if more than one address is generated. However,
437 with -v the child addresses are also verified. Exim was outputting "A
438 failed to verify" and then showing the successful verification for C,
439 with its parentage. It now outputs "B failed to verify", showing B's
440 parentage before showing the successful verification of C.
442 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
443 look up a TXT record in a specific list after matching in a combined
446 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
447 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
448 they consult the DNS. I had assumed they would set it the way they
449 wanted; and indeed my experiments on Linux seem to show that in some
450 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
451 To be on the safe side, however, I have now made the interface to
452 host_find_byname() similar to host_find_bydns(), with an argument
453 containing the DNS resolver options. The host_find_byname() function now
454 sets these options at its start, just as host_find_bydns() does. The smtp
455 transport options dns_qualify_single and dns_search_parents are passed to
456 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
457 of host_find_byname() use the default settings of RES_DEFNAMES
458 (qualify_single) but not RES_DNSRCH (search_parents).
460 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
461 spool_read_header() do less string testing, by means of a preliminary
462 switch on the second character of optional "-foo" lines. (This is
463 overdue, caused by the large number of possibilities that now exist.
464 Originally there were few.) While I was there, I also converted the
465 str(n)cmp tests so they don't re-test the leading "-" and the first
466 character, in the hope this might squeeze out yet more improvement.
468 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
469 flag allowing group syntax was set by the header_syntax check but not
470 turned off, possible causing trouble later; (2) The flag was not being
471 set at all for the header_verify test, causing "group"-style headers to
472 be rejected. I have now set it in this case, and also caused header_
473 verify to ignore an empty address taken from a group. While doing this, I
474 came across some other cases where the code for allowing group syntax
475 while scanning a header line wasn't quite right (mostly, not resetting
476 the flag correctly in the right place). These bugs could have caused
477 trouble for malformed header lines. I hope it is now all correct.
479 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
480 with the "reply" argument non-NULL. The code, however (which originally
481 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
482 but it didn't always do it. This confused somebody who was copying the
483 code for some other use. I have removed all the tests.
485 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
486 feature that was used to support insecure browsers during the U.S. crypto
487 embargo. It requires special client support, and Exim is probably the
488 only MTA that supported it -- and would never use it because real RSA is
489 always available. This code has been removed, because it had the bad
490 effect of slowing Exim down by computing (never used) parameters for the
491 RSA_EXPORT functionality.
493 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
494 authenticator to fail if there's a tab character in the incoming data
495 (there should never be unless someone is messing about, as it's supposed
496 to be base64-encoded). Also added, on Timo's advice, the "secured" option
497 if the connection is using TLS or if the remote IP is the same as the
498 local IP, and the "valid-client-cert option" if a client certificate has
501 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
502 authenticators. This can be used for authorization after authentication
503 succeeds. (In the case of plaintext, it servers for both authentication
506 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
507 if any retry times were supplied.
509 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
510 connection, where there is no client IP address to check. In this
511 situation, the verify now always succeeds.
513 PH/25 Applied John Jetmore's -Mset patch.
515 PH/26 Added -bem to be like -Mset, but loading a message from a file.
517 PH/27 In a string expansion for a processed (not raw) header when multiple
518 headers of the same name were present, leading whitespace was being
519 removed from all of them, but trailing whitespace was being removed only
520 from the last one. Now trailing whitespace is removed from each header
521 before concatenation. Completely empty headers in a concatenation (as
524 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
525 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
527 PH/29 [Removed. This was a change that I later backed out, and forgot to
528 correct the ChangeLog entry (that I had efficiently created) before
529 committing the later change.]
531 PH/30 Exim was sometimes attempting to deliver messages that had suffered
532 address errors (4xx response to RCPT) over the same connection as other
533 messages routed to the same hosts. Such deliveries are always "forced",
534 so retry times are not inspected. This resulted in far too many retries
535 for the affected addresses. The effect occurred only when there were more
536 hosts than the hosts_max_try setting in the smtp transport when it had
537 the 4xx errors. Those hosts that it had tried were not added to the list
538 of hosts for which the message was waiting, so if all were tried, there
539 was no problem. Two fixes have been applied:
541 (i) If there are any address or message errors in an SMTP delivery, none
542 of the hosts (tried or untried) are now added to the list of hosts
543 for which the message is waiting, so the message should not be a
544 candidate for sending over the same connection that was used for a
545 successful delivery of some other message. This seems entirely
546 reasonable: after all the message is NOT "waiting for some host".
547 This is so "obvious" that I'm not sure why it wasn't done
548 previously. Hope I haven't missed anything, but it can't do any
549 harm, as the worst effect is to miss an optimization.
551 (ii) If, despite (i), such a delivery is accidentally attempted, the
552 routing retry time is respected, so at least it doesn't keep
553 hammering the server.
555 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
556 in ${readsocket because some servers need this prod.
558 PH/32 Added some extra debug output when updating a wait-xxx database.
560 PH/33 The hint "could be header name not terminated by colon", which has been
561 given for certain expansion errors for a long time, was not being given
562 for the ${if def:h_colon_omitted{... case.
564 PH/34 The spec says: "With one important exception, whenever a domain list is
565 being scanned, $domain contains the subject domain." There was at least
566 one case where this was not true.
568 PH/35 The error "getsockname() failed: connection reset by peer" was being
569 written to the panic log as well as the main log, but it isn't really
570 panic-worthy as it just means the connection died rather early on. I have
571 removed the panic log writing for the ECONNRESET error when getsockname()
574 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
575 runs only) independently of the message's sender address. This meant
576 that, if the 4xx error was in fact related to the sender, a different
577 message to the same recipient with a different sender could confuse
578 things. In particualar, this can happen when sending to a greylisting
579 server, but other circumstances could also provoke similar problems.
580 I have changed the default so that the retry time for these errors is now
581 based a combination of the sender and recipient addresses. This change
582 can be overridden by setting address_retry_include_sender=false in the
585 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
586 remote server are returned as part of bounce messages. This was not
587 happening for LMTP over a pipe (the lmtp transport), but now it is the
588 same for both kinds of LMTP.
590 PH/38 Despite being documented as not happening, Exim was rewriting addresses
591 in header lines that were in fact CNAMEs. This is no longer the case.
593 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
594 and queue runs started by the daemon processed all messages. This has
595 been fixed so that -R and -S can now usefully be given with -q<time>.
597 PH/40 Import PCRE release 6.7 (fixes some bugs).
599 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
601 PH/42 Give an error if -q is specified more than once.
603 PH/43 Renamed the variables $interface_address and $interface_port as
604 $received_ip_address and $received_port, to make it clear that these
605 values apply to message reception, and not to the outgoing interface when
606 a message is delivered. (The old names remain recognized, of course.)
608 PH/44 There was no timeout on the connect() call when using a Unix domain
609 socket in the ${readsocket expansion. There now is.
611 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
612 be meaningful with "accept".
614 SC/01 Eximstats V1.43
615 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
617 SC/02 Eximstats V1.44
618 Use a glob alias rather than an array ref in the generated
619 parser. This improves both readability and performance.
621 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
622 Collect SpamAssassin and rejection statistics.
623 Don't display local sender or destination tables unless
624 there is data to show.
625 Added average volumes into the top table text output.
627 SC/04 Eximstats V1.46
628 Collect data on the number of addresses (recipients)
629 as well as the number of messages.
631 SC/05 Eximstats V1.47
632 Added 'Message too big' to the list of mail rejection
633 reasons (thanks to Marco Gaiarin).
635 SC/06 Eximstats V1.48
636 Mainlog lines which have GMT offsets and are too short to
637 have a flag are now skipped.
639 SC/07 Eximstats V1.49 (Alain Williams)
640 Added the -emptyok flag.
642 SC/08 Eximstats V1.50
643 Fixes for obtaining the IP address from reject messages.
645 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
646 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
647 whitesspace changes from 4.64-PH/27
649 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
652 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
653 are found, allow negative numbers in numeric criteria)
655 JJ/06 exipick.20061117.2, added new $message_body_missing variable
657 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
658 to match changes made in 4.64-PH/43
660 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
662 PH/47 Put in an explicit test for a DNS lookup of an address record where the
663 "domain" is actually an IP address, and force a failure. This locks out
664 those revolvers/nameservers that support "A-for-A" lookups, in
665 contravention of the specifications.
667 PH/48 When a host name was looked up from an IP address, and the subsequent
668 forward lookup of the name timed out, the host name was left in
669 $sender_host_name, contrary to the specification.
671 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
672 restricted to single-key lookups, Exim was not diagnosing an error if
673 * or *@ was used with a query-style lookup.
675 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
677 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
678 long ago, but noone interested enough thought of it. Let's just say that
679 the "1.1" means that there are some new functions that weren't there at
680 some point in the past.
682 PH/51 Error processing for expansion failure of helo_data from an smtp
683 transport during callout processing was broken.
685 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
686 tested/used via the -bh/-bhc/-bs options.
688 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
689 bug, fixed in subsequent PCRE releases).
691 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
692 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
694 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
700 SC/01 Use a glob alias rather than an array ref in eximstats generated
701 parser. This improves both readability and performance.
703 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
704 Don't display local sender or destination tables in eximstats unless
705 there is data to show.
706 Added average volumes into the eximstats top table text output.
708 SC/03 Collect data on the number of addresses (recipients) as well
709 as the number of messages in eximstats.
711 TF/01 Correct an error in the documentation for the redirect router. Exim
712 does (usually) call initgroups() when daemonizing.
714 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
715 with consistent privilege compared to when running as a daemon.
717 TF/03 Note in the spec that $authenticated_id is not set for local
718 submissions from trusted users.
720 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
721 Thanks to Dean Brooks <dean@iglou.com> for the patch.
723 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
724 by adding some example configuration directives to the default
725 configuration file. A little bit of work is required to uncomment the
726 directives and define how usernames and passwords are checked, but
727 there is now a framework to start from.
729 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
730 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
731 without this. I don't know how relevant this is to other LDAP libraries.
733 PH/02 Add the verb name to the "unknown ACL verb" error.
735 PH/03 Magnus Holmgren's patch for filter_prepend_home.
737 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
739 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
740 directory not expanded when it should be if an expanded home directory
741 was set for the address (which is overridden by the transport).
743 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
746 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
747 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
748 because it is too late at that time, and has no effect.
750 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
751 security issue with \' (bugzilla #107). I could not use the
752 PQescapeStringConn() function, because it needs a PGconn value as one of
755 PH/08 When testing addresses using -bt, indicate those final addresses that
756 are duplicates that would not cause an additional delivery. At least one
757 person was confused, thinking that -bt output corresponded to deliveries.
758 (Suppressing duplicates isn't a good idea as you lose the information
759 about possibly different redirections that led to the duplicates.)
761 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
762 systems where poll() doesn't work, in particular OS X.
764 PH/10 Added more information to debugging output for retry time not reached.
766 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
767 operations in malware.c.
769 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
772 PH/13 If write_rejectlog was set false when logging was sent to syslog with
773 syslog_duplication set false, log lines that would normally be written
774 both the the main log and to the reject log were not written to syslog at
777 PH/14 In the default configuration, change the use of "message" in ACL warn
778 statements to "add_header".
780 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
781 not followed by a command (e.g. "seen endif").
783 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
784 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
787 PH/17 Added extra conditions to the default value of delay_warning_condition
791 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
792 { match{$h_precedence:}{(?i)bulk|list|junk} } \
793 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
796 The Auto-Submitted: and various List- headers are standardised, whereas I
797 don't think Precedence: ever was.
799 PH/18 Refactored debugging code in route_finduser() to show more information,
800 in particular, the error code if getpwnam() issues one.
802 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
803 This is apparently needed in addition to the PH/07 change above to avoid
804 any possible encoding problems.
806 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
807 but not after initializing Perl.
809 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
810 output them only if debugging. By default they are written stderr,
811 apparently, which is not desirable.
813 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
816 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
819 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
821 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
822 authenticated or an ident call has been made. Suppress the default
823 values for $authenticated_id and $authenticated_sender (but permit -oMai
824 and -oMas) when testing with -bh.
826 PH/24 Re-jigged the order of the tests in the default configuration so that the
827 tests for valid domains and recipients precede the DNS black list and CSA
828 tests, on the grounds that those ones are more expensive.
830 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
831 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
832 This bug exists in every version of Exim that I still have, right back to
835 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
836 However, an attempt to turn on case-sensitivity in a regex key by
837 including (?-i) didn't work because the subject string was already
838 lowercased, and the effects were non-intuitive. It turns out that a
839 one-line patch can be used to allow (?-i) to work as expected.
845 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
846 other effects) broke the use of negated acl sub-conditions.
848 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
851 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
852 "Deny" causes Exim to reject the incoming connection with a 554 error.
853 Unfortunately, if there is a major crisis, such as a disk failure,
854 tcp-wrappers gives "deny", whereas what one would like would be some
855 kind of temporary error. A kludge has been added to help with this.
856 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
857 554 error is used if errno is still zero or contains ENOENT (which occurs
858 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
861 PH/03 Add -lutil to the default FreeBSD LIBS setting.
863 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
864 errors. Otherwise a message that provokes a temporary error (when other
865 messages do not) can cause a whole host to time out.
867 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
868 addresses were routed directly to files or pipes from a redirect router.
869 File deliveries just didn't batch; pipe deliveries might have suffered
872 PH/06 A failure to get a lock for a hints database would erroneously always say
873 "Failed to get write lock", even when it was really a read lock.
875 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
876 of 0600. This has been changed to use the value of the lockfile_mode
877 option (which defaults to 0600).
879 PH/08 Applied small patch from the Sieve maintainer.
881 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
882 folder from quota calculations, a direct delivery into this folder messed
883 up the contents of the maildirsize file. This was because the regex was
884 used only to exclude .Trash (or whatever) when the size of the mailbox
885 was calculated. There was no check that a delivery was happening into an
886 excluded directory. This bug has been fixed by ignoring all quota
887 processing for deliveries into excluded directories.
889 PH/10 Added the maildirfolder_create_regex option to appendfile.
895 PH/01 The code for finding all the local interface addresses on a FreeBSD
896 system running IPv6 was broken. This may well have applied to all BSD
897 systems, as well as to others that have similar system calls. The broken
898 code found IPv4 interfaces correctly, but gave incorrect values for the
899 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
900 that it would not match correctly against @[] and not recognize the IPv6
903 PH/02 The ipliteral router was not recognizing addresses of the form user@
904 [ipv6:....] because it didn't know about the "ipv6:" prefix.
906 PH/03 Added disable_ipv6.
908 PH/04 Changed $reply_address to use the raw form of the headers instead of the
909 decoded form, because it is most often used to construct To: headers
910 lines in autoreplies, and the decoded form may well be syntactically
911 invalid. However, $reply_address has leading white space removed, and all
912 newlines turned into spaces so that the autoreply transport does not
915 PH/05 If group was specified without a user on a router, and no group or user
916 was specified on a transport, the group from the router was ignored.
918 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
919 for visible compile-time settings that can be used to change these
920 numbers, for those that want even more. Backwards compatibility with old
921 spool files has been maintained. However, going back to a previous Exim
922 release will lost any variables that are in spool files.
924 PH/07 Two small changes when running in the test harness: increase delay when
925 passing a TCP/IP connection to a new process, in case the original
926 process has to generate a bounce, and remove special handling of
927 127.0.0.2 (sic), which is no longer necessary.
929 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
930 be the same on different OS.
932 PH/09 Moved a debug statement in filter processing to avoid a race problem when
935 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
936 whether --show-vars was specified or not
938 JJ/02 exipick: Added support for new ACL variable spool format introduced
941 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
942 syntactically invalid From: or Reply-to: line, and a filter used this to
943 generate an autoreply, and therefore failed to obtain an address for the
944 autoreply, Exim could try to deliver to a non-existent relative file
945 name, causing unrelated and misleading errors. What now happens is that
946 it logs this as a hard delivery error, but does not attempt to create a
949 PH/11 The exinext utility has a -C option for testing purposes, but although
950 the given file was scanned by exinext itself; it wasn't being passed on
951 when Exim was called.
953 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
954 an end-of-file indication when reading a command response.
956 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
957 compiled. In many other places in Exim, IPv6 addresses are always
958 recognized, so I have changed this. It also means that IPv4 domain
959 literals of the form [IPV4:n.n.n.n] are now always recognized.
961 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
962 used if the router is not running as root, for example, when verifying at
963 ACL time, or when using -bh. The debugging output from this situation was
964 non-existent - all you got was a failure to exec. I have made two
967 (a) Failures to set uid/gid, the current directory, or a process leader
968 in a subprocess such as that created by queryprogram now generate
969 suitable debugging ouput when -d is set.
971 (b) The queryprogram router detects when it is not running as root,
972 outputs suitable debugging information if -d is set, and then runs
973 the subprocess without attempting to change uid/gid.
975 PH/15 Minor change to Makefile for building test_host (undocumented testing
978 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
979 additional section of a DNS packet that returns MX or SRV records.
980 Instead, it always explicitly searches for A/AAAA records. This avoids
981 major problems that occur when a DNS server includes only records of one
982 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
983 fixed another bug: if SRV records were looked up and the corresponding
984 address records were *not* found in the additional section, the port
985 values from the SRV records were lost.
987 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
988 using the correct key (the original address) when searching the retry
989 rules in order to find which one to use for generating the retry hint.
991 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
992 adding the default one. Similarly, if it contains a Reply-To: header, the
993 errors_reply_to option, if set, is not used.
995 PH/19 When calculating a retry time, Exim used to measure the "time since
996 failure" by looking at the "first failed" field in the retry record. Now
997 it does not use this if it is later than than the arrival time of the
998 message. Instead it uses the arrival time. This makes for better
999 behaviour in cases where some deliveries succeed, thus re-setting the
1000 "first failed" field. An example is a quota failure for a huge message
1001 when small messages continue to be delivered. Without this change, the
1002 "time since failure" will always be short, possible causing more frequent
1003 delivery attempts for the huge message than are intended.
1004 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1006 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1007 $1, $2, $3) because the numerical variables can be reset during some
1008 expansion items (e.g. "match"), thereby losing the authentication data.
1010 PH/21 Make -bV show the size of off_t variables so that the test suite can
1011 decide whether to run tests for quotas > 2G.
1013 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1014 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1015 filecount value is greater than 2G or if a quota value is greater than 2G
1016 on a system where the size of off_t is not greater than 4, a panic error
1019 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1020 never match. The debug and -bh output now contains an explicit error
1021 message indicating a malformed IPv4 address or mask.
1023 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1024 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1025 PH/23 above applies.
1027 PH/25 Do not write to syslog when running in the test harness. The only
1028 occasion when this arises is a failure to open the main or panic logs
1029 (for which there is an explicit test).
1031 PH/26 Added the /no_tell option to "control=freeze".
1033 PH/27 If a host name lookup failed very early in a connection, for example, if
1034 the IP address matched host_lookup and the reverse lookup yielded a name
1035 that did not have a forward lookup, an error message of the form "no IP
1036 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1037 could be logged. Now it outputs the IP address instead of "NULL".
1039 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1040 allows the sender and the authenticated sender to be set when
1041 submitting a message from within Exim. Since child_open_exim() is
1042 documented for local_scan(), the new function should be too.
1044 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1045 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1046 results in an empty string is now treated as unset.
1048 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1050 PH/31 Added sender_verify_fail logging option.
1052 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1053 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1054 tidied the source and removed it altogether.
1056 PH/33 When a queue run was abandoned because the load average was too high, a
1057 log line was always written; now it is written only if the queue_run log
1058 selector is set. In addition, the log line for abandonment now contains
1059 information about the queue run such as the pid. This is always present
1060 in "start" and "stop" lines but was omitted from the "abandon" line.
1062 PH/34 Omit spaces between a header name and the colon in the error message that
1063 is given when verify = headers_syntax fails (if there are lots of them,
1064 the message gets confusing).
1066 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1067 names, as there are now some PTR records that contain slashes. This check
1068 is only to protect against broken name servers that fall over on strange
1069 characters, so the fact that it applies to all lookups doesn't matter.
1071 PH/36 Now that the new test suite is complete, we can remove some of the
1072 special code in Exim that was needed for the old test suite. For example,
1073 sorting DNS records because real resolvers return them in an arbitrary
1074 order. The new test suite's fake resolver always returns records in the
1077 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1078 bounces) except when queue_only is set, to avoid logging races between
1079 the different processes.
1081 PH/38 Panic-die if .include specifies a non-absolute path.
1083 PH/39 A tweak to the "H" retry rule from its user.
1085 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1086 a label. They prevented compilation on older perls.
1088 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1089 a warning to be raised on newish perls.
1091 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1092 on queue. Changes to match documented behaviour of showing count of
1093 messages matching specified criteria.
1095 PH/40 Changed the default ident timeout from 30s to 5s.
1097 PH/41 Added support for the use of login_cap features, on those BSD systems
1098 that have them, for controlling the resources used by pipe deliveries.
1100 PH/42 The content-scanning code uses fopen() to create files in which to put
1101 message data. Previously it was not paying any attention to the mode of
1102 the files. Exim runs with umask(0) because the rest of the code creates
1103 files with open(), and sets the required mode explicitly. Thus, these
1104 files were ending up world-writeable. This was not a big issue, because,
1105 being within the spool directory, they were not world-accessible. I have
1106 created a function called modefopen, which takes an additional mode
1107 argument. It sets umask(777), creates the file, chmods it to the required
1108 mode, then resets the umask. All the relevant calls to fopen() in the
1109 content scanning code have been changed to use this function.
1111 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1112 to 24 hours. This avoids potential overflow problems when processing G
1113 and H retry rules. I suspect nobody ever tinkers with this value.
1115 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1117 PH/45 When the plaintext authenticator is running as a client, the server's
1118 challenges are checked to ensure they are valid base64 strings. By
1119 default, the authentication attempt is cancelled if an invalid string is
1120 received. Setting client_ignore_invalid_base64 true ignores these errors.
1121 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1122 they are received. Thus, the responses can be made to depend on the
1123 challenges. If an invalid string is ignored, an empty string is placed in
1126 PH/46 Messages that are created by the autoreply transport now contains a
1127 References: header, in accordance with RFCs 2822 and 3834.
1129 PH/47 Added authenticated_sender_force to the smtp transport.
1131 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1133 PH/49 Installed latest patch from the Sieve maintainer.
1135 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1136 was also set, the appendfile transport was unnecessarily scanning a
1137 directory of message files (e.g. for maildir delivery) to find the count
1138 of files (along with the size), even though it did not need this
1139 information. It now does the scan only if it needs to find either the
1140 size of the count of files.
1142 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1144 PH/52 Two bugs concerned with error handling when the smtp transport is
1147 (i) Exim was not creating retry information for temporary errors given
1148 for individual recipients after the DATA command when the smtp transport
1149 was used in LMTP mode. This meant that they could be retried too
1150 frequently, and not timed out correctly.
1152 (ii) Exim was setting the flag that allows error details to be returned
1153 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1154 recipients that were returned after the DATA command.
1156 PH/53 This is related to PH/52, but is more general: for any failing address,
1157 when detailed error information was permitted to be returned to the
1158 sender, but the error was temporary, then after the final timeout, only
1159 "retry timeout exceeded" was returned. Now it returns the full error as
1160 well as "retry timeout exceeded".
1162 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1163 do this, and (what is worse) MTAs that accept it.
1165 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1166 will now be deprecated.
1168 PH/56 New os.c-cygwin from the Cygwin maintainer.
1170 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1171 formats (previously only available in exim formats via -bpr, -bpru,
1172 and -bpra. Now also available in native and exiqgrep formats)
1174 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1175 with very large, slow to parse queues
1177 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1179 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1181 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1182 responses to authentication challenges, though it was showing the
1183 challenges; (ii) I've removed the CR characters from the debug output for
1186 PH/58 Allow for the insertion of a newline as well as a space when a string
1187 is turned into more than one encoded-word during RFC 2047 encoding. The
1188 Sieve code now uses this.
1190 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1191 data_4xx, lost_connection, tls_required.
1193 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1194 message was being sent as an SMTP response.
1196 PH/61 Add -l and -k options to exicyclog.
1198 PH/62 When verifying, if an address was redirected to one new address, so that
1199 verification continued, and the new address failed or deferred after
1200 having set something in $address_data, the value of $address_data was not
1201 passed back to the ACL. This was different to the case when no
1202 redirection occurred. The value is now passed back in both cases.
1204 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1205 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1206 use login_cap.h, so on its own it isn't the distinguishing feature. The
1207 new name refers directly to the setclassresources() function.
1209 PH/65 Added configuration files for NetBSD3.
1211 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1213 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1214 is preferred over IPv4.
1216 PH/68 The bounce_return_message and bounce_return_body options were not being
1217 honoured for bounces generated during the reception of non-SMTP messages.
1218 In particular, this applied to messages rejected by the ACL. This bug has
1219 been fixed. However, if bounce_return_message is true and bounce_return_
1220 body is false, the headers that are returned for a non-SMTP message
1221 include only those that have been read before the error was detected.
1222 (In the case of an ACL rejection, they have all been read.)
1224 PH/69 The HTML version of the specification is now built in a directory called
1225 spec_html instead of spec.html, because the latter looks like a path with
1226 a MIME-type, and this confuses some software.
1228 PH/70 Catch two compiler warnings in sieve.c.
1230 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1231 function verify_get_ident() calls ip_connect() to connect a socket, but
1232 if the "connect()" function timed out, ip_connect() used to close the
1233 socket. However, verify_get_ident() also closes the socket later, and in
1234 between Exim writes to the log, which may get opened at this point. When
1235 the socket was closed in ip_connect(), the log could get the same file
1236 descriptor number as the socket. This naturally causes chaos. The fix is
1237 not to close the socket in ip_connect(); the socket should be closed by
1238 the function that creates it. There was only one place in the code where
1239 this was missing, in the iplookup router, which I don't think anybody now
1240 uses, but I've fixed it anyway.
1242 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1243 well as to direct DNS lookups. Otherwise the handling of names in host
1244 lists is inconsistent and therefore confusing.
1250 PH/01 Two changes to the default runtime configuration:
1252 (1) Move the checks for relay_from_hosts and authenticated clients from
1253 after to before the (commented out) DNS black list checks.
1255 (2) Add control=submission to the relay_from_hosts and authenticated
1256 clients checks, on the grounds that messages accepted by these
1257 statements are most likely to be submissions.
1259 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1261 (1) Generate an error if the third argument for the ${prvs expansion is
1264 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1267 (3) Reset the variables that are obtained from the first argument of
1268 ${prvscheck and used in the second argument before leaving the code,
1269 because their memory is reclaimed, so using them afterwards may do
1272 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1273 one (it's much easier than Tom thought :-).
1275 (5) Because of (4), we can now allow for the use of $prvscheck_result
1276 inside the third argument.
1278 PH/03 For some reason, the default setting of PATH when running a command from
1279 a pipe transport was just "/usr/bin". I have changed it to
1282 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1283 anything to be listed in the output from -bV.
1285 PH/05 When a filter generated an autoreply, the entire To: header line was
1286 quoted in the delivery log line, like this:
1288 => >A.N.Other <ano@some.domain> <original@ddress> ...
1290 This has been changed so that it extracts the operative address. There
1291 may be more than one such address. If so, they are comma-separated, like
1294 => >ano@some.domain,ona@other.domain <original@ddress> ...
1296 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1297 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1298 not being looked up in the rDNS to get a host name, Exim was showing the
1299 IP address twice in Received: lines, even though the IP addresses were
1300 identical. For example:
1302 Received: from [1.2.3.4] (helo=[1.2.3.4])
1304 However, if the real host name was known, it was omitting the HELO data
1305 if it matched the actual IP address. This has been tidied up so that it
1306 doesn't show the same IP address twice.
1308 PH/07 When both +timestamp and +memory debugging was on, the value given by
1309 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1310 called by the debug printing, thereby overwriting the timestamp buffer.
1311 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1313 PH/08 When the original message was included in an autoreply transport, it
1314 always said "this is a copy of the message, including all the headers",
1315 even if body_only or headers_only was set. It now gives an appropriate
1318 PH/09 Applied a patch from the Sieve maintainer which:
1320 o fixes some comments
1321 o adds the (disabled) notify extension core
1322 o adds some debug output for the result of if/elsif tests
1323 o points to the current vacation draft in the documentation
1324 and documents the missing references header update
1328 o fixes a bug in processing the envelope test (when testing
1329 multiple envelope elements, the last element determinted the
1332 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1333 Electronic Mail") by including:
1335 Auto-submitted: auto-generated
1337 in the messages that it generates (bounce messages and others, such as
1338 warnings). In the case of bounce messages for non-SMTP mesages, there was
1339 also a typo: it was using "Auto_submitted" (underscore instead of
1340 hyphen). Since every message generated by Exim is necessarily in response
1341 to another message, thes have all been changed to:
1343 Auto-Submitted: auto-replied
1345 in accordance with these statements in the RFC:
1347 The auto-replied keyword:
1349 - SHOULD be used on messages sent in direct response to another
1350 message by an automatic process,
1352 - MUST NOT be used on manually-generated messages,
1354 - MAY be used on Delivery Status Notifications (DSNs) and Message
1355 Disposition Notifications (MDNs),
1357 - MUST NOT be used on messages generated by automatic or periodic
1358 processes, except for messages which are automatic responses to
1361 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1362 to the default Received: header definition.
1364 PH/12 Added log selector acl_warn_skipped (default on).
1366 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1367 variables because (a) they are in the wrong storage pool and (b) even if
1368 they were copied, it wouldn't work properly because of the caching.
1370 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1371 checking when decoding. Apparently there are clients that generate
1372 overlong encoded strings. Why am I not surprised?
1374 PH/15 If the first argument of "${if match_address" was not empty, but did not
1375 contain an "@" character, Exim crashed. Now it writes a panic log message
1376 and treats the condition as false.
1378 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1380 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1381 extension "envelope-auth". The code is finished and in agreement with
1382 other implementations, but there is no documentation so far and in fact,
1383 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1384 not changing the active code.
1386 Print executed "if" and "elsif" statements when debugging is used. This
1387 helps a great deal to understand what a filter does.
1389 Document more things not specified clearly in RFC3028. I had all this
1390 sorted out, when out of a sudden new issues came to my mind. Oops."
1392 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1395 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1396 canonical form (as documented). However, after a host name lookup from
1397 the IP address, check_host() was doing a simple string comparison with
1398 addresses acquired from the DNS when checking that the found name did
1399 have the original IP as one of its addresses. Since any found IPv6
1400 addresses are likely to be in abbreviated form, the comparison could
1401 fail. Luckily, there already exists a function for doing the comparison
1402 by converting both addresses to binary, so now that is used instead of
1403 the text comparison.
1405 PH/20 There was another similar case to PH/19, when a complete host name was
1406 given in a host list; looking up its IP address could give an abbreviated
1407 form, whereas the current host's name might or might not be abbreviated.
1408 The same fix has been applied.
1414 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1415 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1418 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1419 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1421 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1423 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1424 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1425 Also, for "Auto-Submitted", treat anything other than "no" as
1426 non-personal, in accordance with RFC 3834. (Previously it treated
1427 anything starting "auto-" as non-personal.)
1429 TF/01 The control=submission/name=... option had a problem with syntax
1430 errors if the name included a slash character. The /name= option
1431 now slurps the rest of the string, so it can include any characters
1432 but it must come last in the list of options (after /sender_retain
1435 PH/05 Some modifications to the interface to the fake nameserver for the new
1443 TK/01 Added the "success_on_redirect" address verification option. See
1444 NewStuff for rationale and an example.
1446 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1448 PH/02 Patch to exigrep to allow it to work on syslog lines.
1450 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1451 fread() to skip over the body file's header line, because in Cygwin the
1452 header line is locked and is inaccessible.
1454 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1455 co-exist for some time) to make it clear that it is the Exim ID that is
1456 referenced, not the Message-ID: header line.
1458 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1459 string_format() function, because snprintf() does not exist on all
1462 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1463 ${stat: expansion item.
1465 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1466 protocol synchronization error", to keep the pedants happy.
1468 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1469 well as for IRIX systems, when gcc is being used. See the host.c source
1472 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1474 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1477 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1478 transport and to the smtp transport in LMTP mode.
1480 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1482 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1483 run a filter in a subprocess. This could lead to confusion in subsequent
1484 lookups in the parent process. There should also be a search_tidyup() at
1485 the end of the subprocess.
1487 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1488 only if the host matched helo_try_verify_hosts, which caused the
1489 verification to occur when the EHLO/HELO command was issued. The ACL just
1490 tested the remembered result. Now, if a previous verification attempt has
1491 not happened, "verify = helo" does it there and then.
1493 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1495 TK/03 Fix log output including CR from clamd.
1497 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1498 exist provoked a memory error which could cause a segfault.
1500 PH/15 Installed PCRE 6.2
1502 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1504 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1505 of the problem. Specifically, suggested +O2 rather than +O1 for the
1508 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1510 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1511 record, the port was not correctly logged when the outgoing_port log
1512 selector was set (it logged the transort's default port).
1514 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1515 fallback_hosts, and "hosts" in the smtp transport.
1517 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1518 host errors such as "Connection refused".
1520 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1521 authentication with radiusclient 0.4.9:
1523 - Error returned from rc_read_config was caught wrongly
1524 - Username/password not passed on to radius server due to wrong length.
1526 The presumption is that some radiusclient API changes for 4.51/PH/17
1527 were not taken care of correctly. The code is still untested by me (my
1528 Linux distribution still has 0.3.2 of radiusclient), but it was
1529 contributed by a Radius user.
1531 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1532 expanding the "port" option of the smtp transport.
1534 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1535 while reading a MIME header. Thanks to Tom Hughes for a patch.
1537 PH/24 Include config.h inside local_scan.h so that configuration settings are
1540 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1541 that in an ACL for RCPT (for example), you can examine exactly what was
1544 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1545 commands, but it was not correctly comparing the address with the actual
1546 client host address. Thus, it would show the EHLO address in Received:
1547 header lines when this was not necessary.
1549 PH/27 Added the % operator to ${eval:}.
1551 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1552 it should be ignoring failures (because with -C, for example, it has lost
1553 privilege). It wasn't ignoring creation failures other than "already
1556 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1559 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1560 because an input line was too long, either on its own, or by virtue of
1561 too many continuations, the temporary file was not being removed, and the
1562 return code was incorrect.
1564 PH/31 Missing "BOOL" in function definition in filtertest.c.
1566 PH/32 Applied Sieve patches from the maintainer.
1568 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1570 PH/33 Added "verify = not_blind".
1572 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1573 Local/Makefile (with some defaults set). These are used in built scripts
1574 such as exicyclog, but they have never been used in the exim_install
1575 script (though there are many overriding facilities there). I have
1576 arranged that the exim_install script now takes note of these two
1579 PH/35 Installed configuration files for Dragonfly.
1581 PH/36 When a locally submitted message by a trusted user did not contain a
1582 From: header, and the sender address was obtained from -f or from an SMTP
1583 MAIL command, and the trusted user did not use -F to supply a sender
1584 name, $originator_name was incorrectly used when constructing a From:
1585 header. Furthermore, $originator_name was used for submission mode
1586 messages from external hosts without From: headers in a similar way,
1587 which is clearly wrong.
1589 PH/37 Added control=suppress_local_fixups.
1591 PH/38 When log_selector = +received_sender was set, and the addition of the
1592 sender made the log line's construction buffer exactly full, or one byte
1593 less than full, an overflow happened when the terminating "\n" was
1596 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1597 when the result of a list match is failure because a DNS lookup failed.
1599 PH/40 RM_COMMAND is now used in the building process.
1601 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1602 the "build-* directories that it finds.
1604 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1605 address in a domain literal was a prefix of an interface address.
1607 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1608 when verifying a sender address, unless rewrite_headers is false.
1610 PH/44 Wrote a long comment about why errors_to addresses are verified as
1611 recipients, not senders.
1613 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1614 the ratelimit ACL was added.
1616 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1618 PH/46 Added extra information about PostgreSQL errors to the error string.
1620 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1621 avoiding the need to install special zones in a real server. This is
1622 backwards compatible; if it can't find the fake resolver, it drops back.
1623 Thus, both old and new test suites can be run.
1625 TF/02 Added util/ratelimit.pl
1627 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1628 clock is set back in time.
1630 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1631 Candler <B.Candler@pobox.com>.
1633 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1634 OK for addresses that are the result of redirections.
1636 PH/48 A number of further additions for the benefit of the new test suite,
1637 including a fake gethostbyname() that interfaces to the fake DNS resolver
1640 TF/06 The fix for widen_domains has also been applied to qualify_single and
1641 search_parents which are the other dnslookup options that can cause
1644 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1647 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1649 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1650 service" if it encountered an MX record with an empty target hostname.
1651 The message is now "an MX or SRV record indicated no SMTP service".
1653 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1654 if the DNS of the sending site is misconfigured. This is quite a
1655 common situation. This change restores the behaviour of treating a
1656 helo verification defer as a failure.
1658 PH/51 If self=fail was set on a router, the bounce message did not include the
1659 actual error message.
1665 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1667 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1668 command itself ended in error, the underlying message about the transport
1669 filter timeout was being overwritten with the pipe command error. Now the
1670 underlying error message should be appended to the second error message.
1672 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1674 PH/02 Reduce the amount of output that "make" produces by default. Full output
1675 can still be requested.
1677 PH/03 The warning log line about a condition test deferring for a "warn" verb
1678 was being output only once per connection, rather than after each
1679 occurrence (because it was using the same function as for successful
1680 "warn" verbs). This seems wrong, so I have changed it.
1682 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1683 it should not have, which might have caused a crash in the right
1684 circumstances, but probably never did.
1686 PH/04 Installed a modified version of Tony Finch's patch to make submission
1687 mode fix the return path as well as the Sender: header line, and to
1688 add a /name= option so that you can make the user's friendly name appear
1691 TF/03 Added the control = fakedefer ACL modifier.
1693 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1694 Mark Lowes for thorough testing.
1696 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1698 TK/03 Merged latest SRS patch from Miles Wilton.
1700 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1701 with the definition in sysexits.h (which is #included earlier).
1702 Fortunately, Exim does not actually use EX_OK. The code used to try to
1703 preserve the sysexits.h value, by assumimg that macro definitions were
1704 scanned for macro replacements. I have been disabused of this notion,
1705 so now the code just undefines EX_OK before #including unistd.h.
1707 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1708 in the smtp transport. When a block could not be written in a single
1709 write() function, the timeout was being re-applied to each part-write.
1710 This seems wrong - if the receiver was accepting one byte at a time it
1711 would take for ever. The timeout is now adjusted when this happens. It
1712 doesn't have to be particularly precise.
1714 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1715 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1717 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1718 without a domain if the check to <postmaster@domain> fails.
1720 SC/01 Eximstats: added -xls and the ability to specify output files
1721 (patch written by Frank Heydlauf).
1723 SC/02 Eximstats: use FileHandles for outputing results.
1725 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1727 SC/04 Eximstats: fixed display of large numbers with -nvr option
1729 SC/05 Eximstats: fixed merging of reports with empty tables.
1731 SC/06 Eximstats: added the -include_original_destination flag
1733 SC/07 Eximstats: removed tabs and trailing whitespace.
1735 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1737 TK/06 MBOX spool code: Add real "From " MBOX separator line
1738 so the .eml file is really in mbox format (even though
1739 most programs do not really care). Patch from Alex Miller.
1741 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1742 The latter is generated from $received_to and is only set if the
1743 message has one envelope recipient. SA can use these headers,
1744 obviously out-of-the-box. Patch from Alex Miller.
1746 PH/08 The ${def test on a variable was returning false if the variable's
1747 value was "0", contrary to what the specification has always said!
1748 The result should be true unless the variable is empty.
1750 PH/09 The syntax error of a character other than { following "${if
1751 def:variable_name" (after optional whitespace) was not being diagnosed.
1752 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1753 accidental colon was present, for example, could give incorrect results.
1755 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1756 result is used (not including appendfile, where other changes are about
1759 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1760 This involved changing a lot of size variables from int to off_t. It
1761 should work with maildirs and everything.
1763 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1764 spamd dying while we are connected to it.
1766 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1769 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1770 function for BDB 4.3.
1772 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1774 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1775 the sources for PCRE have been re-arranged and more files are now
1778 PH/15 The code I had for printing potentially long long variables in PH/11
1779 above was not the best (it lost precision). The length of off_t variables
1780 is now inspected at build time, and an appropriate printing format (%ld
1781 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1782 to be "long long int" or "long int". This is needed for the internal
1783 formatting function string_vformat().
1785 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1786 the configuration file to be ":syslog", then the script "guesses" where
1787 the logs files are, rather than using the compiled in default. In our
1788 case the guess is not the same as the compiled default, so the script
1789 suddenly stopped working when I started to use syslog. The patch checks
1790 to see if log_file_path is "". If so, it attempts to read it from exim
1791 with no configuration file to get the compiled in version, before it
1792 falls back to the previous guessing code."
1794 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1795 implementing BATV in an Exim configuration. See NewStuff for the gory
1798 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1799 Makefile that are specific to HP-UX.
1801 PH/18 If the "use_postmaster" option was set for a recipient callout together
1802 with the "random" option, the postmaster address was used as the MAIL
1803 FROM address for the random test, but not for the subsequent recipient
1804 test. It is now used for both.
1806 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1807 patch removes a few documentation additions to RFC 3028, because the
1808 latest draft now contains them. It adds the new en;ascii-case comparator
1809 and a new error check for 8bit text in MIME parts. Comparator and
1810 require names are now matched exactly. I enabled the subaddress
1811 extension, but it is not well tested yet (read: it works for me)."
1813 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1814 rework some of the code of TK/09 above to avoid the hardwired use of
1815 "%lld" and "long long". Replaced the call to snprintf() with a call to
1818 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1819 records point to non-existent hosts", "retry timeout exceeded", and
1820 "retry time not reached for any host after a long failure period".
1822 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1823 experimental DomainKeys support:
1825 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1826 (2) On an error such as an illegally used "control", the wrong name for
1827 the control was given.
1829 These problems did NOT occur unless DomainKeys support was compiled.
1831 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1833 PH/24 Added ${if match_ip condition.
1835 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1836 databases so that it will be absolutely obvious if a crash occurs in the
1837 DB library. This is a regular occurrence (often caused by mis-matched
1840 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1841 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1842 picked up on a user's system that detects such things. There doesn't seem
1843 to be a gcc warning option for this - only an attribute that has to be
1844 put on the function's prototype. It seems that in Fedora Core 4 they have
1845 set this on a number of new functions. No doubt there will be more in due
1848 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1849 specify a transport. However, if an address that was verified by such a
1850 router was the subject of a callout, Exim crashed because it tried to
1851 read the rcpt_include_affixes from the non-existent transport. Now it
1852 just assumes that the setting of that option is false. This bug was
1853 introduced by 4.51/PH/31.
1855 PH/28 Changed -d+all to exclude +memory, because that information is very
1856 rarely of interest, but it makes the output a lot bigger. People tend to
1857 do -d+all out of habit.
1859 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1860 code in os-type was giving problems when libc.so lives in lib64, like on
1863 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1864 aren't the modern standard, and it seems that some systems' include files
1865 don't always have them. Exim was already checking for some of the newer
1866 ones like T_AAAA, and defining it itself. I've added checks for all the
1867 record types that Exim uses.
1869 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1870 not automatically generating a new one, as it is supposed to. This
1871 prevented TLS from working. If the file did exist, but contained invalid
1872 data, a new version was generated, as expected. It was only the case of a
1873 non-existent file that was broken.
1875 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1876 with a change in libdomainkeys > 0.64.
1878 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1879 from DNS. If the selector record carries the flag, it now has
1880 precedence over the domain-wide flag.
1882 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1884 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1885 the use of a port name that isn't defined in /etc/services) occurred, the
1886 message was deferred as in a normal delivery, and thus remained on the
1887 spool, instead of being failed because of the mua_wrapper setting. This
1888 is now fixed, and I tidied up some of the mua_wrapper messages at the
1891 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1892 the messages in a hash of arrays rather than using individual hashes.
1893 This is a bit cleaner and results in dramatic memory savings, albeit
1894 at a slight CPU cost.
1896 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1897 as requested by Marc Sherman.
1899 SC/10 Eximstats: added histograms for user specified patterns as requested
1902 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1904 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1905 fopen() in the content-scanning modules that did not already have it.
1911 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1912 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1914 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1916 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1918 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1919 to test Sieve filters that use "vacation".
1921 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1922 that changes the way the GnuTLS parameters are stored in the cache file.
1923 The new format can be generated externally. For backward compatibility,
1924 if the data in the cache doesn't make sense, Exim assumes it has read an
1925 old-format file, and it generates new data and writes a new file. This
1926 means that you can't go back to an older release without removing the
1929 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1930 work if there are any delivery delays because "one_time" forces the
1931 parent to be marked "delivered", so its unseen clone is never tried
1932 again. For this reason, Exim now forbids the simultaneous setting of
1935 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1936 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1937 entry if you want to know the details. The fix, however, neglected to
1938 consider the case when local delivery batching is involved. The test for
1939 "previously delivered" was not happening when checking to see if an
1940 address could be batched with a previous (undelivered) one; under
1941 certain circumstances this could lead to multiple deliveries to the same
1944 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1945 in its include files, and this causes problems building Exim.
1947 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1948 header_syntax) but Exim was just ignoring anything given after a slash.
1949 In particular, this caused confusion with an attempt to use "verify =
1950 reverse_host_lookup/defer_ok". An error is now given when options are
1951 supplied for verify items that do not have them. (Maybe reverse_host_
1952 lookup should have a defer_ok option, but that's a different point.)
1954 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1955 defined by RFC 821) to 2048, because there were problems with some AUTH
1956 commands, and RFC 1869 says the size should be increased for extended
1957 SMTP commands that take arguments.
1959 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1962 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1963 "unknown" error; now it says that the functionality isn't in the binary.
1965 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1966 an address' error message when a string expansion fails (syntax or
1967 whatever). Otherwise the password may appear in the log. Following change
1968 PH/42 below, there is no longer a chance of it appearing in a bounce
1971 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1973 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1974 of its addresses was ever tried. (Bugzilla bug #2.)
1976 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1977 the result incorrectly in the debug output. (It correctly added a newline
1978 to what was transported.)
1980 TF/01 Added $received_time.
1982 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1983 commented out examples of how to interface to a virus scanner and to
1984 SpamAssassin. Also added commented examples of av_scanner and
1985 spamd_address settings.
1987 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1988 and controls are allowed in which ACLs. There were a couple of minor
1989 errors. Some of the entries in the conditions table (which is a table of
1990 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1991 negation of where the condition IS allowed.
1993 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1995 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1996 header file does not have a version number, so I've had to invent a new
1997 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1998 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1999 radiusclient), but it was contributed by a Radius user.
2001 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2002 files or directories, for OpenSSL.
2004 PH/19 When an Exim process that is running as root has to create an Exim log
2005 file, it does so in a subprocess that runs as exim:exim so as to get the
2006 ownership right at creation (otherwise, other Exim processes might see
2007 the file with the wrong ownership). There was no test for failure of this
2008 fork() call, which would lead to the process getting stuck as it waited
2009 for a non-existent subprocess. Forks do occasionally fail when resources
2010 run out. I reviewed all the other calls to fork(); they all seem to check
2013 PH/20 When checking for unexpected SMTP input at connect time (before writing
2014 the banner), Exim was not dealing correctly with a non-positive return
2015 from the read() function. If the client had disconnected by this time,
2016 the result was a log entry for a synchronization error with an empty
2017 string after "input=" when read() returned zero. If read() returned -1
2018 (an event I could not check), uninitialized data bytes were printed.
2019 There were reports of junk text (parts of files, etc) appearing after
2022 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2024 PH/22 Added support for macro redefinition, and (re)definition in between
2025 driver and ACL definitions.
2027 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2028 forgetting to use the resulting value; it was using the unexpanded value.
2030 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2031 hadn't been configured. The fix is from Juergen Kreileder, who
2032 understands it better than I do:
2034 "Here's what I see happening with three configured cyrus_sasl
2035 authenticators configured (plain, login, cram-md5):
2037 On startup auth_cyrus_sasl_init() gets called for each of these.
2038 This means three calls to sasl_listmech() without a specified mech_list.
2039 => SASL tests which mechs of all available mechs actually work
2040 => three warnings about OTP not working
2041 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2043 With the patch, sasl_listmech() also gets called three times. But now
2044 SASL's mech_list option is set to the server_mech specified in the the
2045 authenticator. Or in other words, the answer from sasl_listmech()
2046 gets limited to just the mech you're testing for (which is different
2048 => the return list contains just 'plain' or 'login', 'cram-md5' or
2049 nothing depending on the value of ob->server_mech.
2051 I've just tested the patch: Authentication still works fine,
2052 unavailable mechs specified in the exim configuration are still
2053 caught, and the auth.log warnings about OTP are gone."
2055 PH/25 When debugging is enabled, the contents of the command line are added
2056 to the debugging output, even when log_selector=+arguments is not
2059 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2060 answer is "GNU", and only if the return is "GNU/something" is the answer
2063 PH/27 $acl_verify_message is now set immediately after the failure of a
2064 verification in an ACL, and so is available in subsequent modifiers. In
2065 particular, the message can be preserved by coding like this:
2067 warn !verify = sender
2068 set acl_m0 = $acl_verify_message
2070 Previously, $acl_verify_message was set only while expanding "message"
2071 and "log_message" when a very denied access.
2073 PH/28 Modified OS/os.c-Linux with
2075 -#ifndef OS_LOAD_AVERAGE
2076 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2078 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2079 nomenclature these days.)
2081 PH/29 Installed patch from the Sieve maintainer that adds the options
2082 sieve_useraddress and sieve_subaddress to the redirect router.
2084 PH/30 In these circumstances:
2085 . Two addresses routed to the same list of hosts;
2086 . First host does not offer TLS;
2087 . First host accepts first address;
2088 . First host gives temporary error to second address;
2089 . Second host offers TLS and a TLS session is established;
2090 . Second host accepts second address.
2091 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2092 and peerdn, if requested) that were in fact used only for the second
2095 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2096 attention to any local part prefix or suffix that was matched by the
2097 router that accepted the address. It now behaves in the same way as it
2098 does for delivery: the affixes are removed from the local part unless
2099 rcpt_include_affixes is set on the transport.
2101 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2102 timeout during the DATA phase of an incoming message.
2104 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2105 applied a patch sanctioned by the Sieve maintainer.
2107 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2108 the uid or gid is negative. A case of a negative gid caused this to be
2109 noticed. The fix allows for either to be negative.
2111 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2112 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2115 PH/36 The change PH/12 above was broken. Fixed it.
2117 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2118 the grounds that routing the same address twice would always produce the
2119 same answer. This might have been true once, but it is certainly no
2120 longer true now. Routing a child address may depend on the previous
2121 routing that produced that child. Some complicated redirection strategies
2122 went wrong when messages had multiple recipients, and made Exim's
2123 behaviour dependent on the order in which the addresses were given.
2125 I have moved the duplicate checking until after the routing is complete.
2126 Exim scans the addresses that are assigned to local and remote
2127 transports, and removes any duplicates. This means that more work will be
2128 done, as duplicates will always all be routed, but duplicates are
2129 presumably rare, so I don't expect this is of any significance.
2131 For deliveries to pipes, files, and autoreplies, the duplicate checking
2132 still happens during the routing process, since they are not going to be
2135 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2136 It corrects a timeout issue with spamd. This is Ian's comment: "The
2137 background is that sometimes spamd either never reads data from a
2138 connection it has accepted, or it never writes response data. The exiscan
2139 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2140 blindly assumes that writes won't block so it may never time out."
2142 PH/39 Allow G after quota size as well as K and M.
2144 PH/40 The value set for $authenticated_id in an authenticator may not contain
2145 binary zeroes or newlines because the value is written to log lines and
2146 to spool files. There was no check on this. Now the value is run through
2147 the string_printing() function so that such characters are converted to
2148 printable escape sequences.
2150 PH/41 $message_linecount is a new variable that contains the total number of
2151 lines in the message. Compare $body_linecount, which is the count for the
2154 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2155 bounce and delay warning messages, except in certain special cases, which
2158 (a) An SMTP error message from a remote host;
2159 (b) A message specified in a :fail: redirection;
2160 (c) A message specified in a "fail" command in a system filter;
2161 (d) A message specified in a FAIL return from the queryprogram router;
2162 (e) A message specified by the cannot_route_message router option.
2164 In these cases only, Exim does include the error details in bounce and
2165 warning messages. There are also a few cases where bland messages such
2166 as "unrouteable address" or "local delivery error" are given.
2168 PH/43 $value is now also set for the "else" part of a ${run expansion.
2170 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2171 being worked on, but at least Exim now implements the latest version to
2174 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2175 process to complete was treated as a delivery failure, a timeout while
2176 writing the message to the pipe was logged, but erroneously treated as a
2177 successful delivery. Such timeouts include transport filter timeouts. For
2178 consistency with the overall process timeout, these timeouts are now
2179 treated as errors, giving rise to delivery failures by default. However,
2180 there is now a new Boolean option for the pipe transport called
2181 timeout_defer, which, if set TRUE, converts the failures into defers for
2182 both kinds of timeout. A transport filter timeout is now identified in
2185 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2186 systems where "make" and "gmake" are different, calling "gmake" at top
2187 level broke things. I've arranged for the value of $(MAKE) to be passed
2188 from the Makefile to this script so that it can call the same version of
2192 A note about Exim versions 4.44 and 4.50
2193 ----------------------------------------
2195 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2196 changes of various kinds. As a consequence, a big documentation update was
2197 needed. This delayed the release for rather longer than seemed good, especially
2198 in the light of a couple of (minor) security issues. Therefore, the changes
2199 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2200 release. So 4.44 and 4.50 are in effect two different branches that both start
2203 I have left the 4.50 change log unchanged; it contains all the changes since
2204 4.43. The change log for 4.44 is below; many of its items are identical to
2205 those for 4.50. This seems to be the most sensible way to preserve the
2206 historical information.
2212 1. Minor wording change to the doc/README.SIEVE file.
2214 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2215 computation of the current number of files was incorrect.
2217 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2218 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2221 4. Give more explanation in the error message when the command for a transport
2222 filter fails to execute.
2224 5. There are several places where Exim runs a non-Exim command in a
2225 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2226 was being done only for the command run by the queryprogram router. It is
2227 now done for all such subprocesses. The other cases are: ${run, transport
2228 filters, and the commands run by the lmtp and pipe transports.
2230 6. Added CONFIGURE_GROUP build-time option.
2232 7. Some older OS have a limit of 256 on the maximum number of file
2233 descriptors. Exim was using setrlimit() to set 1000 as a large value
2234 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2235 systems. I've change it so that if it can't get 1000, it tries for 256.
2237 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2238 was an oversight, and furthermore, ever since the addition of extra
2239 controls (e.g. 4.43/32), the checks on when to allow different forms of
2240 "control" were broken. There should now be diagnostics for all cases when a
2241 control that does not make sense is encountered.
2243 9. Added the /retain_sender option to "control=submission".
2245 10. $recipients is now available in the predata ACL (oversight).
2247 11. Tidy the search cache before the fork to do a delivery from a message
2248 received from the command line. Otherwise the child will trigger a lookup
2249 failure and thereby defer the delivery if it tries to use (for example) a
2250 cached ldap connection that the parent has called unbind on.
2252 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2253 of $address_data from the recipient verification was clobbered by the
2254 sender verification.
2256 13. The value of address_data from a sender verification is now available in
2257 $sender_address_data in subsequent conditions in the ACL statement.
2259 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2261 15. Added a new option "connect=<time>" to callout options, to set a different
2264 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2265 was its contents. (It was OK if the option was not defined at all.)
2267 17. A "Completed" log line is now written for messages that are removed from
2268 the spool by the -Mrm option.
2270 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2271 information about exactly what failed.
2273 19. Added -dd to debug only the daemon process.
2275 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2276 handles timeouts, both on the server side and network timeouts. Renamed the
2277 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2279 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2280 It is now set to "smtps".
2282 22. $host_address is now set to the target address during the checking of
2283 ignore_target_hosts.
2285 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2286 being passed; this would have caused $sender_host_name to have been used if
2287 matching the list had actually called for a host name (not very likely,
2288 since this list is usually IP addresses). A host name is now passed as
2291 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2292 code that specifies a non-restarting handler (typically sigaction() in
2293 modern systems) in an attempt to fix a rare and obscure crash bug.
2295 25. Narrowed the window for a race in the daemon that could cause it to ignore
2296 SIGCHLD signals. This is not a major problem, because they are used only to
2297 wake it up if nothing else does.
2299 26. A malformed maildirsize file could cause Exim to calculate negative values
2300 for the mailbox size or file count. Odd effects could occur as a result.
2301 The maildirsize information is now recalculated if the size or filecount
2304 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2305 support for a long time. Removed HAVE_SYS_VFS_H.
2307 28. Installed the latest version of exipick from John Jetmore.
2309 29. In an address list, if the pattern was not a regular expression, an empty
2310 subject address (from a bounce message) matched only if the pattern was an
2311 empty string. Non-empty patterns were not even tested. This was the wrong
2312 because it is perfectly reasonable to use an empty address as part of a
2313 database query. An empty address is now tested by patterns that are
2314 lookups. However, all the other forms of pattern expect the subject to
2315 contain a local part and a domain, and therefore, for them, an empty
2316 address still always fails if the pattern is not itself empty.
2318 30. Exim went into a mad DNS loop when attempting to do a callout where the
2319 host was specified on an smtp transport, and looking it up yielded more
2320 than one IP address.
2322 31. Re-factored the code for checking spool and log partition space into a
2323 function that finds that data and another that does the check. The former
2324 is then used to implement four new variables: $spool_space, $log_space,
2325 $spool_inodes, and $log_inodes.
2327 32. The RFC2047 encoding function was originally intended for short strings
2328 such as real names; it was not keeping to the 75-character limit for
2329 encoded words that the RFC imposes. It now respects the limit, and
2330 generates multiple encoded words if necessary. To be on the safe side, I
2331 have increased the buffer size for the ${rfc2047: expansion operator from
2334 33. It is now permitted to omit both strings after an "if" condition; if the
2335 condition is true, the result is "true". As before, when the second string
2336 is omitted, a false condition yields an empty string. This makes it less
2337 cumbersome to write custom ACL and router conditions.
2339 34. Failure to deliver a bounce message always caused it to be frozen, even if
2340 there was an errors_to setting on the router. The errors_to setting is now
2343 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2344 canonical form (fully expanded) before being placed in
2345 $sender_host_address.
2347 36. The table in the code that translates DNS record types into text (T_A to
2348 "A" for instance) was missing entries for NS and CNAME. It is just possible
2349 that this could have caused confusion if both these types were looked up
2350 for the same domain, because the text type is used as part of Exim's
2351 per-process caching. But the chance of anyone hitting this buglet seems
2354 37. The dnsdb lookup has been extended in a number of ways.
2356 (1) There is a new type, "zns", which walks up the domain tree until it
2357 finds some nameserver records. It should be used with care.
2359 (2) There is a new type, "mxh", which is like "mx" except that it returns
2360 just the host names, not the priorities.
2362 (3) It is now possible to give a list of domains (or IP addresses) to be
2363 looked up. The behaviour when one of the lookups defers can be
2364 controlled by a keyword.
2366 (4) It is now possible to specify the separator character for use when
2367 multiple records are returned.
2369 38. The dnslists ACL condition has been extended: it is now possible to supply
2370 a list of IP addresses and/or domains to be looked up in a particular DNS
2373 39. Added log_selector=+queue_time_overall.
2375 40. When running the queue in the test harness, wait just a tad after forking a
2376 delivery process, to get repeatability of debugging output.
2378 41. Include certificate and key file names in error message when GnuTLS fails
2379 to set them up, because the GnuTLS error message doesn't include the name
2380 of the failing file when there is a problem reading it.
2382 42. Allow both -bf and -bF in the same test run.
2384 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2386 44. The "Exiscan patch" is now merged into the mainline Exim source.
2388 45. Sometimes the final signoff response after QUIT could fail to get
2389 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2390 < 0 before doing a fflush(). This bug looks as though it goes back to the
2391 introduction of TLS in release 3.20, but "sometimes" must have been rare
2392 because the tests only now provoked it.
2394 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2395 (this can affect the format of dates).
2397 47. exim_tidydb, when checking for the continued existence of a message for
2398 which it has found a message-specific retry record, was not finding
2399 messages that were in split spool directories. Consequently, it was
2400 deleting retry records that should have stayed in existence.
2402 48. Steve fixed some bugs in eximstats.
2404 49. The SPA authentication driver was not abandoning authentication and moving
2405 on to the next authenticator when an expansion was forced to fail,
2406 contradicting the general specification for all authenticators. Instead it
2407 was generating a temporary error. It now behaves as specified.
2409 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2410 (the order specifies the preference for clients). The order is now AES256,
2411 AES128, 3DES, ARCFOUR128.
2413 51. Small patch to Sieve code - explicitly set From: when generating an
2416 52. Exim crashed if a remote delivery caused a very long error message to be
2417 recorded - for instance if somebody sent an entire SpamAssassin report back
2418 as a large number of 550 error lines. This bug was coincidentally fixed by
2419 increasing the size of one of Exim's internal buffers (big_buffer) that
2420 happened as part of the Exiscan merge. However, to be on the safe side, I
2421 have made the code more robust (and fixed the comments that describe what
2424 53. Now that there can be additional text after "Completed" in log lines (if
2425 the queue_time_overall log selector is set), a one-byte patch to exigrep
2426 was needed to allow it to recognize "Completed" as not the last thing in
2429 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2430 patch that reportedly fixes this has been added. I am not expert enough to
2431 create a test for it. This is what the patch creator wrote:
2433 "I found a little strange behaviour of ldap code when working with
2434 Windows 2003 AD Domain, where users was placed in more than one
2435 Organization Units. When I tried to give exim partial DN, the exit code
2436 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2437 But simultaneously result of request was absolutely normal ldap result,
2438 so I produce this patch..."
2440 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2441 REFERENCE, so I have modified the code to exclude the patch when that macro
2444 55. Some experimental protocols are using DNS PTR records for new purposes. The
2445 keys for these records are domain names, not reversed IP addresses. The
2446 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2447 leaves it alone. Component reversal etc. now happens only for IP addresses.
2450 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2452 57. Double the size of the debug message buffer (to 2048) so that more of very
2453 long debug lines gets shown.
2455 58. The exicyclog utility now does better if the number of log files to keep
2456 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2458 59. Two changes related to the smtp_active_hostname option:
2460 (1) $smtp_active_hostname is now available as a variable.
2461 (2) The default for smtp_banner uses $smtp_active_hostname instead
2462 of $primary_hostname.
2464 60. The host_aton() function is supposed to be passed a string that is known
2465 to be a valid IP address. However, in the case of IPv6 addresses, it was
2466 not checking this. This is a hostage to fortune. Exim now panics and dies
2467 if the condition is not met. A case was found where this could be provoked
2468 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2469 components; fortuitously, this particular loophole had already been fixed
2470 by change 4.50/55 above.
2472 If there are any other similar loopholes, the new check in host_aton()
2473 itself should stop them being exploited. The report I received stated that
2474 data on the command line could provoke the exploit when Exim was running as
2475 exim, but did not say which command line option was involved. All I could
2476 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2477 running as the user.
2480 61. There was a buffer overflow vulnerability in the SPA authentication code
2481 (which came originally from the Samba project). I have added a test to the
2482 spa_base64_to_bits() function which I hope fixes it.
2485 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2486 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2488 63. The daemon start-up calls getloadavg() while still root for those OS that
2489 need the first call to be done as root, but it missed one case: when
2490 deliver_queue_load_max is set with deliver_drop_privilege. This is
2491 necessary for the benefit of the queue runner, because there is no re-exec
2492 when deliver_drop_privilege is set.
2494 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2495 This has been fixed.
2497 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2498 was in use, was not putting the data itself into the right store pool;
2499 consequently, it could be overwritten for a subsequent message in the same
2500 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2503 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2505 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2506 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2507 the function were treating the return as a boolean value, which happened to
2508 work because 0=false and not-0=true, but is not correct code.
2510 68. The host_aton() function was not handling scoped IPv6 addresses (those
2511 with, for example, "%eth0" on the end) correctly.
2513 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2514 negated items (that is, ~something) in unsigned ints. Some compilers
2515 apparently mutter when there is no cast.
2517 70. If an address verification called from an ACL failed, and did not produce a
2518 user-specific message (i.e. there was only a "system" message), nothing was
2519 put in $acl_verify_message. In this situation, it now puts the system
2522 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2523 session; change 4.31/43 added the unwanted input to the log line - except
2524 that it did not do this in the start of session case. It now does.
2526 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2527 This is wrong and can cause the other end to generate a synchronization
2528 error if it is another Exim or anything else that does the synchronization
2529 check. A QUIT command is no longer sent after a timeout.
2531 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2532 during host lookups.
2534 74. The defer_ok option of callout verification was not working if it was used
2535 when verifying addresses in header lines, that is, for this case:
2537 verify = header_sender/callout=defer_ok
2539 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2540 those file descriptors could be used for SMTP connections. If anything
2541 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2542 could be sent to the SMTP client, causing chaos. The daemon now opens
2543 stdin, stdout, and stderr to /dev/null when it puts itself into the
2546 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2547 log by default. The user can override this with suitable Perl magic.
2549 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2550 the log message when discard triggers, was not working for the DATA ACL or
2551 for the non-SMTP ACL.
2553 78. Error message wording change in sieve.c.
2555 79. If smtp_accept_max_per_host was set, the number of connections could be
2556 restricted to fewer than expected, because the daemon was trying to set up
2557 a new connection before checking whether the processes handling previous
2558 connections had finished. The check for completed processes is now done
2559 earlier. On busy systems, this bug wouldn't be noticed because something
2560 else would have woken the daemon, and it would have reaped the completed
2563 80. If a message was submitted locally by a user whose login name contained one
2564 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2565 It caused a spool format error. I have fixed the spool reading code. A
2566 related problem was that the "from" clause in the Received: line became
2567 illegal because of the space(s). It is now covered by ${quote_local_part.
2569 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2572 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2574 83. Patch from Sieve maintainer to fix unterminated string problem in
2575 "vacation" handling.
2577 84. Some minor changes to the Linux configuration files to help with other
2578 OS variants using glibc.
2580 85. One more patch for Sieve to update vacation handling to latest spec.
2583 ----------------------------------------------------
2584 See the note above about the 4.44 and 4.50 releases.
2585 ----------------------------------------------------
2591 1. Change 4.43/35 introduced a bug that caused file counts to be
2592 incorrectly computed when quota_filecount was set in an appendfile
2595 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2596 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2599 3. Give more explanation in the error message when the command for a transport
2600 filter fails to execute.
2602 4. There are several places where Exim runs a non-Exim command in a
2603 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2604 was being done only for the command run by the queryprogram router. It is
2605 now done for all such subprocesses. The other cases are: ${run, transport
2606 filters, and the commands run by the lmtp and pipe transports.
2608 5. Some older OS have a limit of 256 on the maximum number of file
2609 descriptors. Exim was using setrlimit() to set 1000 as a large value
2610 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2611 systems. I've change it so that if it can't get 1000, it tries for 256.
2613 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2614 was an oversight, and furthermore, ever since the addition of extra
2615 controls (e.g. 4.43/32), the checks on when to allow different forms of
2616 "control" were broken. There should now be diagnostics for all cases when a
2617 control that does not make sense is encountered.
2619 7. $recipients is now available in the predata ACL (oversight).
2621 8. Tidy the search cache before the fork to do a delivery from a message
2622 received from the command line. Otherwise the child will trigger a lookup
2623 failure and thereby defer the delivery if it tries to use (for example) a
2624 cached ldap connection that the parent has called unbind on.
2626 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2627 of $address_data from the recipient verification was clobbered by the
2628 sender verification.
2630 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2631 was its contents. (It was OK if the option was not defined at all.)
2633 11. A "Completed" log line is now written for messages that are removed from
2634 the spool by the -Mrm option.
2636 12. $host_address is now set to the target address during the checking of
2637 ignore_target_hosts.
2639 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2640 being passed; this would have caused $sender_host_name to have been used if
2641 matching the list had actually called for a host name (not very likely,
2642 since this list is usually IP addresses). A host name is now passed as
2645 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2646 code that specifies a non-restarting handler (typically sigaction() in
2647 modern systems) in an attempt to fix a rare and obscure crash bug.
2649 15. Narrowed the window for a race in the daemon that could cause it to ignore
2650 SIGCHLD signals. This is not a major problem, because they are used only to
2651 wake it up if nothing else does.
2653 16. A malformed maildirsize file could cause Exim to calculate negative values
2654 for the mailbox size or file count. Odd effects could occur as a result.
2655 The maildirsize information is now recalculated if the size or filecount
2658 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2659 support for a long time. Removed HAVE_SYS_VFS_H.
2661 18. Updated exipick to current release from John Jetmore.
2663 19. Allow an empty sender to be matched against a lookup in an address list.
2664 Previously the only cases considered were a regular expression, or an
2667 20. Exim went into a mad DNS lookup loop when doing a callout where the
2668 host was specified on the transport, if the DNS lookup yielded more than
2671 21. The RFC2047 encoding function was originally intended for short strings
2672 such as real names; it was not keeping to the 75-character limit for
2673 encoded words that the RFC imposes. It now respects the limit, and
2674 generates multiple encoded words if necessary. To be on the safe side, I
2675 have increased the buffer size for the ${rfc2047: expansion operator from
2678 22. Failure to deliver a bounce message always caused it to be frozen, even if
2679 there was an errors_to setting on the router. The errors_to setting is now
2682 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2683 canonical form (fully expanded) before being placed in
2684 $sender_host_address.
2686 24. Updated eximstats to version 1.33
2688 25. Include certificate and key file names in error message when GnuTLS fails
2689 to set them up, because the GnuTLS error message doesn't include the name
2690 of the failing file when there is a problem reading it.
2692 26. Expand error message when OpenSSL has problems setting up cert/key files.
2695 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2696 (this can affect the format of dates).
2698 28. exim_tidydb, when checking for the continued existence of a message for
2699 which it has found a message-specific retry record, was not finding
2700 messages that were in split spool directories. Consequently, it was
2701 deleting retry records that should have stayed in existence.
2703 29. eximstats updated to version 1.35
2704 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2705 1.35 - bugfix such that pie charts by volume are generated correctly
2707 30. The SPA authentication driver was not abandoning authentication and moving
2708 on to the next authenticator when an expansion was forced to fail,
2709 contradicting the general specification for all authenticators. Instead it
2710 was generating a temporary error. It now behaves as specified.
2712 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2713 (the order specifies the preference for clients). The order is now AES256,
2714 AES128, 3DES, ARCFOUR128.
2716 31. Small patch to Sieve code - explicitly set From: when generating an
2719 32. Exim crashed if a remote delivery caused a very long error message to be
2720 recorded - for instance if somebody sent an entire SpamAssassin report back
2721 as a large number of 550 error lines. This bug was coincidentally fixed by
2722 increasing the size of one of Exim's internal buffers (big_buffer) that
2723 happened as part of the Exiscan merge. However, to be on the safe side, I
2724 have made the code more robust (and fixed the comments that describe what
2727 33. Some experimental protocols are using DNS PTR records for new purposes. The
2728 keys for these records are domain names, not reversed IP addresses. The
2729 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2730 leaves it alone. Component reversal etc. now happens only for IP addresses.
2733 34. The host_aton() function is supposed to be passed a string that is known
2734 to be a valid IP address. However, in the case of IPv6 addresses, it was
2735 not checking this. This is a hostage to fortune. Exim now panics and dies
2736 if the condition is not met. A case was found where this could be provoked
2737 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2738 components; fortuitously, this particular loophole had already been fixed
2739 by change 4.50/55 or 4.44/33 above.
2741 If there are any other similar loopholes, the new check in host_aton()
2742 itself should stop them being exploited. The report I received stated that
2743 data on the command line could provoke the exploit when Exim was running as
2744 exim, but did not say which command line option was involved. All I could
2745 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2746 running as the user.
2749 35. There was a buffer overflow vulnerability in the SPA authentication code
2750 (which came originally from the Samba project). I have added a test to the
2751 spa_base64_to_bits() function which I hope fixes it.
2754 36. The daemon start-up calls getloadavg() while still root for those OS that
2755 need the first call to be done as root, but it missed one case: when
2756 deliver_queue_load_max is set with deliver_drop_privilege. This is
2757 necessary for the benefit of the queue runner, because there is no re-exec
2758 when deliver_drop_privilege is set.
2760 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2761 was in use, was not putting the data itself into the right store pool;
2762 consequently, it could be overwritten for a subsequent message in the same
2763 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2766 38. Sometimes the final signoff response after QUIT could fail to get
2767 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2768 < 0 before doing a fflush(). This bug looks as though it goes back to the
2769 introduction of TLS in release 3.20, but "sometimes" must have been rare
2770 because the tests only now provoked it.
2776 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2777 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2778 Now it returns an integer. A number of calls were still expecting a T/F
2779 return. Fortuitously, in all cases, the tests worked in OK situations,
2780 which is the norm. However, things would have gone wrong on any write
2781 failures on the smtp file descriptor. This function is used when sending
2782 messages over SMTP and also when doing verify callouts.
2784 2. When Exim is called to do synchronous delivery of a locally submitted
2785 message (the -odf or -odi options), it no longer closes stderr before doing
2788 3. Implemented the mua_wrapper option.
2790 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2792 5. Implemented the functions header_remove(), header_testname(),
2793 header_add_at_position(), and receive_remove_recipient(), and exported them
2796 6. If an ACL "warn" statement specified the addition of headers, Exim already
2797 inserted X-ACL-Warn: at the start if there was no header name. However, it
2798 was not making this test for the second and subsequent header lines if
2799 there were newlines in the string. This meant that an invalid header could
2800 be inserted if Exim was badly configured.
2802 7. Allow an ACL "warn" statement to add header lines at the start or after all
2803 the Received: headers, as well as at the end.
2805 8. Added the rcpt_4xx retry error code.
2807 9. Added postmaster_mailfrom=xxx to callout verification option.
2809 10. Added mailfrom=xxxx to the callout verification option, for verify=
2812 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2813 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2815 12. Inserted some casts to stop certain compilers warning when using pointer
2816 differences as field lengths or precisions in printf-type calls (mostly
2817 affecting debugging statements).
2819 13. Added optional readline() support for -be (dynamically loaded).
2821 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2822 same clock tick as a message's arrival, so that its received time was the
2823 same as the "first fail" time on the retry record, and that message
2824 remained on the queue past the ultimate address timeout, every queue runner
2825 would try a delivery (because it was past the ultimate address timeout) but
2826 after another failure, the ultimate address timeout, which should have then
2827 bounced the address, did not kick in. This was a "< instead of <=" error;
2828 in most cases the first failure would have been in the next clock tick
2829 after the received time, and all would be well.
2831 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2832 being recognized when the domain list was tested by the match_domain
2833 condition in an expansion string.
2835 16. Added the ${str2b64: operator.
2837 17. Exim was always calling setrlimit() to set a large limit for the number of
2838 processes, without checking whether the existing limit was already
2839 adequate. (It did check for the limit on file descriptors.) Furthermore,
2840 errors from getrlimit() and setrlimit() were being ignored. Now they are
2841 logged to the main and panic logs, but Exim does carry on, to try to do its
2842 job under whatever limits there are.
2844 18. Imported PCRE 5.0.
2846 19. Trivial typo in log message " temporarily refused connection" (the leading
2849 20. If the log selector return_path_on_delivery was set and an address was
2850 redirected to /dev/null, the delivery process crashed because it assumed
2851 that a return path would always be set for a "successful" delivery. In this
2852 case, the whole delivery is bypassed as an optimization, and therefore no
2855 21. Internal re-arrangement: the function for sending a challenge and reading
2856 a response while authentication was assuming a zero-terminated challenge
2857 string. It's now changed to take a pointer and a length, to allow for
2858 binary data in such strings.
2860 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2862 23. Exim was not respecting finduser_retries when seeking the login of the
2863 uid under which it was called; it was always trying 10 times. (The default
2864 setting of finduser_retries is zero.) Also, it was sleeping after the final
2865 failure, which is pointless.
2867 24. Implemented tls_on_connect_ports.
2869 25. Implemented acl_smtp_predata.
2871 26. If the domain in control=submission is set empty, Exim assumes that the
2872 authenticated id is a complete email address when it generates From: or
2873 Sender: header lines.
2875 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2876 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2877 chown and chgrp in /bin and hostname in /usr/bin.
2879 28. Exim was keeping the "process log" file open after each use, just as it
2880 does for the main log. This opens the possibility of it remaining open for
2881 long periods when the USR1 signal hits a daemon. Occasional processlog
2882 errors were reported, that could have been caused by this. Anyway, it seems
2883 much more sensible not to leave this file open at all, so that is what now
2886 29. The long-running daemon process does not normally write to the log once it
2887 has entered its main loop, and it closes the log before doing so. This is
2888 so that log files can straightforwardly be renamed and moved. However,
2889 there are a couple of unusual error situations where the daemon does write
2890 log entries, and I had neglected to close the log afterwards.
2892 30. The text of an SMTP error response that was received during a remote
2893 delivery was being truncated at 512 bytes. This is too short for some of
2894 the long messages that one sometimes sees. I've increased the limit to
2897 31. It is now possible to make retry rules that apply only when a message has a
2898 specific sender, in particular, an empty sender.
2900 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2901 it possible to be selective about when SMTP synchronization is enforced.
2903 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2905 32. Implemented hosts_connection_nolog.
2907 33. Added an ACL for QUIT.
2909 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2912 35. Added mailbox_size and mailbox_filecount to appendfile.
2914 36. Added control = no_multiline_responses to ACLs.
2916 37. There was a bug in the logic of the code that waits for the clock to tick
2917 in the case where the clock went backwards by a substantial amount such
2918 that the microsecond fraction of "now" was more than the microsecond
2919 fraction of "then" (but the whole seconds number was less).
2921 38. Added support for the libradius Radius client library this is found on
2922 FreeBSD (previously only the radiusclient library was supported).
2928 1. When certain lookups returned multiple values in the form name=value, the
2929 quoting of the values was not always being done properly. Specifically:
2930 (a) If the value started with a double quote, but contained no whitespace,
2932 (b) If the value contained whitespace other than a space character (i.e.
2933 tabs or newlines or carriage returns) it was not quoted.
2934 This fix has been applied to the mysql and pgsql lookups by writing a
2935 separate quoting function and calling it from the lookup code. The fix
2936 should probably also be applied to nisplus, ibase and oracle lookups, but
2937 since I cannot test any of those, I have not disturbed their existing code.
2939 2. A hit in the callout cache for a specific address caused a log line with no
2940 reason for rejecting RCPT. Now it says "Previous (cached) callout
2941 verification failure".
2943 3. There was an off-by-one bug in the queryprogram router. An over-long
2944 return line was truncated at 256 instead of 255 characters, thereby
2945 overflowing its buffer with the terminating zero. As well as fixing this, I
2946 have increased the buffer size to 1024 (and made a note to document this).
2948 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2949 when Exim is waiting for an SMTP response from a remote server, Exim
2950 restarts its select() call on the socket, thereby resetting its timeout.
2951 This is not a problem when such interrupts are rare. Somebody set up a cron
2952 job to run exiwhat every 2 minutes, which is less than the normal select()
2953 timeout (5 or 10 minutes). This meant that the select() timeout never
2954 kicked in because it was always reset. I have fixed this by comparing the
2955 time when an interrupt arrives with the time at the start of the first call
2956 to select(). If more time than the timeout has elapsed, the interrupt is
2957 treated as a timeout.
2959 5. Some internal re-factoring in preparation for the addition of Sieve
2960 extensions (by MH). In particular, the "personal" test is moved to a
2961 separate function, and given an option for scanning Cc: and Bcc: (which is
2962 not set for Exim filters).
2964 6. When Exim created an email address using the login of the caller as the
2965 local part (e.g. when creating a From: or Sender: header line), it was not
2966 quoting the local part when it contained special characters such as @.
2968 7. Installed new OpenBSD configuration files.
2970 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2971 try to make them clearer.
2973 9. Callout options, other than the timeout value, were being ignored when
2974 verifying sender addresses in header lines. For example, when using
2976 verify = header_sender/callout=no_cache
2978 the cache was (incorrectly) being used.
2980 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2981 only the Cygwin environment.
2983 11. When return_path_on_delivery was set as a log selector, if different remote
2984 addresses in the same message used different return paths and parallel
2985 remote delivery occurred, the wrong values would sometimes be logged.
2986 (Whenever a remote delivery process finished, the return path value from
2987 the most recently started remote delivery process was logged.)
2989 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2990 lines when AUTH and/or TLS are in use. This is the "received protocol"
2991 field. Exim used to use "asmtp" for authenticated SMTP, without any
2992 indication (in the protocol name) for TLS use. Now it follows the RFC and
2993 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2994 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2995 names appear in log lines as well as in Received: header lines.
2997 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2998 extensions, and comparison tests, and to fix some bugs.
3000 14. Changes to the "personal" filter test:
3002 (1) The test was buggy in that it was just doing the equivalent of
3003 "contains" tests on header lines. For example, if a user's address was
3004 anne@some.where, the "personal" test would incorrectly be true for
3006 To: susanne@some.where
3008 This test is now done by extracting each address from the header in turn,
3009 and checking the entire address. Other tests that are part of "personal"
3010 are now done using regular expressions (for example, to check local parts
3011 of addresses in From: header lines).
3013 (2) The list of non-personal local parts in From: addresses has been
3014 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3015 taken from the Sieve specification recommendations.
3017 (3) If the message contains any header line starting with "List-" it is
3018 treated as non-personal.
3020 (4) The test for "circular" in the Subject: header line has been removed
3021 because it now seems ill-conceived.
3023 15. Minor typos in src/EDITME comments corrected.
3025 16. Installed latest exipick from John Jetmore.
3027 17. If headers_add on a router specified a text string that was too long for
3028 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3029 of string_sprintf() is now avoided.
3031 18. $message_body_size was not set (it was always zero) when running the DATA
3032 ACL and the local_scan() function.
3034 19. For the "mail" command in an Exim filter, no default was being set for
3035 the once_repeat time, causing a random time value to be used if "once" was
3036 specified. (If the value happened to be <= 0, no repeat happened.) The
3037 default is now 0s, meaning "never repeat". The "vacation" command was OK
3038 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3039 (I found it when inspecting the code).
3041 20. There is now an overall timeout for performing a callout verification. It
3042 defaults to 4 times the callout timeout, which applies to individual SMTP
3043 commands during the callout. The overall timeout applies when there is more
3044 than one host that can be tried. The timeout is checked before trying the
3045 next host. This prevents very long delays if there are a large number of
3046 hosts and all are timing out (e.g. when the network connections are timing
3047 out). The value of the overall timeout can be changed by specifying an
3048 additional sub-option for "callout", called "maxwait". For example:
3050 verify = sender/callout=5s,maxwait=20s
3052 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3053 to the end before writing, but this should make it even safer).
3055 22. Exim was forgetting that it had advertised PIPELINING for the second and
3056 subsequent messages on an SMTP connection. It was also not resetting its
3057 memory on STARTTLS and an internal HELO.
3059 23. When Exim logs an SMTP synchronization error within a session, it now
3060 records whether PIPELINING has been advertised or not.
3062 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3063 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3064 rather than long int.
3066 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3068 26. Added the never_mail option to autoreply.
3074 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3075 crash if the getsockname() call failed; this can happen if a connection is
3076 closed very soon after it is established. The problem was simply in the
3077 order in which certain operations were done, causing Exim to try to write
3078 to the SMTP stream before it had set up the file descriptor. The bug has
3079 been fixed by making things happen in the correct order.
3085 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3086 before the connection was closed, thus losing the rejection response.
3088 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3089 some early Solaris releases, but causes trouble in current releases where
3090 socklen_t is defined.
3092 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3095 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3098 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3099 that releasing the top of it at the end releases what was used for sub-
3100 expansions (unless the block got too big). However, discard this block if
3101 the first thing is a variable or header, so that we can use its block when
3102 it is dynamic (useful for very large $message_headers, for example).
3104 6. Lookups now cache *every* query, not just the most recent. A new, separate
3105 store pool is used for this. It can be recovered when all lookup caches are
3106 flushed. Lookups now release memory at the end of their result strings.
3107 This has involved some general refactoring of the lookup sources.
3109 7. Some code has been added to the store_xxx() functions to reduce the amount
3110 of flapping under certain conditions.
3112 8. log_incoming_interface used to affect only the <= reception log lines. Now
3113 it causes the local interface and port to be added to several more SMTP log
3114 lines, for example "SMTP connection from", and rejection lines.
3116 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3118 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3120 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3121 could be overwritten at the end of the current message (or the start of a
3122 new message if it was set in a HELO ACL). The value is now preserved for
3123 the duration of the SMTP connection.
3125 12. If a transport had a headers_rewrite setting, and a matching header line
3126 contained an unqualified address, that address was qualified, even if it
3127 did not match any rewriting rules. The underlying bug was that the values
3128 of the flags that permit the existence of unqualified sender and recipient
3129 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3130 non-local messages, and by -bnq for local messages) were not being
3131 preserved with the message after it was received.
3133 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3134 "next input=" as part of the text comprising the host identity instead of
3135 the correct text. The code was using the same buffer for two different
3136 strings. However, depending on which order the printing function evaluated
3137 its arguments, the bug did not always show up. Under Linux, for example, my
3138 test suite worked just fine.
3140 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3141 doesn't work with some older versions of Perl. It has been changed to "my",
3142 which in any case is probably the better facility to use.
3144 15. A really picky compiler found some instances of statements for creating
3145 error messages that either had too many or two few arguments for the format
3148 16. The size of the buffer for calls to the DNS resolver has been increased
3149 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3150 for addresses that have a lot of PTR records. This alleviates a problem; it
3151 does not fully solve it.
3153 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3154 buffer now truncates the list and logs the incident, which is the same
3155 action as happens when Exim is looking up a host name and its aliases.
3156 Previously in this situation something unpredictable would happen;
3157 sometimes it was "internal error: store_reset failed".
3159 18. If a server dropped the connection unexpectedly when an Exim client was
3160 using GnuTLS and trying to read a response, the client delivery process
3161 crashed while trying to generate an error log message.
3163 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3164 string, for example:
3166 warn message = H1: something\nH2: something
3168 the text was added as a single header line from Exim's point of view
3169 though it ended up OK in the delivered message. However, searching for the
3170 second and subsequent header lines using $h_h2: did not work. This has been
3171 fixed. Similarly, if a system filter added multiple headers in this way,
3172 the routers could not see them.
3174 20. Expanded the error message when iplsearch is called with an invalid key to
3175 suggest using net-iplsearch in a host list.
3177 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3178 ACLs are no longer actually imposed (and a message to that effect is
3181 22. If a "gecos" field in a passwd entry contained escaped characters, in
3182 particular, if it contained a \" sequence, Exim got it wrong when building
3183 a From: or a Sender: header from that name. A second bug also caused
3184 incorrect handling when an unquoted " was present following a character
3185 that needed quoting.
3187 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3188 was not being matched caselessly.
3190 24. Arranged for all hyphens in the exim.8 source to be escaped with
3193 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3194 properly. Recipient callout cache records were still being keyed to include
3195 the sender, even when use_sender was set false. This led to far more
3196 callouts that were necessary. The sender is no longer included in the key
3197 when use_sender is false.
3199 26. Added "control = submission" modifier to ACLs.
3201 27. Added the ${base62d: operator to decode base 62 numbers.
3203 28. dnsdb lookups can now access SRV records.
3205 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3206 the configuration file.
3208 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3209 (-v) mode. This makes the output for a verbose queue run more intelligible.
3211 31. Added a use_postmaster feature to recipient callouts.
3213 32. Added the $body_zerocount variable, containing the number of binary zero
3214 bytes in the message body.
3216 33. The time of last modification of the "new" subdirectory is now used as the
3217 "mailbox time last read" when there is a quota error for a maildir
3220 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3222 35. Added +ignore_unknown as a special item in host lists.
3224 36. Code for decoding IPv6 addresses in host lists is now included, even if
3225 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3226 address was recognized as an IP address, but was then not correctly decoded
3227 into binary, causing unexpected and incorrect effects when compared with
3234 1. Very minor rewording of debugging text in manualroute to say "list of
3235 hosts" instead of "hostlist".
3237 2. If verify=header_syntax was set, and a header line with an unqualified
3238 address (no domain) and a large number of spaces between the end of the
3239 name and the colon was received, the reception process suffered a buffer
3240 overflow, and (when I tested it) crashed. This was caused by some obsolete
3241 code that should have been removed. The fix is to remove it!
3243 3. When running in the test harness, delay a bit after writing a bounce
3244 message to get a bit more predictability in the log output.
3246 4. Added a call to search_tidyup() just before forking a reception process. In
3247 theory, someone could use a lookup in the expansion of smtp_accept_max_
3248 per_host which, without the tidyup, could leave open a database connection.
3250 5. Added the variables $recipient_data and $sender_data which get set from a
3251 lookup success in an ACL "recipients" or "senders" condition, or a router
3252 "senders" option, similar to $domain_data and $local_part_data.
3254 6. Moved the writing of debug_print from before to after the "senders" test
3257 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3258 problems for message scanning, either using a data ACL, or using
3259 local_scan() because the Received: header was not generated till after they
3260 were called (in order to set the time as the time of reception completion).
3261 I have revised the way this works. The header is now generated after the
3262 body is received, but before the ACL or local_scan() are called. After they
3263 are run, the timestamp in the header is updated.
3269 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3270 before starting a queue runner without re-exec. This happened only when
3271 deliver_drop_privilege was set or when the Exim user was set to root. The
3272 effect of the bug was that timeouts during subsequent deliveries caused
3273 crashes instead of being properly handled. The handler is now left at its
3274 default (and expected) setting.
3276 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3277 message, again when deliver_drop_privilege is set or Exim is run as root.
3278 The bug described in (1) was not present in this case, but the tidying up
3279 of the other signals was missing. I have made the two cases consistent.
3281 3. The ignore_target_hosts setting on a manualroute router was being ignored
3282 for hosts that were looked up using the /MX notation.
3284 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3287 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3288 operated on the sender address. After changing the $sender_address to <>
3289 for the sender address verify, Exim was re-instated it as the original
3290 (before rewriting) address, but remembering that it had rewritten it, so it
3291 wasn't rewriting it again. This bug also had the effect of breaking the
3292 sender address verification caching when the sender address was rewritten.
3294 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3295 This has been changed so that if the ip literal address matches
3296 ignore_target_hosts, the router declines.
3298 7. Added expansion conditions match_domain, match_address, and match_local_
3299 part (NOT match_host).
3301 8. The placeholder for the Received: header didn't have a length field set.
3303 9. Added code to Exim itself and to exim_lock to test for a specific race
3304 condition that could lead to file corruption when using MBX delivery. The
3305 issue is with the lockfile that is created in /tmp. If this file is removed
3306 after a process has opened it but before that process has acquired a lock,
3307 there is the potential for a second process to recreate the file and also
3308 acquire a lock. This could lead to two Exim processes writing to the file
3309 at the same time. The added code performs the same test as UW imapd; it
3310 checks after acquiring the lock that its file descriptor still refers to
3311 the same named file.
3313 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3314 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3317 11. Added the smtp_active_hostname option. If used, this will typically be made
3318 to depend on the incoming interface address. Because $interface_address is
3319 not set up until the daemon has forked a reception process, error responses
3320 that can happen earlier (such as "too many connections") no longer contain
3323 12. If an expansion in a condition on a "warn" statement fails because a lookup
3324 defers, the "warn" statement is abandoned, and the next ACL statement is
3325 processed. Previously this caused the whole ACL to be aborted.
3327 13. Added the iplsearch lookup type.
3329 14. Added ident_timeout as a log selector.
3331 15. Added tls_certificate_verified as a log selector.
3333 16. Added a global option tls_require_ciphers (compare the smtp transport
3334 option of the same name). This controls incoming TLS connections.
3336 17. I finally figured out how to make tls_require_ciphers do a similar thing
3337 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3338 before starting the TLS session.
3340 18. Tabs are now shown as \t in -bP output.
3342 19. If the log selector return_path_on_delivery was set, Exim crashed when
3343 bouncing a message because it had too many Received: header lines.
3345 20. If two routers both had headers_remove settings, and the first one included
3346 a superfluous trailing colon, the final name in the first list and the
3347 first name in the second list were incorrectly joined into one item (with a
3348 colon in the middle).
3354 1. Added -C and -D options to the exinext utility, mainly to make it easier
3355 to include in the automated testing, but these could be helpful when
3356 multiple configurations are in use.
3358 2. The exinext utility was not formatting the output nicely when there was
3359 an alternate port involved in the retry record key, nor when there was a
3360 message id as well (for retries that were specific to a specific message
3361 and a specific host). It was also confused by IPv6 addresses, because of
3362 the additional colons they contain. I have fixed the IPv4 problem, and
3363 patched it up to do a reasonable job for IPv6.
3365 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3366 delivery, the log line now contains "pipelined" if PIPELINING was used.
3368 4. An SMTP transport process used to panic and die if the bind() call to set
3369 an explicit outgoing interface failed. This has been changed; it is now
3370 treated in the same way as a connect() failure.
3372 5. A reference to $sender_host_name in the part of a conditional expansion
3373 that was being skipped was still causing a DNS lookup. This no longer
3376 6. The def: expansion condition was not recognizing references to header lines
3377 that used bh_ and bheader_.
3379 7. Added the _cache feature to named lists.
3381 8. The code for checking quota_filecount in the appendfile transport was
3382 allowing one more file than it should have been.
3384 9. For compatibility with Sendmail, the command line option
3392 and sets the incoming protocol and host name (for trusted callers). The
3393 host name and its colon can be omitted when only the protocol is to be set.
3394 Note the Exim already has two private options, -pd and -ps, that refer to
3395 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3396 "s", but I don't think that's a major issue.
3398 10. A number of refactoring changes to the code, none of which should affect
3401 (a) The number of logging options was getting close to filling up the
3402 32-bit word that was used as a bit map. I have split them into two classes:
3403 those that are passed in the argument to log_write(), and those that are
3404 only ever tested independently outside of that function. These are now in
3405 separate 32-bit words, so there is plenty of room for expansion again.
3406 There is no change in the user interface or the logging behaviour.
3408 (b) When building, for example, log lines, the code previously used a
3409 macro that called string_cat() twice, in order to add two strings. This is
3410 not really sufficiently general. Furthermore, there was one instance where
3411 it was actually wrong because one of the argument was used twice, and in
3412 one call a function was used. (As it happened, calling the function twice
3413 did not affect the overall behaviour.) The macro has been replaced by a
3414 function that can join an arbitrary number of extra strings onto a growing
3417 (c) The code for expansion conditions now uses a table and a binary chop
3418 instead of a serial search (which was left over from when there were very
3419 few conditions). Also, it now recognizes conditions like "pam" even when
3420 the relevant support is not compiled in: a suitably worded error message is
3421 given if an attempt is made to use such a condition.
3423 11. Added ${time_interval:xxxxx}.
3425 12. A bug was causing one of the ddress fields not to be passed back correctly
3426 from remote delivery subprocesses. The field in question was not being
3427 subsequently used, so this caused to problems in practice.
3429 13. Added new log selectors queue_time and deliver_time.
3431 14. Might have fixed a bug in maildirsizefile handling that threw up
3432 "unexpected character" debug warnings, and recalculated the data
3433 unnecessarily. In any case, I expanded the warning message to give more
3436 15. Added the message "Restricted characters in address" to the statements in
3437 the default ACL that block characters like @ and % in local parts.
3439 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3440 Three changes have been made:
3442 (a) There was a serious bug; a negative response to MAIL caused the whole
3443 recipient domain to be cached as invalid, thereby blocking all messages
3444 to all local parts at the same domain, from all senders. This bug has
3445 been fixed. The domain is no longer cached after a negative response to
3446 MAIL if the sender used is not empty.
3448 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3451 (c) A new callout option, "use_sender" has been added for people who want
3452 the modified behaviour.
3458 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3461 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3462 indeed breaks things for older releases.
3464 3. Added additional logging to the case where there is a problem reading data
3465 from a filter that is running in a subprocess using a pipe, in order to
3466 try to track down a specific problem.
3468 4. Testing facility fudge: when running in the test harness and attempting
3469 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3470 getting "No route to host". Convert this to a timeout.
3472 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3475 6. Some OS don't have socklen_t but use size_t instead. This affects the
3476 fifth argument of getsockopt() amongst other things. This is now
3477 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3478 can be set for individual OS. I have set it for SunOS5, OSF1, and
3479 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3480 some earlier ones do not.
3482 7. Change 4.30/15 was not doing the test caselessly.
3484 8. The standard form for an IPv6 address literal was being rejected by address
3485 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3486 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3487 this, as well as the form without the "IPv6" on the front (but only when
3488 address literals are enabled, of course).
3490 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3492 10. Exim crashed if a message with an empty sender address specified by -f
3493 encountered a router with an errors_to setting. This could be provoked only
3494 by a command such as
3498 where an empty string was supplied; "<>" did not hit this bug.
3500 11. Installed PCRE release 4.5.
3502 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3503 remained set. It is now erased.
3505 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3506 times from message ids (which are base 36 rather than the normal 62).
3508 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3509 were being counted as actual protocol errors, and logged if the log
3510 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3511 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3512 and DATA following a set of rejected RCPTs do not count as protocol errors.
3513 In other words, Exim assumes they were pipelined, though this may not
3514 actually be the case. Of course, in all cases the client gets an
3515 appropriate error code.
3517 15. If a lookup fails in an ACL condition, a message about the failure may
3518 be available; it is used if testing the ACL cannot continue, because most
3519 such messages specify what the cause of the deferral is. However, some
3520 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3521 that caused an old message to be retained and used if a later statement
3522 caused a defer, replacing the real cause of the deferral.
3524 16. If an IP address had so many PTR records that the DNS lookup buffer
3525 was not large enough to hold them, Exim could crash while trying to process
3526 the truncated data. It now detects and logs this case.
3528 17. Further to 4.21/58, another change has been made: if (and only if) the
3529 first line of a message (the first header line) ends with CRLF, a bare LF
3530 in a subsequent header line has a space inserted after it, so as not to
3531 terminate the header.
3533 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3534 unnecessarily, used atoi() instead of strtol(), and didn't check the
3535 termination when getting file sizes from file names by regex.
3537 19. Completely re-implemented the support for maildirsize files, in the light
3538 of a number of problems with the previous contributed implementation
3539 (4.30/29). In particular:
3541 . If the quota is zero, the maildirsize file is maintained, but no quota is
3544 . If the maildir directory does not exist, it is created before any attempt
3545 to write a maildirsize file.
3547 . The quota value in the file is just a cache; if the quota is changed in
3548 the transport, the new value overrides.
3550 . A regular expression is available for excluding directories from the
3553 20. The autoreply transport checks the characters in options that define the
3554 message's headers; it allows continued headers, but it was checking with
3555 isspace() after an embedded newline instead of explicitly looking for a
3558 21. If all the "regular" hosts to which an address was routed had passed their
3559 expiry times, and had not reached their retry times, the address was
3560 bounced, even if fallback hosts were defined. Now Exim should go on to try
3563 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3564 equivalent code in the SMTP transport. Some hosts send humungous responses
3565 to HELO/EHLO, more than 1024 it seems.
3567 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3568 gives compiler warnings in some environments. I've now done it "properly",
3571 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3572 (because of problems with the built-in one) was declared to return uschar *
3573 instead of char *, causing compiler failure.
3575 25. Fixed a file descriptor leak when processing alias/forward files.
3577 26. Fixed a minor format string issue in dbfn.c.
3579 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3581 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3582 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3585 29. When a sender address is verified, it is cached, to save repeating the test
3586 when there is more than one recipient in a message. However, when the
3587 verification involves a callout, it is possible for different callout
3588 options to be set for different recipients. It is too complicated to keep
3589 track of this in the cache, so now Exim always runs a verification when a
3590 callout is required, relying on the callout cache for the optimization.
3591 The overhead is duplication of the address routing, but this should not be
3594 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3595 to be verified with callout=postmaster, and the main callout worked but the
3596 postmaster check failed, the verification correctly failed. However, if a
3597 subsequent RCPT command asked for sender verification *without* the
3598 postmaster check, incorrect caching caused this verification also to fail,
3601 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3602 it was not caching the DNS options (qualify_single, search_parents) that
3603 were used when the lookup failed. A subsequent lookup with different
3604 options therefore always gave the same answer, though there were cases
3605 where it should not have. (Example: a "domains = !$mx_any" option on a
3606 dnslookup router: the "domains" option is always processed without any
3607 widening, but the router might have qualify_single set.) Now Exim uses the
3608 cached value only when the same options are set.
3610 32. Added John Jetmore's "exipick" utility to the distribution.
3612 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3613 than a timeout (e.g. a certificate is required, and is not provided), an
3614 Exim server now closes the connection immediately. Previously it waited for
3615 the client to close - but if the client is SSL, it seems that they each
3616 wait for each other, leading to a delay before one of them times out.
3618 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3619 maintained 0.8.x compatibility because I don't think many are using it, and
3620 it is clearly obsolete.
3622 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3625 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3626 client certificate was expired. A simple patch fixes this, though I don't
3627 understand the full logic of why the verify callback is called multiple
3630 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3631 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3632 which causes problems with some clients (such as the Certicom SSL Plus
3633 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3634 disables the coutermeasure allowing Eudora to connect."
3636 38. Exim was not checking that a write() to a log file succeeded. This could
3637 lead to Bad Things if a log got too big, in particular if it hit a file
3638 size limit. Exim now panics and dies if it cannot write to a log file, just
3639 as it does if it cannot open a log file.
3641 39. Modified OS/Makefile-Linux so that it now contains
3643 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3645 The two -D definitions ensure that Exim is compiled with large file
3646 support, which makes it possible to handle log files that are bigger than
3649 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3650 instance) a domain was checked against a named list that involved a lookup,
3651 causing $domain_data to be set, then another domain was checked against the
3652 same list, then the first domain was re-checked, the value of $domain_data
3653 after the final check could be wrong. In particular, if the second check
3654 failed, it could be set empty. This bug probably also applied to
3657 41. The strip_trailing_dot option was not being applied to the address given
3658 with the -f command-line option.
3660 42. The code for reading a message's header from the spool was incrementing
3661 $received_count, but never initializing it. This meant that the value was
3662 incorrect (doubled) while delivering a message in the same process in which
3663 it was received. In the most common configuration of Exim, this never
3664 happens - a fresh exec is done - but it can happen when
3665 deliver_drop_privilege is set.
3667 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3668 it now includes up to 150 characters of the unexpected data in the log
3671 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3672 and building data strings. The size of both of these buffers was 10 000
3673 bytes - far larger than anybody would *ever* want, thought I. Needless to
3674 say, somebody hit the limit. I have increased the maximum line length to
3675 20 000 and the maximum data length of concatenated lines to 100 000. I have
3676 also fixed two bugs, because there was no checking on these buffers. Tsk,
3677 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3678 buffer is too small.
3680 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3681 lsearch lookups. Now it does.
3683 46. When parsing a route_list item in a manualroute router, a fixed-length
3684 buffer was used for the list of hosts. I made this 1024 bytes long,
3685 thinking that nobody would ever have a list of hosts that long. Wrong.
3686 Somebody had a whole pile of complicated expansion conditions, and the
3687 string was silently truncated, leading to an expansion error. It turns out
3688 that it is easier to change to an unlimited length (owing to other changes
3689 that have happened since this code was originally written) than to build
3690 structure for giving a limitation error. The length of the item that
3691 expands into the list of hosts is now unlimited.
3693 47. The lsearch lookup could not handle data where the length of text line was
3694 more than 4095 characters. Such lines were truncated, leading to shortened
3695 data being returned. It should now handle lines of any length.
3697 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3698 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3701 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3702 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3703 no longer happens in comment lines. A list of replacements is now placed
3704 at the head of all of the source files, except those whose only change is
3705 to replace PERL_COMMAND in the very first #! line.
3707 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3708 messages on the queue, with a bottom-up merge sort, using code contributed
3709 by Michael Haardt. This should make operations like -bp somewhat faster on
3710 large queues. It won't affect queue runners, except when queue_run_in_order
3713 51. Installed eximstats 1.31 in the distribution.
3715 52. Added support for SRV lookups to the dnslookup router.
3717 53. If an ACL referred to $message_body or $message_body_end, the value was not
3718 reset for any messages that followed in the same SMTP session.
3720 54. The store-handling optimization for building very long strings was not
3721 differentiating between the different store pools. I don't think this
3722 actually made any difference in practice, but I've tidied it.
3724 55. While running the routers to verify a sender address, $sender_address
3725 was still set to the sender address. This is wrong, because when routing to
3726 send a bounce to the sender, it would be empty. Therefore, I have changed
3727 it so that, while verifying a sender address, $sender_address is set to <>.
3728 (There is no change to what happens when verifying a recipient address.)
3730 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3731 target A or AAAA records (if not already returned) without resetting the
3732 qualify_single or search_parents options of the DNS resolver. These are
3733 inappropriate in this case because the targets of MX and SRV records must
3734 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3735 target that, when qualified, matched something in the local domain. These
3736 two options are now turned off when doing these lookups.
3738 57. It seems that at least some releases of Reiserfs (which does not have the
3739 concept of a fixed number of inodes) returns zero and not -1 for the
3740 number of available inodes. This interacted badly with check_spool_inodes,
3741 which assumed that -1 was the "no such thing" setting. What I have done is
3742 to check that the total number of inodes is greater than zero before doing
3743 the test of how many are available.
3745 58. When a "warn" ACL statement has a log_message modifier, the message is
3746 remembered, and not repeated. This is to avoid a lot of repetition when a
3747 message has many recipients that cause the same warning to be written.
3748 Howewer, Exim was preserving the list of already written lines for an
3749 entire SMTP session, which doesn't seem right. The memory is now reset if a
3750 new message is started.
3752 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3753 debugging output unless log_rewrite was also set.
3755 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3756 of (int)(handle) when we know that handle contains (void *)(-1).
3758 61. The Exim daemon panic-logs an error return when it closes the incoming
3759 connection. However "connection reset by peer" seems to be common, and
3760 isn't really an error worthy of noting specially, so that particular error
3763 62. When Exim is trying to find all the local interfaces, it used to panic and
3764 die if the ioctl to get the interface flags failed. However, it seems that
3765 on at least one OS (Solaris 9) it is possible to have an interface that is
3766 included in the list of interfaces, but for which you get a failure error
3767 for this call. This happens when the interface is not "plumbed" into a
3768 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3769 failure of the "get flags" call assumes that the interface is down.
3771 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3772 makes life easier for people who are doing arithmetic on fields extracted
3773 from dates, where you often get leading zeros that should not be
3774 interpreted as octal.
3776 64. Added qualify_domain to the redirect router, to override the global
3779 65. If a pathologically long header line contained very many addresses (the
3780 report of this problem mentioned 10 000) and each of them was rewritten,
3781 Exim could use up a very large amount of memory. (It kept on making new
3782 copies of the header line as it rewrote, and never released the old ones.)
3783 At the expense of a bit more processing, the header rewriting function has
3784 been changed so that it no longer eats memory in this way.
3786 66. The generation of the Received: header has been moved from the time that a
3787 message starts to be received, to the time that it finishes. The timestamp
3788 in the Received: header should now be very close to that of the <= log
3789 line. There are two side-effects of this change:
3791 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3792 logged header lines no longer include the local Received: line, because
3793 it has not yet been created. The same applies to a copy of the message
3794 that is returned to a non-SMTP sender when a message is rejected.
3796 (b) When a filter file is tested using -bf, no additional Received: header
3797 is added to the test message. After some thought, I decided that this
3800 This change does not affect the value of $received_for. It is still set
3801 after address rewriting, but before local_scan() is called.
3803 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3805 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3806 gave an unhelpful panic error message, and a defer error. I have managed to
3807 change this behaviour so that it now rejects any supplied certificate,
3808 which seems right, as the list of acceptable certificates is empty.
3810 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3811 gave an unhelpful defer error. I have not managed to make this reject any
3812 supplied certificates, but the error message it gives is "no certificate
3813 supplied", which is not helpful.
3815 70. exigrep's output now also includes lines that are not associated with any
3816 message, but which match the given pattern. Implemented by a patch from
3817 Martin Sluka, which also tidied up the Perl a bit.
3819 71. Recipient callout verification, like sender verification, was using <> in
3820 the MAIL FROM command. This isn't really the right thing, since the actual
3821 sender may affect whether the remote host accepts the recipient or not. I
3822 have changed it to use the actual sender in the callout; this means that
3823 the cache record is now keyed on a recipient/sender pair, not just the
3824 recipient address. There doesn't seem to be a real danger of callout loops,
3825 since a callout by the remote host to check the sender would use <>.
3826 [SEE ABOVE: changed after hitting problems.]
3828 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3829 temporary errors. However, in the case of such a code being given after
3830 the end of a data transmission (i.e. after ".") Exim was failing to write
3831 a retry record for the message. (Yes, there was some broken host that was
3832 actually sending 8xx at this point.)
3834 73. An unknown lookup type in a host list could cause Exim to panic-die when
3835 the list was checked. (An example that provoked this was putting <; in the
3836 middle of a list instead of at the start.) If this happened during a DATA
3837 ACL check, a -D file could be left lying around. This kind of configuration
3838 error no longer causes Exim to die; instead it causes a defer errror. The
3839 incident is still logged to the main and panic logs.
3841 74. Buglet left over from Exim 3 conversion. The message "too many messages
3842 in one connection" was written to the rejectlog but not the mainlog, except
3843 when address rewriting (yes!) was being logged.
3845 75. Added write_rejectlog option.
3847 76. When a system filter was run not as root (that is, when system_filter_user
3848 was set), the values of the $n variables were not being returned to the
3849 main process; thus, they were not subsequently available in the $sn
3852 77. Added +return_path_on_delivery log selector.
3854 78. A connection timeout was being treated differently from recipients deferred
3855 when testing hosts_max_try with a message that was older than the host's
3856 retry timeout. (The host should not be counted, thus allowing all hosts to
3857 be tried at least once before bouncing.) This may have been the cause of an
3858 occasionally reported bug whereby a message would remain on the queue
3859 longer than the retry timeout, but would be bounced if a delivery was
3860 forced. I say "may" because I never totally pinned down the problem;
3861 setting up timeout/retry tests is difficult. See also the next item.
3863 79. The ultimate address timeout was not being applied to errors that involved
3864 a combination of host plus message (for example, a timeout on a MAIL
3865 command). When an address resolved to a number of possible hosts, and they
3866 were not all tried for each delivery (e.g. because of hosts_max_try), a
3867 message could remain on the queue longer than the retry timeout.
3869 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3872 81. Fixed an obscure SMTP outgoing bug which required at least the following
3873 conditions: (a) there was another message waiting for the same server;
3874 (b) the server returned 5xx to all RCPT commands in the first message so
3875 that the message was not completed; (c) the server dropped the connection
3876 or gave a negative response to the RSET that Exim sends to abort the
3877 transaction. The observed case was a dropped connection after DATA that had
3878 been sent in pipelining mode. That is, the server had advertised PIPELINING
3879 but was not implementing it correctly. The effect of the bug was incorrect
3880 behaviour, such as trying another host, and this could lead to a crash.
3886 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3887 and daemon.c were passed as pointers to ints; they should have been
3888 pointers to socklen_t variables (which are typically unsigned ints).
3890 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3893 3. Fixed a really odd bug that affected only the testing scheme; patching a
3894 certain fixed string in the binary changed the value of another string that
3895 happened to be identical to the end of the original first string.
3897 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3898 name", it returns that address as the IP address. On some operating
3899 systems (e.g. Solaris), it also passes back the IP address string as the
3900 "host name". However, on others (e.g. Linux), it passes back an empty
3901 string. Exim wasn't checking for this, and was changing the host name to an
3902 empty string, assuming it had been canonicized.
3904 5. Although rare, it is permitted to have more than one PTR record for a given
3905 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3906 all the names associated with an address, because they do in Solaris.
3907 However, it seems that they do not in Linux for data that comes from the
3908 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3909 I found this out when I moved to a new Linux workstation and tried to run
3910 the Exim test suite.
3912 To get round this problem I have changed the code so that it now does its
3913 own call to the DNS to look up PTR records when searching for a host name.
3914 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3915 addresses that are only in /etc/hosts are still found.
3917 This behaviour is, however, controlled by an option called host_lookup_
3918 order, which defaults to "bydns:byaddr". If people want to use the other
3919 order, or indeed, just use one or the other means of lookup, they can
3920 specify it in this variable.
3922 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3923 some operating systems, this comes back from gethostbyaddr() as an empty
3924 string, and this is what Exim used to test for. However, it seems that in
3925 other systems, "." is yielded. Exim now tests for this case too.
3927 7. The values of check_spool_space and check_log_space are now held internally
3928 as a number of kilobytes instead of an absolute number of bytes. If a
3929 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3930 kilobyte. This means that much larger values can be stored.
3932 8. Exim monitor: an attempt to get the action menu when not actually pointing
3933 at a message produces an empty menu entitled "No message selected". This
3934 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3935 no entries in it ("Shell widget menu has zero width and/or height"). So I
3936 have added a single, blank menu entry in this case.
3938 9. Added ${quote_local_part.
3940 10. MIME decoding is now applied to the contents of Subject: header lines when
3943 11. Now that a reference to $sender_host_address automatically causes a reverse
3944 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3945 host lookup before query-style lookups in lists that might use this
3946 variable. This has therefore been abolished, and the "net-" prefix is no
3947 longer necessary for query-style lookups.
3949 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3950 have been a typo for LFLAGS, so it has been changed.
3952 13. The install script calls Exim with "-C /dev/null" in order to find the
3953 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3954 to be output. Howeve, since Exim outputs its version number before the
3955 error, it didn't break the script. It just looked ugly. I fixed this by
3956 always allowing "-C /dev/null" if the caller is root.
3958 14. Ignore overlarge ACL variable number when reading spool file - insurance
3959 against a later release with more variables having written the file.
3961 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3962 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3963 this, as well as the form without the "IPv6" on the front.
3965 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3966 OS/Makefile-Darwin file.
3968 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3969 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3971 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3972 variable. However, some OS use ints for inodes. I've added cast to long int
3973 to get rid of the compiler warning.
3975 19. I had forgotten to lock out "/../" in configuration file names when
3976 ALT_CONFIG_PREFIX was set.
3978 20. Routers used for verification do not need to specify transports. However,
3979 if such a router generated a host list, and callout was configured, Exim
3980 crashed, because it could not find a port number from the (non-existent)
3981 transport. It now assumes port 25 in this circumstance.
3983 21. Added the -t option to exigrep.
3985 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3986 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3987 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3988 them is set, LOOKUP_LSEARCH is forced.
3990 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3992 24. Added sender and host information to the "rejected by local_scan()" log
3993 line; previously there was no indication of these.
3995 25. Added .include_if_exists.
3997 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3998 Linux. It turns out that not all file systems support this. Apparently some
3999 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4000 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4001 not supported on the file descriptor, is now ignored when Exim is trying to
4002 sync a directory. This applies only to Linux.
4004 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4006 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4007 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4010 29. Added support for maildirsize files from supplied patch (modified a bit).
4012 30. The use of :fail: followed by an empty string could lead Exim to respond to
4013 sender verification failures with (e.g.):
4015 550 Verification failed for <xxx>
4016 550 Sender verify failed
4018 where the first response line was missing the '-' that indicates it is not
4019 the final line of the response.
4021 31. The loop for finding the name of the user that called Exim had a hardwired
4022 limit of 10; it now uses the value of finduser_retries, which is used for
4023 all other user lookups.
4025 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4028 33. Exim was neglecting to zero errno before one call of strtol() when
4029 expanding a string and expecting an integer value. On some systems this
4030 resulted in spurious "integer overflow" errors. Also, it was casting the
4031 result into an int without checking.
4033 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4034 did not work. The code looks as if it has *never* worked, though it appears
4035 to have been documented since at least releast 1.62. I have made it work.
4037 35. The "timeout_DNS" error in retry rules, also documented since at least
4038 1.62, also never worked. As it isn't clear exactly what this means, and
4039 clearly it isn't a major issue, I have abolished the feature by treating it
4040 as "timeout", and writing a warning to the main and panic logs.
4042 36. The display of retry rules for -brt wasn't always showing the error code
4045 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4046 timeout_connect_A, timeout_connect_MX.
4048 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4049 to the empty sender.
4051 39. The daemon was not analysing the content of -oX till after it had closed
4052 stderr and disconnected from the controlling terminal. This meant that any
4053 syntax errors were only noted on the panic log, and the return code from
4054 the command was 0. By re-arranging the code a little, I've made the
4055 decoding happen first, so such errors now appear on stderr, and the return
4056 code is 1. However, the actual setting up of the sockets still happens in
4057 the disconnected process, so errors there are still only recorded on the
4060 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4061 connections (as happens on some IP stacks) was logged at start up time as
4062 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4063 it from "IPv6 and IPv4", which means that two separate sockets are being
4066 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4067 says whether AF_INET or AF_INET6 was passed as an argument.
4069 42. Exiwhat output was messed up when time zones were included in log
4072 43. Exiwhat now gives more information about the daemon's listening ports,
4073 and whether -tls-on-connect was used.
4075 44. The "port" option of the smtp transport is now expanded.
4077 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4078 silently ignored. Now an error message is written to the main and panic
4081 46. There's a new ACL modifier called "logwrite" which writes to a log file
4082 as soon as it is encountered.
4084 47. Added $local_user_uid and $local_user_gid at routing time.
4086 48. Exim crashed when trying to verify a sender address that was being
4089 49. Exim was recognizing only a space character after ".include". It now also
4090 recognizes a tab character.
4092 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4093 extracting the relevant information from the specification. The man page no
4094 longer contains scrambled data for the -d option, and I've added a section
4095 at the front about calling Exim under different names.
4097 51. Added "extra_headers" argument to the "mail" command in filter files.
4099 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4102 53. Installed eximstats 1.29.
4104 54. Added transport_filter_timeout as a generic transport option.
4106 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4107 Cc: header lines. This was required by RFC 822, but it not required by RFC
4110 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4111 incoming messages that did not have them. Now it does so only if the
4112 message originates locally, that is, if there is no associated remote host
4113 address. When Resent- header lines are present, this applies to the Resent-
4114 lines rather than the non-Resent- lines.
4116 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4117 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4119 58. Messages for configuration errors now include the name of the main
4120 configuration file - useful now that there may be more than one file in a
4121 list (.included file names were always shown).
4123 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4124 for those rare installations that do not start the daemon as root or run it
4125 setuid root. I've cut out the call to initgroups() if the daemon is not
4128 60. The Exim user and group can now be bound into the binary as text strings
4129 that are looked up at the start of Exim's processing.
4131 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4133 62. Added $mailstore_basename variable.
4135 63. Installed patch to sieve.c from Michael Haardt.
4137 64. When Exim failed to open the panic log after failing to open the main log,
4138 the original message it was trying to log was written to stderr and debug
4139 output, but if they were not available (the usual case in production), it
4140 was lost. Now it is written to syslog before the two lines that record the
4141 failures to open the logs.
4143 65. Users' Exim filters run in subprocesses under the user's uid. It is
4144 possible for a "deliver" command or an alias in a "personal" command to
4145 provoke an address rewrite. If logging of address rewriting is configured,
4146 this fails because the process is not running as root or exim. There may be
4147 a better way of dealing with this, but for the moment (because 4.30 needs
4148 to be released), I have disabled address rewrite logging when running a
4149 filter in a non-root, non-exim process.
4155 1. The buildconfig auxiliary program wasn't quoting the value set for
4156 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4157 not defined. This bug was masked in 4.22 by the effect that was fixed in
4160 2. Some messages that were rejected after a message id was allocated were
4161 shown as "incomplete" by exigrep. It no longer does this for messages that
4162 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4164 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4165 have allow_domain_literals set, the ID did not get logged in the <= line.
4166 Domain literals are now always recognized in Message-ID: header lines.
4168 4. The first argument for a ${extract expansion item is the key name or field
4169 number. Leading and trailing spaces in this item were not being ignored,
4170 causing some misleading effects.
4172 5. When deliver_drop_privilege was set, single queue runner processes started
4173 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4174 same command in the process it spins off) were not dropping privilege.
4176 6. When the daemon running as "exim" started a queue runner, it always
4177 re-executed Exim in the spun-off process. This is a waste of effort when
4178 deliver_drop_privilege is set. The new process now just calls the
4179 queue-runner function directly.
4185 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4188 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4189 ignored. Though the use of -oP was forcing the writing of a pid file, it
4190 was always written to the default place.
4192 3. If the message "no IP address found for host xxxx" is generated during
4193 incoming verification, it is now followed by identification of the incoming
4194 connection (so you can more easily find what provoked it).
4196 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4198 5. Added some features to "harden" Exim a bit more against certain attacks:
4200 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4201 be put in Local/Makefile. This is like the never_users runtime option,
4202 but it cannot be overridden. The default setting is "root".
4204 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4205 prefix string with which any file named in a -C command line option
4208 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4209 is retained for -C and -D only if the caller of Exim is root. Without
4210 it, the exim user may also use -C and -D and retain privilege.
4212 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4213 command line option is disabled.
4215 6. Macro names set by the -D option must start with an upper case letter, just
4216 like macro names defined in the configuration file.
4218 7. Added "dereference=" facility to LDAP.
4220 8. Two instances of the typo "uknown" in the source files are fixed.
4222 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4223 the Configure-Makefile script screwed up while processing it.
4225 10. Incorporated PCRE 4.4.
4227 11. The SMTP synchronization check was not operating right at the start of an
4228 SMTP session. For example, it could not catch a HELO sent before the client
4229 waited for the greeting. There is now a check for outstanding input at the
4230 point when the greeting is written. Because of the duplex, asynchronous
4231 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4232 way, but not yet received, when the check is performed.
4234 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4235 on TCP/IP sockets, because this apparently causes some broken clients to
4238 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4239 unchanged) from the Cygwin maintainer.
4241 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4242 was supported instead of testing for mbx. Effectively a typo.
4244 15. The spa authenticator server code was not checking that the input it
4245 received was valid base64.
4247 16. The debug output line for the "set" modifier in ACLs was not showing the
4248 name of the variable that was being set.
4250 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4252 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4253 lookup on its own. Something else was needed to trigger the lookup. For
4254 example, a match in host_lookup or the need for a host name in a host list.
4255 Now, if $sender_host_name is referenced and the host name has not yet been
4256 looked up, a lookup is performed. If the lookup fails, the variable remains
4257 empty, and $host_lookup_failed is set to "1".
4259 19. Added "eqi" as a case-independent comparison operator.
4261 20. The saslauthd authentication condition could segfault if neither service
4262 nor realm was specified.
4264 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4265 error message that was logged was misleading, and incoming SMTP
4266 connections were dropped. The message is now more accurate, and temporary
4267 errors are given to SMTP connections.
4269 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4270 (except RSET) with a 421 error, until QUIT is received. However, it was
4271 failing to send a response to QUIT.
4273 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4274 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4275 ACL is now run after the helo_try_verify_hosts code.
4277 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4278 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4279 case-independent, so other case variants are also recognized). Apparently
4280 some systems use these upper case variants.
4282 25. If more than two messages were waiting for the same host, and a transport
4283 filter was specified for the transport, Exim sent two messages over the
4284 same TCP/IP connection, and then failed with "socket operation on non-
4285 socket" when it tried to send the third.
4287 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4289 27. The extern definition of crypt16() in expand.c was not being excluded when
4290 the OS had its own crypt16() function.
4292 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4293 as a preferred synonym for return_size_limit, both as an option and as an
4296 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4298 30. Changed the default configuration ACL to relax the local part checking rule
4299 for addresses that are not in any local domains. For these addresses,
4300 slashes and pipe symbols are allowed within local parts, but the sequence
4301 /../ is explicitly forbidden.
4303 31. SPA server authentication was not clearing the challenge buffer before
4306 32. log_message in a "warn" ACL statement was writing to the reject log as
4307 well as to the main log, which contradicts the documentation and doesn't
4308 seem right (because no rejection is happening). So I have stopped it.
4310 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4311 However, I am unable to do any testing of this.
4313 34. Fixed an infelicity in the appendfile transport. When checking directories
4314 for a mailbox, to see if any needed to be created, it was accidentally
4315 using path names with one or more superfluous leading slashes; tracing
4316 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4318 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4319 contents are added to the log line that is written for every discarded
4320 recipient. (Previously a log_message setting was ignored.)
4322 36. The ${quote: operator now quotes the string if it is empty.
4324 37. The install script runs exim in order to find its version number. If for
4325 some reason other than non-existence or emptiness, which it checks, it
4326 could not run './exim', it was installing it with an empty version number,
4327 i.e. as "exim-". This error state is now caught, and the installation is
4330 38. An argument was missing from the function that creates an error message
4331 when Exim fails to connect to the socket for saslauthd authentication.
4332 This could cause Exim to crash, or give a corrupted message.
4334 39. Added isip, isip4, and isip6 to ${if conditions.
4336 40. The ACL variables $acl_xx are now saved with the message, and can be
4337 accessed later in routers, transports, and filters.
4339 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4340 strings in the file are not string-expanded.
4342 42. If a MAIL command specified a SIZE value that was too large to fit into an
4343 int variable, the check against message_size_limit failed. Such values are
4344 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4345 day this will have to be increased, but I don't think I want to be around
4346 when emails are that large.
4353 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4354 iconv() is not standard in FreeBSD.
4356 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4357 IPv6 enabled. The observed symptom was a segmentation fault on return from
4358 the function os_common_find_running_interfaces() in src/os.c.
4360 3. In the check_special_case() function in daemon.c I had used "errno" as an
4361 argument name, which causes warnings on some systems. This was basically a
4362 typo, since it was named "eno" in the comments!
4364 4. The code that waits for the clock to tick (at a resolution of some fraction
4365 of a second) so as to ensure message-id uniqueness was always waiting for
4366 at least one whole tick, when it could have waited for less. [This is
4367 almost certainly not relevant at current processor speeds, where it is
4368 unlikely to ever wait at all. But we try to future-proof.]
4370 5. The function that sleeps for a time interval that includes fractions of a
4371 second contained a race. It did not block SIGALRM between setting the
4372 timer, and suspending (a couple of lines later). If the interval was short
4373 and the sigsuspend() was delayed until after it had expired, the suspension
4374 never ended. On busy systems this could lead to processes getting stuck for
4377 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4378 process, before it forks any delivery processes. The open lookup caching
4379 mechanism meant that the open file or database connection was passed into
4380 the delivery process. The problem was that delivery processes always tidy
4381 up cached lookup data. This could cause a problem for the next delivery
4382 process started by the queue runner, because the external queue runner
4383 process does not know about the closure. So the next delivery process
4384 still has data in the lookup cache. In the case of a file lookup, there was
4385 no problem because closing a file descriptor in a subprocess doesn't affect
4386 the parent. However, if the lookup was caching a connection to a database,
4387 the connection was closed, and the second delivery process was likely to
4388 see errors such as "PGSQL: query failed: server closed the connection
4389 unexpectedly". The problem has been fixed by closing all cached lookups
4390 in a queue runner before running a delivery process.
4392 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4393 seem to have the "const" qualifier which it has on other OS. I've
4396 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4397 *of the same type* (client or server) with the same public name that an
4398 error should be diagnosed.
4400 9. When Exim looked up a host name for an IP address, but failed to find the
4401 original IP address when looking up the host name (a safety check), it
4402 output the message "<ip address> does not match any IP for NULL", which was
4403 confusing, to say the least. The bug was that the host name should have
4404 appeared instead of "NULL".
4406 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4407 user that is built into the binary, and the -C or -D options is used, root
4408 privilege is dropped before the configuration file is read. In addition,
4409 logging is switched to stderr instead of the normal log files. If the
4410 configuration then re-defines the Exim user, the unprivileged environment
4411 is probably not what is expected, so Exim logs a panic warning message (but
4414 However, if deliver_drop_privilege is set, the unprivileged state may well
4415 be exactly what is intended, so the warning has been cut out in that case,
4416 and Exim is allowed to try to write to its normal log files.
4422 1. smtp_return_error_details was not giving details for temporary sender
4423 or receiver verification errors.
4425 2. Diagnose a configuration error if two authenticators have the same public
4428 3. Exim used not to create the message log file for a message until the first
4429 delivery attempt. This could be confusing when incoming messages were held
4430 for policy or load reasons. The message log file is now created at the time
4431 the message is received, and an initial "Received" line is written to it.
4433 4. The automatically generated man page for command line options had a minor
4434 bug that caused no ill effects; however, a more serious problem was that
4435 the procedure for building the man page automatically didn't always
4436 operate. Consequently, release 4.20 contains an out-of-date version. This
4437 shouldn't happen again.
4439 5. When building Exim with embedded Perl support, the script that builds the
4440 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4441 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4443 6. The freeze_tell option was not being used for messages that were frozen on
4444 arrival, either by an ACL or by local_scan().
4446 7. Added the smtp_incomplete_transaction log selector.
4448 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4449 was accepting AUTH without a new EHLO.
4451 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4452 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4455 10. Logging of TCP/IP connections (when configured) now happens in the main
4456 daemon process instead of the child process, so that the TCP/IP connection
4457 count is more accurate (but it can never be perfect).
4459 11. The use of "drop" in a nested ACL was not being handled correctly in the
4460 outer ACL. Now, if condition failure induced by the nested "drop" causes
4461 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4462 or "require"), the connection is dropped.
4464 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4465 that yield "discard" can now be used with an "accept" or a "discard" verb,
4466 but an error is generated for any others (because I can't see a useful way
4467 to define what should happen).
4469 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4470 are now processed in the same way as lines in the Exim configuration file.
4471 In particular, continuation lines are supported.
4473 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4475 15. Added -ti meaning -t -i.
4477 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4478 domains, and warn by logging if others are found.
4480 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4481 returned from GIFCONF when Exim is trying to find the list of interfaces on
4482 a host. The code in os.c has been modified to copy each ifreq to an aligned
4483 structure in all cases.
4485 Also, in some cases, the returned ifreq's were being copied to a 'struct
4486 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4487 means the last couple of bytes of an IPv6 address could be chopped if the
4488 ifreq contained only a normal sockaddr (14 bytes storage).
4490 18. Named domain lists were not supported in the hosts_treat_as_local option.
4491 An entry such as +xxxx was not recognized, and was treated as a literal
4494 19. Ensure that header lines added by a DATA ACL are included in the reject log
4495 if the ACL subsequently rejects the message.
4497 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4498 MD5 (which is deprecated).
4500 21. When testing a filter file using -bf, Exim was writing a message when it
4501 took the sender from a "From " line in the message, but it was not doing so
4502 when it took $return_path from a Return-Path: header line. It now does.
4504 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4505 with a valid header line field name (a series of printing characters
4506 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4508 23. Changed "disc" in the source to "disk" to conform to the documentation and
4509 the book and for uniformity.
4511 24. Ignore Sendmail's -Ooption=value command line item.
4513 25. When execve() failed while trying to run a command in a pipe transport,
4514 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4515 could be confused with a return value of 69 from the command itself. This
4516 has been changed to 127, the value the shell returns if it is asked to run
4517 a non-existent command. The wording for the related log line suggests a
4518 non-existent command as the problem.
4520 26. If received_header_text expands to an empty string, do not add a Received:
4521 header line to the message. (Well, it adds a token one on the spool, but
4522 marks it "old" so that it doesn't get used or transmitted.)
4524 27. Installed eximstats 1.28 (addition of -nt option).
4526 28. There was no check for failure on the call to getsockname() in the daemon
4527 code. This can fail if there is a shortage of resources on the system, with
4528 ENOMEM, for example. A temporary error is now given on failure.
4530 29. Contrary to the C standard, it seems that in some environments, the
4531 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4532 program. Exim now does this explicitly; it affects the formatting of
4533 timestamps using strftime().
4535 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4536 complaints. I've now initialized all the variables, to avoid this.
4538 32. Header lines added by a system filter were not being "seen" during
4539 transport-time rewrites.
4541 33. The info_callback() function passed to OpenSSL is set up with type void
4542 (*)(SSL *, int, int), as described somewhere. However, when calling the
4543 function (actually a macro) that sets it up, the type void(*)() is
4544 expected. I've put in a cast to prevent warnings from picky compilers.
4546 34. If a DNS black list lookup found a CNAME record, but there were no A
4547 records associated with the domain it pointed at, Exim crashed.
4549 35. If a DNS black list lookup returned more than one A record, Exim ignored
4550 all but the first. It now scans all returned addresses if a particular IP
4551 value is being sought. In this situation, the contents of the
4552 $dnslist_value variable are a list of all the addresses, separated by a
4555 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4556 to accept a host name and all its aliases if the forward lookup for any of
4557 them yielded the IP address of the incoming connection. Now it accepts only
4558 those names whose forward lookup yields the correct IP address. Any other
4559 names are discarded. This closes a loophole whereby a rogue DNS
4560 administrator could create reverse DNS records to break through a
4561 wildcarded host restriction in an ACL.
4563 37. If a user filter or a system filter that ran in a subprocess used any of
4564 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4565 the wrong values were passed to the pipe command ($thisaddress had the
4566 value of $0, $0 had the value of $1, etc). This bug was introduced by
4567 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4569 38. Improved the line breaking for long SMTP error messages from ACLs.
4570 Previously, if there was no break point between 40 and 75 characters, Exim
4571 left the rest of the message alone. Two changes have been made: (a) I've
4572 reduced the minimum length to 35 characters; (b) if it can't find a break
4573 point between 35 and 75 characters, it looks ahead and uses the first one
4574 that it finds. This may give the occasional overlong line, but at least the
4575 remaining text gets split now.
4577 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4578 file descriptors might be low, and that setting 1000 would always raise it.
4579 It turns out that in some environments, the limit is already over 1000 and
4580 that lowering it causes trouble. So now Exim takes care not to decrease it.
4582 40. When delivering a message, the value of $return_path is set to $sender_
4583 address at the start of routing (routers may change the value). By an
4584 oversight, this default was not being set up when an address was tested by
4585 -bt or -bv, which affected the outcome if any router or filter referred to
4588 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4589 log only when "message" or "log_message" are set. However, if one of the
4590 conditions was an address verification, or a call to a nested ACL, the
4591 messages generated by the underlying test were being passed through. This
4592 no longer happens. The underlying message is available in $acl_verify_
4593 message for both "message" and "log_message" expansions, so it can be
4594 passed through if needed.
4596 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4597 new expansion $bh_ to give the encoded byte string without charset
4598 translation. Translation happens only if iconv() is available; HAVE_ICONV
4599 indicates this at build time. HEADERS_CHARSET gives the charset to
4600 translate to; headers_charset can change it in the configuration, and
4601 "headers charset" can change it in an individual filter file.
4603 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4604 that creates RFC 2047 encoded "words" labels them as that charset instead
4605 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4606 expansion operator; (ii) when Exim creates a From: line for a local
4607 message; (iii) when a header line is rewritten to include a "phrase" part.
4609 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4610 buggy, causing it to skip the first lines of messages whose message ID
4611 ended in 'D'. This would not have bitten before Exim release 4.14, because
4612 message IDs were unlikely to end in 'D' before then. The effect was to have
4613 incorrect size information for certain domains.
4615 45. #include "config.h" was missing at the start of the crypt16.c module. This
4616 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4619 46. If there was a timeout during a "random" callout check, Exim treated it as
4620 a failure of the random address, and carried on sending RSET and the real
4621 address. If the delay was just some slowness somewhere, the response to the
4622 original RCPT would be taken as a response to RSET and so on, causing
4623 mayhem of various kinds.
4625 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4626 when I implemented it. It didn't allow for the fact that some option values
4627 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4628 the right test for positive values.
4630 48. Domain names in DNS records are case-independent. Exim always looks them up
4631 in lower case. Some resolvers return domain names in exactly the case they
4632 appear in the zone file, that is, they may contain uppercase letters. Not
4633 all resolvers do this - some return always lower case. Exim was treating a
4634 change of case by a resolver as a change of domain, similar to a widening
4635 of a domain abbreviation. This triggered its re-routing code and so it was
4636 trying to route what was effectively the same domain again. This normally
4637 caused routing to fail (because the router wouldn't handle the domain
4638 twice). Now Exim checks for this case specially, and just changes the
4639 casing of the domain that it ultimately uses when it transmits the message
4642 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4645 50. If a filter generated a file delivery with a non-absolute name (possible if
4646 no home directory exists for the router), the forbid_file option was not
4649 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4650 the existing equality matching.
4652 52. Exim was using ints instead of ino_t variables in some places where it was
4653 dealing with inode numbers.
4655 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4656 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4657 called TMPDIR, and if it finds it is different, it changes its value.
4659 54. The smtp_printf() function is now made available to local_scan() so
4660 additional output lines can be written before returning. There is also an
4661 smtp_fflush() function to enable the detection of a dropped connection.
4662 The variables smtp_input and smtp_batched_input are exported to
4665 55. Changed the default runtime configuration: the message "Unknown user"
4666 has been removed from the ACL, and instead placed on the localuser router,
4667 using the cannot_route_message feature. This means that any verification
4668 failures that generate their own messages won't get overridden. Similarly,
4669 the "Unrouteable address" message that was in the ACL for unverifiable
4670 relay addresses has also been removed.
4672 56. Added hosts_avoid_esmtp to the smtp transport.
4674 57. The exicyclog script was not checking for the esoteric option
4675 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4676 will work only if exicyclog is run under the appropriate euid.
4678 58. Following a discussion on the list, the rules by which Exim recognises line
4679 endings on incoming messages have been changed. The -dropcr and drop_cr
4680 options are now no-ops, retained only for backwards compatibility. The
4681 following line terminators are recognized: LF CRLF CR. However, special
4682 processing applies to CR:
4684 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4685 nor a local message in the state where . is a terminator.
4687 (ii) If a bare CR is encountered in a header line, an extra space is added
4688 after the line terminator so as not to end the header. The reasoning
4689 behind this is that bare CRs in header lines are most likely either
4690 to be mistakes, or people trying to play silly games.
4692 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4693 was being incorrectly given as 18 bytes larger than it should have been.
4694 This is a VOB (very old bug).
4696 60. This may never have affected anything current, but just in case it has:
4697 When the local host is found other than at the start of a list of hosts,
4698 the local host, those with the same MX, and any that follow, are discarded.
4699 When the list in question was part of a longer list of hosts, the following
4700 hosts (not currently being processed) were also being discarded. This no
4701 longer happens. I'm not sure if this situation could ever has previously
4704 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4707 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4708 header. This is something that is recommended in a new Internet Draft, and
4709 is something that is documented as being done by Sendmail. There are two
4710 possible values. For messages generated by the autoreply transport, Exim
4713 Auto-Submitted: auto-replied
4715 whereas for all other generated messages (e.g. bounces) it adds
4717 Auto-Submitted: auto-generated
4719 63. The "personal" condition in filters now includes a test for the
4720 Auto-Submitted: header. If it contains the string "auto-" the message it
4721 not considered personal.
4723 64. Added rcpt_include_affixes as a generic transport option.
4725 65. Added queue_only_override (default true).
4727 66. Added the syslog_duplication option.
4729 67. If what should have been the first header line of a message consisted of
4730 a space followed by a colon, Exim was mis-interpreting it as a header line.
4731 It isn't of course - it is syntactically invalid and should therefore be
4732 treated as the start of the message body. The misbehaviour could have
4733 caused a number of strange effects, including loss of data in subsequent
4734 header lines, and spool format errors.
4736 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4737 client host had authenticated. This control can now be exercised by an ACL
4738 for more flexibility.
4740 69. By default, callouts do not happen when testing with -bh. There is now a
4741 variant, -bhc, which does actually run the callout code, including
4742 consulting and updating the callout cache.
4744 70. Added support for saslauthd authentication, courtesy of Alexander
4747 71. If statvfs() failed on the spool or log directories while checking their
4748 size for availability, Exim confusingly gave the error "space shortage".
4749 Furthermore, in debugging mode it crashed with a floating point exception.
4750 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4751 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4752 serious problem, Exim now writes to the main and panic logs when this
4753 happens, with details of the failure. It then refuses to accept the
4754 incoming message, giving the message "spool directory problem" or "log
4755 directory problem" with a 421 code for SMTP messages.
4757 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4758 0, 1, and 2 exist, because some OS complain for execs without them (see
4759 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4760 descriptors. However, the code omitted to check that the open succeeded,
4761 causing mysterious errors if for some reason the permissions on /dev/null
4762 got screwed. Now Exim writes a message to the main and panic logs, and
4763 bombs out if it can't open /dev/null.
4765 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4766 interact so that it is all more flexible. It is supposed to remain
4767 backwards compatible. Also added extra_local_interfaces.
4769 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4770 to bomb out with an assertion failure - to the client this appears as a
4771 connection drop. This problem occurs in the part of the code that was taken
4772 from the Samba project. Fortunately, the assertion is in a very simple
4773 function, so I have fixed this by reproducing the function inline in the
4774 one place where it is called, and arranging for authentication to fail
4775 instead of killing the process with assert().
4777 75. The SPA client code was not working when the server requested OEM rather
4778 than Unicode encoding.
4780 76. Added code to make require_files with a specific uid setting more usable in
4781 the case where statting the file as root fails - usually a non-root-mounted
4782 NFS file system. When this happens and the failure is EACCES, Exim now
4783 forks a subprocess and does the per-uid checking as the relevant uid.
4785 77. Added process_log_path.
4787 78. If log_file_path was not explicitly set, a setting of check_log_space or
4788 check_log_inodes was ignored.
4790 79. If a space check for the spool or log partitions fails, the incident is now
4791 logged. Of course, in the latter case the data may get lost...
4793 80. Added the %p formatting code to string_format() so that it can be used to
4794 print addresses in debug_print(). Adjusted all the address printing in the
4795 debugging in store.c to use %p rather than %d.
4797 81. There was a concern that a line of code in smtp_in.c could overflow a
4798 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4799 initially expressed, the concern was not well-founded, because trailing
4800 spaces are removed early. However, if the trailing spaces were followed by
4801 a NULL, they did not get removed, so the overflow was possible. Two fixes
4804 (a) I re-wrote the offending code in a cleaner fashion.
4805 (b) If an incoming SMTP command contains a NULL character, it is rejected
4808 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4809 runs initgroups(), so that if the Exim user is in any additional groups,
4810 they will be used during message reception.
4816 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob