1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.502 2007/04/13 15:13:47 ph10 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
10 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
13 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
16 PH/02 In an ACL statement such as
18 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
20 if a client was not listed at all, or was listed with a value other than
21 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
22 the condition was not true (as it should be), so access was not denied.
23 The bug was that the ! inversion was incorrectly passed on to the second
24 item. This has been fixed.
26 PH/03 Added additional dnslists conditions == and =& which are different from
27 = and & when the dns lookup returns more than one IP address.
29 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
30 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
32 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
33 FSYNC, which compiles an option called disable_fsync that allows for
34 bypassing fsync(). The documentation is heavily laced with warnings.
36 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
38 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
39 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
40 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
41 including adding "make clean"; (3) Added -fPIC when compiling the test
42 dynamically loaded module, to get rid of a warning.
44 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
45 message fails, move_frozen_messages = true and ignore_bounce_errors_after
46 = 0s. The bug is otherwise harmless.
48 PH/07 There was a bug in the dovecot authenticator such that the value of
49 $auth1 could be overwritten, and so not correctly preserved, after a
50 successful authentication. This usually meant that the value preserved by
51 the server_setid option was incorrect.
53 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
55 PH/09 Installed PCRE release 7.0.
57 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
58 run for batched SMTP input. It is now run at the start of every message
59 in the batch. While fixing this I discovered that the process information
60 (output by running exiwhat) was not always getting set for -bs and -bS
61 input. This is fixed, and it now also says "batched" for BSMTP.
63 PH/11 Added control=no_pipelining.
65 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
66 patch, slightly modified), and move the expansion of helo_data till after
67 the connection is made in the smtp transport (so it can use these
70 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
72 PH/14 Added log_selector = +pid.
74 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
76 PH/16 Add ${if forany and ${if forall.
78 PH/17 Added dsn_from option to vary the From: line in DSNs.
80 PH/18 Flush SMTP output before performing a callout, unless control =
81 no_callout_flush is set.
83 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
84 was true (the default) a successful delivery failed to delete the retry
85 item, thus causing premature timeout of the address. The bug is now
88 PH/20 Added hosts_avoid_pipelining to the smtp transport.
90 PH/21 Long custom messages for fakedefer and fakereject are now split up
91 into multiline reponses in the same way that messages for "deny" and
92 other ACL rejections are.
94 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
95 with slight modification.
97 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
98 draft, changing the syntax and factoring some duplicate code".
100 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
101 for deliveries of the second and subsequent messages over the same SMTP
104 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
105 ${reduce, with only minor "tidies".
107 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
109 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
110 expansion side effects.
112 PH/27 When a message times out after an over-quota error from an Exim-imposed
113 quota, the bounce message says "mailbox is full". This message was not
114 being given when it was a system quota that was exceeded. It now should
117 MH/03 Made $recipients available in local_scan(). local_scan() already has
118 better access to the recipient list through recipients_list[], but
119 $recipients can be useful in postmaster-provided expansion strings.
121 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
122 in the case of a MAIL command with additional options following the
123 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
124 were accidentally chopped off.
126 PH/29 SMTP synchronization checks are implemented when a command is read -
127 there is a check that no more input is waiting when there shouldn't be
128 any. However, for some commands, a delay in an ACL can mean that it is
129 some time before the response is written. In this time, more input might
130 arrive, invalidly. So now there are extra checks after an ACL has run for
131 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
132 pipelining has not been advertised.
134 PH/30 MH's patch to allow iscntrl() characters to be list separators.
136 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
137 returned in the SMTP response when smtp_return_error_details was false.
140 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
141 instead of the C I/O that was originally supplied, because problems were
144 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
145 Exim which did not show up earlier: it was assuming that a call to
146 SSL_CTX_set_info_callback() might give an error value. In fact, there is
147 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
148 was a macro that became an assignment, so it seemed to work. This has
149 changed to a proper function call with a void return, hence the compile
150 error. Exim's code has been fixed.
152 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
155 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
157 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
159 PH/37 If a message is not accepted after it has had an id assigned (e.g.
160 because it turns out to be too big or there is a timeout) there is no
161 "Completed" line in the log. When some messages of this type were
162 selected by exigrep, they were listed as "not completed". Others were
163 picked up by some special patterns. I have improved the selection
164 criteria to be more general.
166 PH/38 The host_find_failed option in the manualroute router can now be set
167 to "ignore", to completely ignore a host whose IP address cannot be
168 found. If all hosts are ignored, the behaviour is controlled by the new
169 host_all_ignored option.
171 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
172 homing or because of multiple MX records with /mx) generated more than
173 one IP address, and the following item turned out to be the local host,
174 all the secondary addresses of the first item were incorrectly removed
175 from the list, along with the local host and any following hosts (which
176 is what is supposed to happen).
178 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
179 whoever called Exim into the -H file. In the case of the daemon it was
180 behaving confusingly. When first started, it used values for whoever
181 started the daemon, but after a SIGHUP it used the Exim user (because it
182 calls itself on a restart). I have changed the code so that it now always
185 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
186 message are rejected with the same error (e.g. no authentication or bad
187 sender address), and a DATA command is nevertheless sent (as can happen
188 with PIPELINING or a stupid MUA), the error message that was given to the
189 RCPT commands is included in the rejection of the DATA command. This is
190 intended to be helpful for MUAs that show only the final error to their
193 PH/42 Another patch from the Sieve maintainer.
195 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
196 Eximstats - Fixed some broken HTML links and added missing column headers
198 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
199 columns for Rejects, Temp Rejects, Ham, and Spam rows.
201 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
203 PH/43 Yet another patch from the Sieve maintainer.
205 PH/44 I found a way to check for a TCP/IP connection going away before sending
206 the response to the final '.' that terminates a message. At this time,
207 there should not be any pending input - the client should be waiting for
208 the response. Therefore, a select() can be used: if it shows that the
209 input is "ready", there is either input waiting, or the socket has been
210 closed. Both cases are errors. (It's a bit more complicated than this
211 because of buffering, but that's the essence of it.) Previously, Exim
212 would have sent an OK response which the client would never have see.
213 This could lead to message repetition. This fix should cure that.
219 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
220 fixed by 4.65/MH/01 (is this a record?) are fixed:
222 (i) An empty string was always treated as zero by the numeric comparison
223 operators. This behaviour has been restored.
225 (ii) It is documented that the numeric comparison operators always treat
226 their arguments as decimal numbers. This was broken in that numbers
227 starting with 0 were being interpreted as octal.
229 While fixing these problems I realized that there was another issue that
230 hadn't been noticed. Values of message_size_limit (both the global option
231 and the transport option) were treated as octal if they started with 0.
232 The documentation was vague. These values are now always treated as
233 decimal, and I will make that clear in the documentation.
239 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
240 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
243 MH/01 Don't check that the operands of numeric comparison operators are
244 integers when their expansion is in "skipping" mode (fixes bug
245 introduced by 4.64-PH/07).
247 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
248 child addresses, Exim now panics and dies. Previously, because the count
249 is held in a short int, deliveries were likely to be lost. As such a
250 large number of recipients for a single message is ridiculous
251 (performance will be very, very poor), I have chosen to impose a limit
252 rather than extend the field.
258 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
259 leftover -K file (the existence of which was triggered by #402).
260 While we were at it, introduced process PID as part of the -K
261 filename. This should rule out race conditions when creating
264 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
265 processing considerably. Previous code took too long for large mails,
266 triggering a timeout which in turn triggers #401.
268 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
269 in the DK code in transports.c. sendfile() is not really portable,
270 hence the _LINUX specificness.
272 TF/01 In the add_headers option to the mail command in an Exim filter,
273 there was a bug that Exim would claim a syntax error in any
274 header after the first one which had an odd number of characters
277 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
278 callout verification, Exim cached a "reject" for the entire domain. This
279 is correct for most verifications, but it is not correct for a recipient
280 verification with use_sender or use_postmaster set, because in that case
281 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
282 case of MAIL FROM:<> rejection from other early rejections (e.g.
283 rejection of HELO). When verifying a recipient using a non-null MAIL
284 address, the cache is ignored if it shows MAIL FROM:<> rejection.
285 Whatever the result of the callout, the value of the domain cache is
286 left unchanged (for any other kind of callout, getting as far as trying
287 RCPT means that the domain itself is ok).
289 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
292 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
293 manifest itself as EPIPE rather than ECONNECT. When tidying away a
294 session, the daemon ignores ECONNECT errors and logs others; it now
295 ignores EPIPE as well.
297 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
298 (quoted-printable decoding).
300 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
301 later the small subsequent patch to fix an introduced bug.
303 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
305 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
307 PH/08 An error is now given if message_size_limit is specified negative.
309 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
310 to be given (somewhat) arbitrary names.
312 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
315 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
316 miscellaneous code fixes
318 PH/10 Added the log_reject_target ACL modifier to specify where to log
321 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
322 hostname. This is wrong, because it relates to the incoming message (and
323 probably the interface on which it is arriving) and not to the outgoing
324 callout (which could be using a different interface). This has been
325 changed to use the value of the helo_data option from the smtp transport
326 instead - this is what is used when a message is actually being sent. If
327 there is no remote transport (possible with a router that sets up host
328 addresses), $smtp_active_hostname is used.
330 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
331 tweaks were necessary in order to get it to work (see also 21 below):
332 (a) The code assumed that strncpy() returns a negative number on buffer
333 overflow, which isn't the case. Replaced with Exim's string_format()
335 (b) There were several signed/unsigned issues. I just did the minimum
336 hacking in of casts. There is scope for a larger refactoring.
337 (c) The code used strcasecmp() which is not a standard C function.
338 Replaced with Exim's strcmpic() function.
339 (d) The code set only $1; it now sets $auth1 as well.
340 (e) A simple test gave the error "authentication client didn't specify
341 service in request". It would seem that Dovecot has changed its
342 interface. Fortunately there's a specification; I followed it and
343 changed what the client sends and it appears to be working now.
345 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
348 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
349 address A is aliased to B and C, where B exists and C does not. Without
350 -v the output is "A verified" because verification stops after a
351 successful redirection if more than one address is generated. However,
352 with -v the child addresses are also verified. Exim was outputting "A
353 failed to verify" and then showing the successful verification for C,
354 with its parentage. It now outputs "B failed to verify", showing B's
355 parentage before showing the successful verification of C.
357 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
358 look up a TXT record in a specific list after matching in a combined
361 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
362 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
363 they consult the DNS. I had assumed they would set it the way they
364 wanted; and indeed my experiments on Linux seem to show that in some
365 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
366 To be on the safe side, however, I have now made the interface to
367 host_find_byname() similar to host_find_bydns(), with an argument
368 containing the DNS resolver options. The host_find_byname() function now
369 sets these options at its start, just as host_find_bydns() does. The smtp
370 transport options dns_qualify_single and dns_search_parents are passed to
371 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
372 of host_find_byname() use the default settings of RES_DEFNAMES
373 (qualify_single) but not RES_DNSRCH (search_parents).
375 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
376 spool_read_header() do less string testing, by means of a preliminary
377 switch on the second character of optional "-foo" lines. (This is
378 overdue, caused by the large number of possibilities that now exist.
379 Originally there were few.) While I was there, I also converted the
380 str(n)cmp tests so they don't re-test the leading "-" and the first
381 character, in the hope this might squeeze out yet more improvement.
383 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
384 flag allowing group syntax was set by the header_syntax check but not
385 turned off, possible causing trouble later; (2) The flag was not being
386 set at all for the header_verify test, causing "group"-style headers to
387 be rejected. I have now set it in this case, and also caused header_
388 verify to ignore an empty address taken from a group. While doing this, I
389 came across some other cases where the code for allowing group syntax
390 while scanning a header line wasn't quite right (mostly, not resetting
391 the flag correctly in the right place). These bugs could have caused
392 trouble for malformed header lines. I hope it is now all correct.
394 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
395 with the "reply" argument non-NULL. The code, however (which originally
396 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
397 but it didn't always do it. This confused somebody who was copying the
398 code for some other use. I have removed all the tests.
400 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
401 feature that was used to support insecure browsers during the U.S. crypto
402 embargo. It requires special client support, and Exim is probably the
403 only MTA that supported it -- and would never use it because real RSA is
404 always available. This code has been removed, because it had the bad
405 effect of slowing Exim down by computing (never used) parameters for the
406 RSA_EXPORT functionality.
408 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
409 authenticator to fail if there's a tab character in the incoming data
410 (there should never be unless someone is messing about, as it's supposed
411 to be base64-encoded). Also added, on Timo's advice, the "secured" option
412 if the connection is using TLS or if the remote IP is the same as the
413 local IP, and the "valid-client-cert option" if a client certificate has
416 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
417 authenticators. This can be used for authorization after authentication
418 succeeds. (In the case of plaintext, it servers for both authentication
421 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
422 if any retry times were supplied.
424 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
425 connection, where there is no client IP address to check. In this
426 situation, the verify now always succeeds.
428 PH/25 Applied John Jetmore's -Mset patch.
430 PH/26 Added -bem to be like -Mset, but loading a message from a file.
432 PH/27 In a string expansion for a processed (not raw) header when multiple
433 headers of the same name were present, leading whitespace was being
434 removed from all of them, but trailing whitespace was being removed only
435 from the last one. Now trailing whitespace is removed from each header
436 before concatenation. Completely empty headers in a concatenation (as
439 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
440 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
442 PH/29 [Removed. This was a change that I later backed out, and forgot to
443 correct the ChangeLog entry (that I had efficiently created) before
444 committing the later change.]
446 PH/30 Exim was sometimes attempting to deliver messages that had suffered
447 address errors (4xx response to RCPT) over the same connection as other
448 messages routed to the same hosts. Such deliveries are always "forced",
449 so retry times are not inspected. This resulted in far too many retries
450 for the affected addresses. The effect occurred only when there were more
451 hosts than the hosts_max_try setting in the smtp transport when it had
452 the 4xx errors. Those hosts that it had tried were not added to the list
453 of hosts for which the message was waiting, so if all were tried, there
454 was no problem. Two fixes have been applied:
456 (i) If there are any address or message errors in an SMTP delivery, none
457 of the hosts (tried or untried) are now added to the list of hosts
458 for which the message is waiting, so the message should not be a
459 candidate for sending over the same connection that was used for a
460 successful delivery of some other message. This seems entirely
461 reasonable: after all the message is NOT "waiting for some host".
462 This is so "obvious" that I'm not sure why it wasn't done
463 previously. Hope I haven't missed anything, but it can't do any
464 harm, as the worst effect is to miss an optimization.
466 (ii) If, despite (i), such a delivery is accidentally attempted, the
467 routing retry time is respected, so at least it doesn't keep
468 hammering the server.
470 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
471 in ${readsocket because some servers need this prod.
473 PH/32 Added some extra debug output when updating a wait-xxx database.
475 PH/33 The hint "could be header name not terminated by colon", which has been
476 given for certain expansion errors for a long time, was not being given
477 for the ${if def:h_colon_omitted{... case.
479 PH/34 The spec says: "With one important exception, whenever a domain list is
480 being scanned, $domain contains the subject domain." There was at least
481 one case where this was not true.
483 PH/35 The error "getsockname() failed: connection reset by peer" was being
484 written to the panic log as well as the main log, but it isn't really
485 panic-worthy as it just means the connection died rather early on. I have
486 removed the panic log writing for the ECONNRESET error when getsockname()
489 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
490 runs only) independently of the message's sender address. This meant
491 that, if the 4xx error was in fact related to the sender, a different
492 message to the same recipient with a different sender could confuse
493 things. In particualar, this can happen when sending to a greylisting
494 server, but other circumstances could also provoke similar problems.
495 I have changed the default so that the retry time for these errors is now
496 based a combination of the sender and recipient addresses. This change
497 can be overridden by setting address_retry_include_sender=false in the
500 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
501 remote server are returned as part of bounce messages. This was not
502 happening for LMTP over a pipe (the lmtp transport), but now it is the
503 same for both kinds of LMTP.
505 PH/38 Despite being documented as not happening, Exim was rewriting addresses
506 in header lines that were in fact CNAMEs. This is no longer the case.
508 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
509 and queue runs started by the daemon processed all messages. This has
510 been fixed so that -R and -S can now usefully be given with -q<time>.
512 PH/40 Import PCRE release 6.7 (fixes some bugs).
514 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
516 PH/42 Give an error if -q is specified more than once.
518 PH/43 Renamed the variables $interface_address and $interface_port as
519 $received_ip_address and $received_port, to make it clear that these
520 values apply to message reception, and not to the outgoing interface when
521 a message is delivered. (The old names remain recognized, of course.)
523 PH/44 There was no timeout on the connect() call when using a Unix domain
524 socket in the ${readsocket expansion. There now is.
526 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
527 be meaningful with "accept".
529 SC/01 Eximstats V1.43
530 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
532 SC/02 Eximstats V1.44
533 Use a glob alias rather than an array ref in the generated
534 parser. This improves both readability and performance.
536 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
537 Collect SpamAssassin and rejection statistics.
538 Don't display local sender or destination tables unless
539 there is data to show.
540 Added average volumes into the top table text output.
542 SC/04 Eximstats V1.46
543 Collect data on the number of addresses (recipients)
544 as well as the number of messages.
546 SC/05 Eximstats V1.47
547 Added 'Message too big' to the list of mail rejection
548 reasons (thanks to Marco Gaiarin).
550 SC/06 Eximstats V1.48
551 Mainlog lines which have GMT offsets and are too short to
552 have a flag are now skipped.
554 SC/07 Eximstats V1.49 (Alain Williams)
555 Added the -emptyok flag.
557 SC/08 Eximstats V1.50
558 Fixes for obtaining the IP address from reject messages.
560 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
561 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
562 whitesspace changes from 4.64-PH/27
564 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
567 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
568 are found, allow negative numbers in numeric criteria)
570 JJ/06 exipick.20061117.2, added new $message_body_missing variable
572 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
573 to match changes made in 4.64-PH/43
575 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
577 PH/47 Put in an explicit test for a DNS lookup of an address record where the
578 "domain" is actually an IP address, and force a failure. This locks out
579 those revolvers/nameservers that support "A-for-A" lookups, in
580 contravention of the specifications.
582 PH/48 When a host name was looked up from an IP address, and the subsequent
583 forward lookup of the name timed out, the host name was left in
584 $sender_host_name, contrary to the specification.
586 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
587 restricted to single-key lookups, Exim was not diagnosing an error if
588 * or *@ was used with a query-style lookup.
590 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
592 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
593 long ago, but noone interested enough thought of it. Let's just say that
594 the "1.1" means that there are some new functions that weren't there at
595 some point in the past.
597 PH/51 Error processing for expansion failure of helo_data from an smtp
598 transport during callout processing was broken.
600 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
601 tested/used via the -bh/-bhc/-bs options.
603 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
604 bug, fixed in subsequent PCRE releases).
606 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
607 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
609 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
615 SC/01 Use a glob alias rather than an array ref in eximstats generated
616 parser. This improves both readability and performance.
618 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
619 Don't display local sender or destination tables in eximstats unless
620 there is data to show.
621 Added average volumes into the eximstats top table text output.
623 SC/03 Collect data on the number of addresses (recipients) as well
624 as the number of messages in eximstats.
626 TF/01 Correct an error in the documentation for the redirect router. Exim
627 does (usually) call initgroups() when daemonizing.
629 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
630 with consistent privilege compared to when running as a daemon.
632 TF/03 Note in the spec that $authenticated_id is not set for local
633 submissions from trusted users.
635 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
636 Thanks to Dean Brooks <dean@iglou.com> for the patch.
638 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
639 by adding some example configuration directives to the default
640 configuration file. A little bit of work is required to uncomment the
641 directives and define how usernames and passwords are checked, but
642 there is now a framework to start from.
644 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
645 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
646 without this. I don't know how relevant this is to other LDAP libraries.
648 PH/02 Add the verb name to the "unknown ACL verb" error.
650 PH/03 Magnus Holmgren's patch for filter_prepend_home.
652 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
654 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
655 directory not expanded when it should be if an expanded home directory
656 was set for the address (which is overridden by the transport).
658 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
661 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
662 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
663 because it is too late at that time, and has no effect.
665 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
666 security issue with \' (bugzilla #107). I could not use the
667 PQescapeStringConn() function, because it needs a PGconn value as one of
670 PH/08 When testing addresses using -bt, indicate those final addresses that
671 are duplicates that would not cause an additional delivery. At least one
672 person was confused, thinking that -bt output corresponded to deliveries.
673 (Suppressing duplicates isn't a good idea as you lose the information
674 about possibly different redirections that led to the duplicates.)
676 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
677 systems where poll() doesn't work, in particular OS X.
679 PH/10 Added more information to debugging output for retry time not reached.
681 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
682 operations in malware.c.
684 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
687 PH/13 If write_rejectlog was set false when logging was sent to syslog with
688 syslog_duplication set false, log lines that would normally be written
689 both the the main log and to the reject log were not written to syslog at
692 PH/14 In the default configuration, change the use of "message" in ACL warn
693 statements to "add_header".
695 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
696 not followed by a command (e.g. "seen endif").
698 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
699 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
702 PH/17 Added extra conditions to the default value of delay_warning_condition
706 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
707 { match{$h_precedence:}{(?i)bulk|list|junk} } \
708 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
711 The Auto-Submitted: and various List- headers are standardised, whereas I
712 don't think Precedence: ever was.
714 PH/18 Refactored debugging code in route_finduser() to show more information,
715 in particular, the error code if getpwnam() issues one.
717 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
718 This is apparently needed in addition to the PH/07 change above to avoid
719 any possible encoding problems.
721 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
722 but not after initializing Perl.
724 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
725 output them only if debugging. By default they are written stderr,
726 apparently, which is not desirable.
728 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
731 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
734 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
736 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
737 authenticated or an ident call has been made. Suppress the default
738 values for $authenticated_id and $authenticated_sender (but permit -oMai
739 and -oMas) when testing with -bh.
741 PH/24 Re-jigged the order of the tests in the default configuration so that the
742 tests for valid domains and recipients precede the DNS black list and CSA
743 tests, on the grounds that those ones are more expensive.
745 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
746 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
747 This bug exists in every version of Exim that I still have, right back to
750 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
751 However, an attempt to turn on case-sensitivity in a regex key by
752 including (?-i) didn't work because the subject string was already
753 lowercased, and the effects were non-intuitive. It turns out that a
754 one-line patch can be used to allow (?-i) to work as expected.
760 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
761 other effects) broke the use of negated acl sub-conditions.
763 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
766 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
767 "Deny" causes Exim to reject the incoming connection with a 554 error.
768 Unfortunately, if there is a major crisis, such as a disk failure,
769 tcp-wrappers gives "deny", whereas what one would like would be some
770 kind of temporary error. A kludge has been added to help with this.
771 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
772 554 error is used if errno is still zero or contains ENOENT (which occurs
773 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
776 PH/03 Add -lutil to the default FreeBSD LIBS setting.
778 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
779 errors. Otherwise a message that provokes a temporary error (when other
780 messages do not) can cause a whole host to time out.
782 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
783 addresses were routed directly to files or pipes from a redirect router.
784 File deliveries just didn't batch; pipe deliveries might have suffered
787 PH/06 A failure to get a lock for a hints database would erroneously always say
788 "Failed to get write lock", even when it was really a read lock.
790 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
791 of 0600. This has been changed to use the value of the lockfile_mode
792 option (which defaults to 0600).
794 PH/08 Applied small patch from the Sieve maintainer.
796 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
797 folder from quota calculations, a direct delivery into this folder messed
798 up the contents of the maildirsize file. This was because the regex was
799 used only to exclude .Trash (or whatever) when the size of the mailbox
800 was calculated. There was no check that a delivery was happening into an
801 excluded directory. This bug has been fixed by ignoring all quota
802 processing for deliveries into excluded directories.
804 PH/10 Added the maildirfolder_create_regex option to appendfile.
810 PH/01 The code for finding all the local interface addresses on a FreeBSD
811 system running IPv6 was broken. This may well have applied to all BSD
812 systems, as well as to others that have similar system calls. The broken
813 code found IPv4 interfaces correctly, but gave incorrect values for the
814 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
815 that it would not match correctly against @[] and not recognize the IPv6
818 PH/02 The ipliteral router was not recognizing addresses of the form user@
819 [ipv6:....] because it didn't know about the "ipv6:" prefix.
821 PH/03 Added disable_ipv6.
823 PH/04 Changed $reply_address to use the raw form of the headers instead of the
824 decoded form, because it is most often used to construct To: headers
825 lines in autoreplies, and the decoded form may well be syntactically
826 invalid. However, $reply_address has leading white space removed, and all
827 newlines turned into spaces so that the autoreply transport does not
830 PH/05 If group was specified without a user on a router, and no group or user
831 was specified on a transport, the group from the router was ignored.
833 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
834 for visible compile-time settings that can be used to change these
835 numbers, for those that want even more. Backwards compatibility with old
836 spool files has been maintained. However, going back to a previous Exim
837 release will lost any variables that are in spool files.
839 PH/07 Two small changes when running in the test harness: increase delay when
840 passing a TCP/IP connection to a new process, in case the original
841 process has to generate a bounce, and remove special handling of
842 127.0.0.2 (sic), which is no longer necessary.
844 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
845 be the same on different OS.
847 PH/09 Moved a debug statement in filter processing to avoid a race problem when
850 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
851 whether --show-vars was specified or not
853 JJ/02 exipick: Added support for new ACL variable spool format introduced
856 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
857 syntactically invalid From: or Reply-to: line, and a filter used this to
858 generate an autoreply, and therefore failed to obtain an address for the
859 autoreply, Exim could try to deliver to a non-existent relative file
860 name, causing unrelated and misleading errors. What now happens is that
861 it logs this as a hard delivery error, but does not attempt to create a
864 PH/11 The exinext utility has a -C option for testing purposes, but although
865 the given file was scanned by exinext itself; it wasn't being passed on
866 when Exim was called.
868 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
869 an end-of-file indication when reading a command response.
871 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
872 compiled. In many other places in Exim, IPv6 addresses are always
873 recognized, so I have changed this. It also means that IPv4 domain
874 literals of the form [IPV4:n.n.n.n] are now always recognized.
876 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
877 used if the router is not running as root, for example, when verifying at
878 ACL time, or when using -bh. The debugging output from this situation was
879 non-existent - all you got was a failure to exec. I have made two
882 (a) Failures to set uid/gid, the current directory, or a process leader
883 in a subprocess such as that created by queryprogram now generate
884 suitable debugging ouput when -d is set.
886 (b) The queryprogram router detects when it is not running as root,
887 outputs suitable debugging information if -d is set, and then runs
888 the subprocess without attempting to change uid/gid.
890 PH/15 Minor change to Makefile for building test_host (undocumented testing
893 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
894 additional section of a DNS packet that returns MX or SRV records.
895 Instead, it always explicitly searches for A/AAAA records. This avoids
896 major problems that occur when a DNS server includes only records of one
897 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
898 fixed another bug: if SRV records were looked up and the corresponding
899 address records were *not* found in the additional section, the port
900 values from the SRV records were lost.
902 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
903 using the correct key (the original address) when searching the retry
904 rules in order to find which one to use for generating the retry hint.
906 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
907 adding the default one. Similarly, if it contains a Reply-To: header, the
908 errors_reply_to option, if set, is not used.
910 PH/19 When calculating a retry time, Exim used to measure the "time since
911 failure" by looking at the "first failed" field in the retry record. Now
912 it does not use this if it is later than than the arrival time of the
913 message. Instead it uses the arrival time. This makes for better
914 behaviour in cases where some deliveries succeed, thus re-setting the
915 "first failed" field. An example is a quota failure for a huge message
916 when small messages continue to be delivered. Without this change, the
917 "time since failure" will always be short, possible causing more frequent
918 delivery attempts for the huge message than are intended.
919 [Note: This change was subsequently modified - see PH/04 for 4.62.]
921 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
922 $1, $2, $3) because the numerical variables can be reset during some
923 expansion items (e.g. "match"), thereby losing the authentication data.
925 PH/21 Make -bV show the size of off_t variables so that the test suite can
926 decide whether to run tests for quotas > 2G.
928 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
929 mailbox_size, and mailbox_filecount in the appendfile transport. If a
930 filecount value is greater than 2G or if a quota value is greater than 2G
931 on a system where the size of off_t is not greater than 4, a panic error
934 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
935 never match. The debug and -bh output now contains an explicit error
936 message indicating a malformed IPv4 address or mask.
938 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
939 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
942 PH/25 Do not write to syslog when running in the test harness. The only
943 occasion when this arises is a failure to open the main or panic logs
944 (for which there is an explicit test).
946 PH/26 Added the /no_tell option to "control=freeze".
948 PH/27 If a host name lookup failed very early in a connection, for example, if
949 the IP address matched host_lookup and the reverse lookup yielded a name
950 that did not have a forward lookup, an error message of the form "no IP
951 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
952 could be logged. Now it outputs the IP address instead of "NULL".
954 PH/28 An enabling patch from MH: add new function child_open_exim2() which
955 allows the sender and the authenticated sender to be set when
956 submitting a message from within Exim. Since child_open_exim() is
957 documented for local_scan(), the new function should be too.
959 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
960 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
961 results in an empty string is now treated as unset.
963 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
965 PH/31 Added sender_verify_fail logging option.
967 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
968 needed by RFC 822 but not by RFC 2822 was commented out. I have now
969 tidied the source and removed it altogether.
971 PH/33 When a queue run was abandoned because the load average was too high, a
972 log line was always written; now it is written only if the queue_run log
973 selector is set. In addition, the log line for abandonment now contains
974 information about the queue run such as the pid. This is always present
975 in "start" and "stop" lines but was omitted from the "abandon" line.
977 PH/34 Omit spaces between a header name and the colon in the error message that
978 is given when verify = headers_syntax fails (if there are lots of them,
979 the message gets confusing).
981 PH/35 Change the default for dns_check_names_pattern to allow slashes within
982 names, as there are now some PTR records that contain slashes. This check
983 is only to protect against broken name servers that fall over on strange
984 characters, so the fact that it applies to all lookups doesn't matter.
986 PH/36 Now that the new test suite is complete, we can remove some of the
987 special code in Exim that was needed for the old test suite. For example,
988 sorting DNS records because real resolvers return them in an arbitrary
989 order. The new test suite's fake resolver always returns records in the
992 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
993 bounces) except when queue_only is set, to avoid logging races between
994 the different processes.
996 PH/38 Panic-die if .include specifies a non-absolute path.
998 PH/39 A tweak to the "H" retry rule from its user.
1000 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1001 a label. They prevented compilation on older perls.
1003 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1004 a warning to be raised on newish perls.
1006 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1007 on queue. Changes to match documented behaviour of showing count of
1008 messages matching specified criteria.
1010 PH/40 Changed the default ident timeout from 30s to 5s.
1012 PH/41 Added support for the use of login_cap features, on those BSD systems
1013 that have them, for controlling the resources used by pipe deliveries.
1015 PH/42 The content-scanning code uses fopen() to create files in which to put
1016 message data. Previously it was not paying any attention to the mode of
1017 the files. Exim runs with umask(0) because the rest of the code creates
1018 files with open(), and sets the required mode explicitly. Thus, these
1019 files were ending up world-writeable. This was not a big issue, because,
1020 being within the spool directory, they were not world-accessible. I have
1021 created a function called modefopen, which takes an additional mode
1022 argument. It sets umask(777), creates the file, chmods it to the required
1023 mode, then resets the umask. All the relevant calls to fopen() in the
1024 content scanning code have been changed to use this function.
1026 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1027 to 24 hours. This avoids potential overflow problems when processing G
1028 and H retry rules. I suspect nobody ever tinkers with this value.
1030 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1032 PH/45 When the plaintext authenticator is running as a client, the server's
1033 challenges are checked to ensure they are valid base64 strings. By
1034 default, the authentication attempt is cancelled if an invalid string is
1035 received. Setting client_ignore_invalid_base64 true ignores these errors.
1036 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1037 they are received. Thus, the responses can be made to depend on the
1038 challenges. If an invalid string is ignored, an empty string is placed in
1041 PH/46 Messages that are created by the autoreply transport now contains a
1042 References: header, in accordance with RFCs 2822 and 3834.
1044 PH/47 Added authenticated_sender_force to the smtp transport.
1046 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1048 PH/49 Installed latest patch from the Sieve maintainer.
1050 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1051 was also set, the appendfile transport was unnecessarily scanning a
1052 directory of message files (e.g. for maildir delivery) to find the count
1053 of files (along with the size), even though it did not need this
1054 information. It now does the scan only if it needs to find either the
1055 size of the count of files.
1057 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1059 PH/52 Two bugs concerned with error handling when the smtp transport is
1062 (i) Exim was not creating retry information for temporary errors given
1063 for individual recipients after the DATA command when the smtp transport
1064 was used in LMTP mode. This meant that they could be retried too
1065 frequently, and not timed out correctly.
1067 (ii) Exim was setting the flag that allows error details to be returned
1068 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1069 recipients that were returned after the DATA command.
1071 PH/53 This is related to PH/52, but is more general: for any failing address,
1072 when detailed error information was permitted to be returned to the
1073 sender, but the error was temporary, then after the final timeout, only
1074 "retry timeout exceeded" was returned. Now it returns the full error as
1075 well as "retry timeout exceeded".
1077 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1078 do this, and (what is worse) MTAs that accept it.
1080 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1081 will now be deprecated.
1083 PH/56 New os.c-cygwin from the Cygwin maintainer.
1085 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1086 formats (previously only available in exim formats via -bpr, -bpru,
1087 and -bpra. Now also available in native and exiqgrep formats)
1089 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1090 with very large, slow to parse queues
1092 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1094 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1096 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1097 responses to authentication challenges, though it was showing the
1098 challenges; (ii) I've removed the CR characters from the debug output for
1101 PH/58 Allow for the insertion of a newline as well as a space when a string
1102 is turned into more than one encoded-word during RFC 2047 encoding. The
1103 Sieve code now uses this.
1105 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1106 data_4xx, lost_connection, tls_required.
1108 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1109 message was being sent as an SMTP response.
1111 PH/61 Add -l and -k options to exicyclog.
1113 PH/62 When verifying, if an address was redirected to one new address, so that
1114 verification continued, and the new address failed or deferred after
1115 having set something in $address_data, the value of $address_data was not
1116 passed back to the ACL. This was different to the case when no
1117 redirection occurred. The value is now passed back in both cases.
1119 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1120 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1121 use login_cap.h, so on its own it isn't the distinguishing feature. The
1122 new name refers directly to the setclassresources() function.
1124 PH/65 Added configuration files for NetBSD3.
1126 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1128 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1129 is preferred over IPv4.
1131 PH/68 The bounce_return_message and bounce_return_body options were not being
1132 honoured for bounces generated during the reception of non-SMTP messages.
1133 In particular, this applied to messages rejected by the ACL. This bug has
1134 been fixed. However, if bounce_return_message is true and bounce_return_
1135 body is false, the headers that are returned for a non-SMTP message
1136 include only those that have been read before the error was detected.
1137 (In the case of an ACL rejection, they have all been read.)
1139 PH/69 The HTML version of the specification is now built in a directory called
1140 spec_html instead of spec.html, because the latter looks like a path with
1141 a MIME-type, and this confuses some software.
1143 PH/70 Catch two compiler warnings in sieve.c.
1145 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1146 function verify_get_ident() calls ip_connect() to connect a socket, but
1147 if the "connect()" function timed out, ip_connect() used to close the
1148 socket. However, verify_get_ident() also closes the socket later, and in
1149 between Exim writes to the log, which may get opened at this point. When
1150 the socket was closed in ip_connect(), the log could get the same file
1151 descriptor number as the socket. This naturally causes chaos. The fix is
1152 not to close the socket in ip_connect(); the socket should be closed by
1153 the function that creates it. There was only one place in the code where
1154 this was missing, in the iplookup router, which I don't think anybody now
1155 uses, but I've fixed it anyway.
1157 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1158 well as to direct DNS lookups. Otherwise the handling of names in host
1159 lists is inconsistent and therefore confusing.
1165 PH/01 Two changes to the default runtime configuration:
1167 (1) Move the checks for relay_from_hosts and authenticated clients from
1168 after to before the (commented out) DNS black list checks.
1170 (2) Add control=submission to the relay_from_hosts and authenticated
1171 clients checks, on the grounds that messages accepted by these
1172 statements are most likely to be submissions.
1174 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1176 (1) Generate an error if the third argument for the ${prvs expansion is
1179 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1182 (3) Reset the variables that are obtained from the first argument of
1183 ${prvscheck and used in the second argument before leaving the code,
1184 because their memory is reclaimed, so using them afterwards may do
1187 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1188 one (it's much easier than Tom thought :-).
1190 (5) Because of (4), we can now allow for the use of $prvscheck_result
1191 inside the third argument.
1193 PH/03 For some reason, the default setting of PATH when running a command from
1194 a pipe transport was just "/usr/bin". I have changed it to
1197 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1198 anything to be listed in the output from -bV.
1200 PH/05 When a filter generated an autoreply, the entire To: header line was
1201 quoted in the delivery log line, like this:
1203 => >A.N.Other <ano@some.domain> <original@ddress> ...
1205 This has been changed so that it extracts the operative address. There
1206 may be more than one such address. If so, they are comma-separated, like
1209 => >ano@some.domain,ona@other.domain <original@ddress> ...
1211 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1212 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1213 not being looked up in the rDNS to get a host name, Exim was showing the
1214 IP address twice in Received: lines, even though the IP addresses were
1215 identical. For example:
1217 Received: from [1.2.3.4] (helo=[1.2.3.4])
1219 However, if the real host name was known, it was omitting the HELO data
1220 if it matched the actual IP address. This has been tidied up so that it
1221 doesn't show the same IP address twice.
1223 PH/07 When both +timestamp and +memory debugging was on, the value given by
1224 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1225 called by the debug printing, thereby overwriting the timestamp buffer.
1226 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1228 PH/08 When the original message was included in an autoreply transport, it
1229 always said "this is a copy of the message, including all the headers",
1230 even if body_only or headers_only was set. It now gives an appropriate
1233 PH/09 Applied a patch from the Sieve maintainer which:
1235 o fixes some comments
1236 o adds the (disabled) notify extension core
1237 o adds some debug output for the result of if/elsif tests
1238 o points to the current vacation draft in the documentation
1239 and documents the missing references header update
1243 o fixes a bug in processing the envelope test (when testing
1244 multiple envelope elements, the last element determinted the
1247 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1248 Electronic Mail") by including:
1250 Auto-submitted: auto-generated
1252 in the messages that it generates (bounce messages and others, such as
1253 warnings). In the case of bounce messages for non-SMTP mesages, there was
1254 also a typo: it was using "Auto_submitted" (underscore instead of
1255 hyphen). Since every message generated by Exim is necessarily in response
1256 to another message, thes have all been changed to:
1258 Auto-Submitted: auto-replied
1260 in accordance with these statements in the RFC:
1262 The auto-replied keyword:
1264 - SHOULD be used on messages sent in direct response to another
1265 message by an automatic process,
1267 - MUST NOT be used on manually-generated messages,
1269 - MAY be used on Delivery Status Notifications (DSNs) and Message
1270 Disposition Notifications (MDNs),
1272 - MUST NOT be used on messages generated by automatic or periodic
1273 processes, except for messages which are automatic responses to
1276 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1277 to the default Received: header definition.
1279 PH/12 Added log selector acl_warn_skipped (default on).
1281 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1282 variables because (a) they are in the wrong storage pool and (b) even if
1283 they were copied, it wouldn't work properly because of the caching.
1285 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1286 checking when decoding. Apparently there are clients that generate
1287 overlong encoded strings. Why am I not surprised?
1289 PH/15 If the first argument of "${if match_address" was not empty, but did not
1290 contain an "@" character, Exim crashed. Now it writes a panic log message
1291 and treats the condition as false.
1293 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1295 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1296 extension "envelope-auth". The code is finished and in agreement with
1297 other implementations, but there is no documentation so far and in fact,
1298 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1299 not changing the active code.
1301 Print executed "if" and "elsif" statements when debugging is used. This
1302 helps a great deal to understand what a filter does.
1304 Document more things not specified clearly in RFC3028. I had all this
1305 sorted out, when out of a sudden new issues came to my mind. Oops."
1307 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1310 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1311 canonical form (as documented). However, after a host name lookup from
1312 the IP address, check_host() was doing a simple string comparison with
1313 addresses acquired from the DNS when checking that the found name did
1314 have the original IP as one of its addresses. Since any found IPv6
1315 addresses are likely to be in abbreviated form, the comparison could
1316 fail. Luckily, there already exists a function for doing the comparison
1317 by converting both addresses to binary, so now that is used instead of
1318 the text comparison.
1320 PH/20 There was another similar case to PH/19, when a complete host name was
1321 given in a host list; looking up its IP address could give an abbreviated
1322 form, whereas the current host's name might or might not be abbreviated.
1323 The same fix has been applied.
1329 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1330 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1333 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1334 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1336 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1338 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1339 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1340 Also, for "Auto-Submitted", treat anything other than "no" as
1341 non-personal, in accordance with RFC 3834. (Previously it treated
1342 anything starting "auto-" as non-personal.)
1344 TF/01 The control=submission/name=... option had a problem with syntax
1345 errors if the name included a slash character. The /name= option
1346 now slurps the rest of the string, so it can include any characters
1347 but it must come last in the list of options (after /sender_retain
1350 PH/05 Some modifications to the interface to the fake nameserver for the new
1358 TK/01 Added the "success_on_redirect" address verification option. See
1359 NewStuff for rationale and an example.
1361 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1363 PH/02 Patch to exigrep to allow it to work on syslog lines.
1365 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1366 fread() to skip over the body file's header line, because in Cygwin the
1367 header line is locked and is inaccessible.
1369 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1370 co-exist for some time) to make it clear that it is the Exim ID that is
1371 referenced, not the Message-ID: header line.
1373 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1374 string_format() function, because snprintf() does not exist on all
1377 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1378 ${stat: expansion item.
1380 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1381 protocol synchronization error", to keep the pedants happy.
1383 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1384 well as for IRIX systems, when gcc is being used. See the host.c source
1387 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1389 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1392 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1393 transport and to the smtp transport in LMTP mode.
1395 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1397 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1398 run a filter in a subprocess. This could lead to confusion in subsequent
1399 lookups in the parent process. There should also be a search_tidyup() at
1400 the end of the subprocess.
1402 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1403 only if the host matched helo_try_verify_hosts, which caused the
1404 verification to occur when the EHLO/HELO command was issued. The ACL just
1405 tested the remembered result. Now, if a previous verification attempt has
1406 not happened, "verify = helo" does it there and then.
1408 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1410 TK/03 Fix log output including CR from clamd.
1412 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1413 exist provoked a memory error which could cause a segfault.
1415 PH/15 Installed PCRE 6.2
1417 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1419 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1420 of the problem. Specifically, suggested +O2 rather than +O1 for the
1423 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1425 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1426 record, the port was not correctly logged when the outgoing_port log
1427 selector was set (it logged the transort's default port).
1429 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1430 fallback_hosts, and "hosts" in the smtp transport.
1432 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1433 host errors such as "Connection refused".
1435 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1436 authentication with radiusclient 0.4.9:
1438 - Error returned from rc_read_config was caught wrongly
1439 - Username/password not passed on to radius server due to wrong length.
1441 The presumption is that some radiusclient API changes for 4.51/PH/17
1442 were not taken care of correctly. The code is still untested by me (my
1443 Linux distribution still has 0.3.2 of radiusclient), but it was
1444 contributed by a Radius user.
1446 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1447 expanding the "port" option of the smtp transport.
1449 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1450 while reading a MIME header. Thanks to Tom Hughes for a patch.
1452 PH/24 Include config.h inside local_scan.h so that configuration settings are
1455 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1456 that in an ACL for RCPT (for example), you can examine exactly what was
1459 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1460 commands, but it was not correctly comparing the address with the actual
1461 client host address. Thus, it would show the EHLO address in Received:
1462 header lines when this was not necessary.
1464 PH/27 Added the % operator to ${eval:}.
1466 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1467 it should be ignoring failures (because with -C, for example, it has lost
1468 privilege). It wasn't ignoring creation failures other than "already
1471 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1474 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1475 because an input line was too long, either on its own, or by virtue of
1476 too many continuations, the temporary file was not being removed, and the
1477 return code was incorrect.
1479 PH/31 Missing "BOOL" in function definition in filtertest.c.
1481 PH/32 Applied Sieve patches from the maintainer.
1483 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1485 PH/33 Added "verify = not_blind".
1487 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1488 Local/Makefile (with some defaults set). These are used in built scripts
1489 such as exicyclog, but they have never been used in the exim_install
1490 script (though there are many overriding facilities there). I have
1491 arranged that the exim_install script now takes note of these two
1494 PH/35 Installed configuration files for Dragonfly.
1496 PH/36 When a locally submitted message by a trusted user did not contain a
1497 From: header, and the sender address was obtained from -f or from an SMTP
1498 MAIL command, and the trusted user did not use -F to supply a sender
1499 name, $originator_name was incorrectly used when constructing a From:
1500 header. Furthermore, $originator_name was used for submission mode
1501 messages from external hosts without From: headers in a similar way,
1502 which is clearly wrong.
1504 PH/37 Added control=suppress_local_fixups.
1506 PH/38 When log_selector = +received_sender was set, and the addition of the
1507 sender made the log line's construction buffer exactly full, or one byte
1508 less than full, an overflow happened when the terminating "\n" was
1511 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1512 when the result of a list match is failure because a DNS lookup failed.
1514 PH/40 RM_COMMAND is now used in the building process.
1516 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1517 the "build-* directories that it finds.
1519 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1520 address in a domain literal was a prefix of an interface address.
1522 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1523 when verifying a sender address, unless rewrite_headers is false.
1525 PH/44 Wrote a long comment about why errors_to addresses are verified as
1526 recipients, not senders.
1528 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1529 the ratelimit ACL was added.
1531 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1533 PH/46 Added extra information about PostgreSQL errors to the error string.
1535 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1536 avoiding the need to install special zones in a real server. This is
1537 backwards compatible; if it can't find the fake resolver, it drops back.
1538 Thus, both old and new test suites can be run.
1540 TF/02 Added util/ratelimit.pl
1542 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1543 clock is set back in time.
1545 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1546 Candler <B.Candler@pobox.com>.
1548 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1549 OK for addresses that are the result of redirections.
1551 PH/48 A number of further additions for the benefit of the new test suite,
1552 including a fake gethostbyname() that interfaces to the fake DNS resolver
1555 TF/06 The fix for widen_domains has also been applied to qualify_single and
1556 search_parents which are the other dnslookup options that can cause
1559 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1562 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1564 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1565 service" if it encountered an MX record with an empty target hostname.
1566 The message is now "an MX or SRV record indicated no SMTP service".
1568 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1569 if the DNS of the sending site is misconfigured. This is quite a
1570 common situation. This change restores the behaviour of treating a
1571 helo verification defer as a failure.
1573 PH/51 If self=fail was set on a router, the bounce message did not include the
1574 actual error message.
1580 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1582 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1583 command itself ended in error, the underlying message about the transport
1584 filter timeout was being overwritten with the pipe command error. Now the
1585 underlying error message should be appended to the second error message.
1587 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1589 PH/02 Reduce the amount of output that "make" produces by default. Full output
1590 can still be requested.
1592 PH/03 The warning log line about a condition test deferring for a "warn" verb
1593 was being output only once per connection, rather than after each
1594 occurrence (because it was using the same function as for successful
1595 "warn" verbs). This seems wrong, so I have changed it.
1597 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1598 it should not have, which might have caused a crash in the right
1599 circumstances, but probably never did.
1601 PH/04 Installed a modified version of Tony Finch's patch to make submission
1602 mode fix the return path as well as the Sender: header line, and to
1603 add a /name= option so that you can make the user's friendly name appear
1606 TF/03 Added the control = fakedefer ACL modifier.
1608 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1609 Mark Lowes for thorough testing.
1611 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1613 TK/03 Merged latest SRS patch from Miles Wilton.
1615 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1616 with the definition in sysexits.h (which is #included earlier).
1617 Fortunately, Exim does not actually use EX_OK. The code used to try to
1618 preserve the sysexits.h value, by assumimg that macro definitions were
1619 scanned for macro replacements. I have been disabused of this notion,
1620 so now the code just undefines EX_OK before #including unistd.h.
1622 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1623 in the smtp transport. When a block could not be written in a single
1624 write() function, the timeout was being re-applied to each part-write.
1625 This seems wrong - if the receiver was accepting one byte at a time it
1626 would take for ever. The timeout is now adjusted when this happens. It
1627 doesn't have to be particularly precise.
1629 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1630 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1632 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1633 without a domain if the check to <postmaster@domain> fails.
1635 SC/01 Eximstats: added -xls and the ability to specify output files
1636 (patch written by Frank Heydlauf).
1638 SC/02 Eximstats: use FileHandles for outputing results.
1640 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1642 SC/04 Eximstats: fixed display of large numbers with -nvr option
1644 SC/05 Eximstats: fixed merging of reports with empty tables.
1646 SC/06 Eximstats: added the -include_original_destination flag
1648 SC/07 Eximstats: removed tabs and trailing whitespace.
1650 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1652 TK/06 MBOX spool code: Add real "From " MBOX separator line
1653 so the .eml file is really in mbox format (even though
1654 most programs do not really care). Patch from Alex Miller.
1656 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1657 The latter is generated from $received_to and is only set if the
1658 message has one envelope recipient. SA can use these headers,
1659 obviously out-of-the-box. Patch from Alex Miller.
1661 PH/08 The ${def test on a variable was returning false if the variable's
1662 value was "0", contrary to what the specification has always said!
1663 The result should be true unless the variable is empty.
1665 PH/09 The syntax error of a character other than { following "${if
1666 def:variable_name" (after optional whitespace) was not being diagnosed.
1667 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1668 accidental colon was present, for example, could give incorrect results.
1670 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1671 result is used (not including appendfile, where other changes are about
1674 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1675 This involved changing a lot of size variables from int to off_t. It
1676 should work with maildirs and everything.
1678 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1679 spamd dying while we are connected to it.
1681 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1684 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1685 function for BDB 4.3.
1687 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1689 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1690 the sources for PCRE have been re-arranged and more files are now
1693 PH/15 The code I had for printing potentially long long variables in PH/11
1694 above was not the best (it lost precision). The length of off_t variables
1695 is now inspected at build time, and an appropriate printing format (%ld
1696 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1697 to be "long long int" or "long int". This is needed for the internal
1698 formatting function string_vformat().
1700 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1701 the configuration file to be ":syslog", then the script "guesses" where
1702 the logs files are, rather than using the compiled in default. In our
1703 case the guess is not the same as the compiled default, so the script
1704 suddenly stopped working when I started to use syslog. The patch checks
1705 to see if log_file_path is "". If so, it attempts to read it from exim
1706 with no configuration file to get the compiled in version, before it
1707 falls back to the previous guessing code."
1709 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1710 implementing BATV in an Exim configuration. See NewStuff for the gory
1713 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1714 Makefile that are specific to HP-UX.
1716 PH/18 If the "use_postmaster" option was set for a recipient callout together
1717 with the "random" option, the postmaster address was used as the MAIL
1718 FROM address for the random test, but not for the subsequent recipient
1719 test. It is now used for both.
1721 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1722 patch removes a few documentation additions to RFC 3028, because the
1723 latest draft now contains them. It adds the new en;ascii-case comparator
1724 and a new error check for 8bit text in MIME parts. Comparator and
1725 require names are now matched exactly. I enabled the subaddress
1726 extension, but it is not well tested yet (read: it works for me)."
1728 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1729 rework some of the code of TK/09 above to avoid the hardwired use of
1730 "%lld" and "long long". Replaced the call to snprintf() with a call to
1733 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1734 records point to non-existent hosts", "retry timeout exceeded", and
1735 "retry time not reached for any host after a long failure period".
1737 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1738 experimental DomainKeys support:
1740 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1741 (2) On an error such as an illegally used "control", the wrong name for
1742 the control was given.
1744 These problems did NOT occur unless DomainKeys support was compiled.
1746 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1748 PH/24 Added ${if match_ip condition.
1750 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1751 databases so that it will be absolutely obvious if a crash occurs in the
1752 DB library. This is a regular occurrence (often caused by mis-matched
1755 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1756 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1757 picked up on a user's system that detects such things. There doesn't seem
1758 to be a gcc warning option for this - only an attribute that has to be
1759 put on the function's prototype. It seems that in Fedora Core 4 they have
1760 set this on a number of new functions. No doubt there will be more in due
1763 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
1764 specify a transport. However, if an address that was verified by such a
1765 router was the subject of a callout, Exim crashed because it tried to
1766 read the rcpt_include_affixes from the non-existent transport. Now it
1767 just assumes that the setting of that option is false. This bug was
1768 introduced by 4.51/PH/31.
1770 PH/28 Changed -d+all to exclude +memory, because that information is very
1771 rarely of interest, but it makes the output a lot bigger. People tend to
1772 do -d+all out of habit.
1774 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
1775 code in os-type was giving problems when libc.so lives in lib64, like on
1778 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
1779 aren't the modern standard, and it seems that some systems' include files
1780 don't always have them. Exim was already checking for some of the newer
1781 ones like T_AAAA, and defining it itself. I've added checks for all the
1782 record types that Exim uses.
1784 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
1785 not automatically generating a new one, as it is supposed to. This
1786 prevented TLS from working. If the file did exist, but contained invalid
1787 data, a new version was generated, as expected. It was only the case of a
1788 non-existent file that was broken.
1790 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
1791 with a change in libdomainkeys > 0.64.
1793 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
1794 from DNS. If the selector record carries the flag, it now has
1795 precedence over the domain-wide flag.
1797 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
1799 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
1800 the use of a port name that isn't defined in /etc/services) occurred, the
1801 message was deferred as in a normal delivery, and thus remained on the
1802 spool, instead of being failed because of the mua_wrapper setting. This
1803 is now fixed, and I tidied up some of the mua_wrapper messages at the
1806 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
1807 the messages in a hash of arrays rather than using individual hashes.
1808 This is a bit cleaner and results in dramatic memory savings, albeit
1809 at a slight CPU cost.
1811 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
1812 as requested by Marc Sherman.
1814 SC/10 Eximstats: added histograms for user specified patterns as requested
1817 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
1819 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
1820 fopen() in the content-scanning modules that did not already have it.
1826 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1827 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1829 TK/02 Fix ACL "control" statement not being available in MIME ACL.
1831 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1833 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1834 to test Sieve filters that use "vacation".
1836 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1837 that changes the way the GnuTLS parameters are stored in the cache file.
1838 The new format can be generated externally. For backward compatibility,
1839 if the data in the cache doesn't make sense, Exim assumes it has read an
1840 old-format file, and it generates new data and writes a new file. This
1841 means that you can't go back to an older release without removing the
1844 PH/03 A redirect router that has both "unseen" and "one_time" set does not
1845 work if there are any delivery delays because "one_time" forces the
1846 parent to be marked "delivered", so its unseen clone is never tried
1847 again. For this reason, Exim now forbids the simultaneous setting of
1850 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1851 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1852 entry if you want to know the details. The fix, however, neglected to
1853 consider the case when local delivery batching is involved. The test for
1854 "previously delivered" was not happening when checking to see if an
1855 address could be batched with a previous (undelivered) one; under
1856 certain circumstances this could lead to multiple deliveries to the same
1859 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1860 in its include files, and this causes problems building Exim.
1862 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1863 header_syntax) but Exim was just ignoring anything given after a slash.
1864 In particular, this caused confusion with an attempt to use "verify =
1865 reverse_host_lookup/defer_ok". An error is now given when options are
1866 supplied for verify items that do not have them. (Maybe reverse_host_
1867 lookup should have a defer_ok option, but that's a different point.)
1869 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1870 defined by RFC 821) to 2048, because there were problems with some AUTH
1871 commands, and RFC 1869 says the size should be increased for extended
1872 SMTP commands that take arguments.
1874 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1877 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1878 "unknown" error; now it says that the functionality isn't in the binary.
1880 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1881 an address' error message when a string expansion fails (syntax or
1882 whatever). Otherwise the password may appear in the log. Following change
1883 PH/42 below, there is no longer a chance of it appearing in a bounce
1886 PH/11 Installed exipick version 20050225.0 from John Jetmore.
1888 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1889 of its addresses was ever tried. (Bugzilla bug #2.)
1891 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1892 the result incorrectly in the debug output. (It correctly added a newline
1893 to what was transported.)
1895 TF/01 Added $received_time.
1897 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1898 commented out examples of how to interface to a virus scanner and to
1899 SpamAssassin. Also added commented examples of av_scanner and
1900 spamd_address settings.
1902 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1903 and controls are allowed in which ACLs. There were a couple of minor
1904 errors. Some of the entries in the conditions table (which is a table of
1905 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1906 negation of where the condition IS allowed.
1908 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1910 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1911 header file does not have a version number, so I've had to invent a new
1912 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1913 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1914 radiusclient), but it was contributed by a Radius user.
1916 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1917 files or directories, for OpenSSL.
1919 PH/19 When an Exim process that is running as root has to create an Exim log
1920 file, it does so in a subprocess that runs as exim:exim so as to get the
1921 ownership right at creation (otherwise, other Exim processes might see
1922 the file with the wrong ownership). There was no test for failure of this
1923 fork() call, which would lead to the process getting stuck as it waited
1924 for a non-existent subprocess. Forks do occasionally fail when resources
1925 run out. I reviewed all the other calls to fork(); they all seem to check
1928 PH/20 When checking for unexpected SMTP input at connect time (before writing
1929 the banner), Exim was not dealing correctly with a non-positive return
1930 from the read() function. If the client had disconnected by this time,
1931 the result was a log entry for a synchronization error with an empty
1932 string after "input=" when read() returned zero. If read() returned -1
1933 (an event I could not check), uninitialized data bytes were printed.
1934 There were reports of junk text (parts of files, etc) appearing after
1937 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1939 PH/22 Added support for macro redefinition, and (re)definition in between
1940 driver and ACL definitions.
1942 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1943 forgetting to use the resulting value; it was using the unexpanded value.
1945 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1946 hadn't been configured. The fix is from Juergen Kreileder, who
1947 understands it better than I do:
1949 "Here's what I see happening with three configured cyrus_sasl
1950 authenticators configured (plain, login, cram-md5):
1952 On startup auth_cyrus_sasl_init() gets called for each of these.
1953 This means three calls to sasl_listmech() without a specified mech_list.
1954 => SASL tests which mechs of all available mechs actually work
1955 => three warnings about OTP not working
1956 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1958 With the patch, sasl_listmech() also gets called three times. But now
1959 SASL's mech_list option is set to the server_mech specified in the the
1960 authenticator. Or in other words, the answer from sasl_listmech()
1961 gets limited to just the mech you're testing for (which is different
1963 => the return list contains just 'plain' or 'login', 'cram-md5' or
1964 nothing depending on the value of ob->server_mech.
1966 I've just tested the patch: Authentication still works fine,
1967 unavailable mechs specified in the exim configuration are still
1968 caught, and the auth.log warnings about OTP are gone."
1970 PH/25 When debugging is enabled, the contents of the command line are added
1971 to the debugging output, even when log_selector=+arguments is not
1974 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1975 answer is "GNU", and only if the return is "GNU/something" is the answer
1978 PH/27 $acl_verify_message is now set immediately after the failure of a
1979 verification in an ACL, and so is available in subsequent modifiers. In
1980 particular, the message can be preserved by coding like this:
1982 warn !verify = sender
1983 set acl_m0 = $acl_verify_message
1985 Previously, $acl_verify_message was set only while expanding "message"
1986 and "log_message" when a very denied access.
1988 PH/28 Modified OS/os.c-Linux with
1990 -#ifndef OS_LOAD_AVERAGE
1991 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1993 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1994 nomenclature these days.)
1996 PH/29 Installed patch from the Sieve maintainer that adds the options
1997 sieve_useraddress and sieve_subaddress to the redirect router.
1999 PH/30 In these circumstances:
2000 . Two addresses routed to the same list of hosts;
2001 . First host does not offer TLS;
2002 . First host accepts first address;
2003 . First host gives temporary error to second address;
2004 . Second host offers TLS and a TLS session is established;
2005 . Second host accepts second address.
2006 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2007 and peerdn, if requested) that were in fact used only for the second
2010 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2011 attention to any local part prefix or suffix that was matched by the
2012 router that accepted the address. It now behaves in the same way as it
2013 does for delivery: the affixes are removed from the local part unless
2014 rcpt_include_affixes is set on the transport.
2016 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2017 timeout during the DATA phase of an incoming message.
2019 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2020 applied a patch sanctioned by the Sieve maintainer.
2022 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2023 the uid or gid is negative. A case of a negative gid caused this to be
2024 noticed. The fix allows for either to be negative.
2026 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2027 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2030 PH/36 The change PH/12 above was broken. Fixed it.
2032 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2033 the grounds that routing the same address twice would always produce the
2034 same answer. This might have been true once, but it is certainly no
2035 longer true now. Routing a child address may depend on the previous
2036 routing that produced that child. Some complicated redirection strategies
2037 went wrong when messages had multiple recipients, and made Exim's
2038 behaviour dependent on the order in which the addresses were given.
2040 I have moved the duplicate checking until after the routing is complete.
2041 Exim scans the addresses that are assigned to local and remote
2042 transports, and removes any duplicates. This means that more work will be
2043 done, as duplicates will always all be routed, but duplicates are
2044 presumably rare, so I don't expect this is of any significance.
2046 For deliveries to pipes, files, and autoreplies, the duplicate checking
2047 still happens during the routing process, since they are not going to be
2050 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2051 It corrects a timeout issue with spamd. This is Ian's comment: "The
2052 background is that sometimes spamd either never reads data from a
2053 connection it has accepted, or it never writes response data. The exiscan
2054 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2055 blindly assumes that writes won't block so it may never time out."
2057 PH/39 Allow G after quota size as well as K and M.
2059 PH/40 The value set for $authenticated_id in an authenticator may not contain
2060 binary zeroes or newlines because the value is written to log lines and
2061 to spool files. There was no check on this. Now the value is run through
2062 the string_printing() function so that such characters are converted to
2063 printable escape sequences.
2065 PH/41 $message_linecount is a new variable that contains the total number of
2066 lines in the message. Compare $body_linecount, which is the count for the
2069 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2070 bounce and delay warning messages, except in certain special cases, which
2073 (a) An SMTP error message from a remote host;
2074 (b) A message specified in a :fail: redirection;
2075 (c) A message specified in a "fail" command in a system filter;
2076 (d) A message specified in a FAIL return from the queryprogram router;
2077 (e) A message specified by the cannot_route_message router option.
2079 In these cases only, Exim does include the error details in bounce and
2080 warning messages. There are also a few cases where bland messages such
2081 as "unrouteable address" or "local delivery error" are given.
2083 PH/43 $value is now also set for the "else" part of a ${run expansion.
2085 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2086 being worked on, but at least Exim now implements the latest version to
2089 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2090 process to complete was treated as a delivery failure, a timeout while
2091 writing the message to the pipe was logged, but erroneously treated as a
2092 successful delivery. Such timeouts include transport filter timeouts. For
2093 consistency with the overall process timeout, these timeouts are now
2094 treated as errors, giving rise to delivery failures by default. However,
2095 there is now a new Boolean option for the pipe transport called
2096 timeout_defer, which, if set TRUE, converts the failures into defers for
2097 both kinds of timeout. A transport filter timeout is now identified in
2100 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2101 systems where "make" and "gmake" are different, calling "gmake" at top
2102 level broke things. I've arranged for the value of $(MAKE) to be passed
2103 from the Makefile to this script so that it can call the same version of
2107 A note about Exim versions 4.44 and 4.50
2108 ----------------------------------------
2110 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2111 changes of various kinds. As a consequence, a big documentation update was
2112 needed. This delayed the release for rather longer than seemed good, especially
2113 in the light of a couple of (minor) security issues. Therefore, the changes
2114 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2115 release. So 4.44 and 4.50 are in effect two different branches that both start
2118 I have left the 4.50 change log unchanged; it contains all the changes since
2119 4.43. The change log for 4.44 is below; many of its items are identical to
2120 those for 4.50. This seems to be the most sensible way to preserve the
2121 historical information.
2127 1. Minor wording change to the doc/README.SIEVE file.
2129 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2130 computation of the current number of files was incorrect.
2132 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2133 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2136 4. Give more explanation in the error message when the command for a transport
2137 filter fails to execute.
2139 5. There are several places where Exim runs a non-Exim command in a
2140 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2141 was being done only for the command run by the queryprogram router. It is
2142 now done for all such subprocesses. The other cases are: ${run, transport
2143 filters, and the commands run by the lmtp and pipe transports.
2145 6. Added CONFIGURE_GROUP build-time option.
2147 7. Some older OS have a limit of 256 on the maximum number of file
2148 descriptors. Exim was using setrlimit() to set 1000 as a large value
2149 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2150 systems. I've change it so that if it can't get 1000, it tries for 256.
2152 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2153 was an oversight, and furthermore, ever since the addition of extra
2154 controls (e.g. 4.43/32), the checks on when to allow different forms of
2155 "control" were broken. There should now be diagnostics for all cases when a
2156 control that does not make sense is encountered.
2158 9. Added the /retain_sender option to "control=submission".
2160 10. $recipients is now available in the predata ACL (oversight).
2162 11. Tidy the search cache before the fork to do a delivery from a message
2163 received from the command line. Otherwise the child will trigger a lookup
2164 failure and thereby defer the delivery if it tries to use (for example) a
2165 cached ldap connection that the parent has called unbind on.
2167 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2168 of $address_data from the recipient verification was clobbered by the
2169 sender verification.
2171 13. The value of address_data from a sender verification is now available in
2172 $sender_address_data in subsequent conditions in the ACL statement.
2174 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2176 15. Added a new option "connect=<time>" to callout options, to set a different
2179 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2180 was its contents. (It was OK if the option was not defined at all.)
2182 17. A "Completed" log line is now written for messages that are removed from
2183 the spool by the -Mrm option.
2185 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2186 information about exactly what failed.
2188 19. Added -dd to debug only the daemon process.
2190 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2191 handles timeouts, both on the server side and network timeouts. Renamed the
2192 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2194 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2195 It is now set to "smtps".
2197 22. $host_address is now set to the target address during the checking of
2198 ignore_target_hosts.
2200 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2201 being passed; this would have caused $sender_host_name to have been used if
2202 matching the list had actually called for a host name (not very likely,
2203 since this list is usually IP addresses). A host name is now passed as
2206 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2207 code that specifies a non-restarting handler (typically sigaction() in
2208 modern systems) in an attempt to fix a rare and obscure crash bug.
2210 25. Narrowed the window for a race in the daemon that could cause it to ignore
2211 SIGCHLD signals. This is not a major problem, because they are used only to
2212 wake it up if nothing else does.
2214 26. A malformed maildirsize file could cause Exim to calculate negative values
2215 for the mailbox size or file count. Odd effects could occur as a result.
2216 The maildirsize information is now recalculated if the size or filecount
2219 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2220 support for a long time. Removed HAVE_SYS_VFS_H.
2222 28. Installed the latest version of exipick from John Jetmore.
2224 29. In an address list, if the pattern was not a regular expression, an empty
2225 subject address (from a bounce message) matched only if the pattern was an
2226 empty string. Non-empty patterns were not even tested. This was the wrong
2227 because it is perfectly reasonable to use an empty address as part of a
2228 database query. An empty address is now tested by patterns that are
2229 lookups. However, all the other forms of pattern expect the subject to
2230 contain a local part and a domain, and therefore, for them, an empty
2231 address still always fails if the pattern is not itself empty.
2233 30. Exim went into a mad DNS loop when attempting to do a callout where the
2234 host was specified on an smtp transport, and looking it up yielded more
2235 than one IP address.
2237 31. Re-factored the code for checking spool and log partition space into a
2238 function that finds that data and another that does the check. The former
2239 is then used to implement four new variables: $spool_space, $log_space,
2240 $spool_inodes, and $log_inodes.
2242 32. The RFC2047 encoding function was originally intended for short strings
2243 such as real names; it was not keeping to the 75-character limit for
2244 encoded words that the RFC imposes. It now respects the limit, and
2245 generates multiple encoded words if necessary. To be on the safe side, I
2246 have increased the buffer size for the ${rfc2047: expansion operator from
2249 33. It is now permitted to omit both strings after an "if" condition; if the
2250 condition is true, the result is "true". As before, when the second string
2251 is omitted, a false condition yields an empty string. This makes it less
2252 cumbersome to write custom ACL and router conditions.
2254 34. Failure to deliver a bounce message always caused it to be frozen, even if
2255 there was an errors_to setting on the router. The errors_to setting is now
2258 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2259 canonical form (fully expanded) before being placed in
2260 $sender_host_address.
2262 36. The table in the code that translates DNS record types into text (T_A to
2263 "A" for instance) was missing entries for NS and CNAME. It is just possible
2264 that this could have caused confusion if both these types were looked up
2265 for the same domain, because the text type is used as part of Exim's
2266 per-process caching. But the chance of anyone hitting this buglet seems
2269 37. The dnsdb lookup has been extended in a number of ways.
2271 (1) There is a new type, "zns", which walks up the domain tree until it
2272 finds some nameserver records. It should be used with care.
2274 (2) There is a new type, "mxh", which is like "mx" except that it returns
2275 just the host names, not the priorities.
2277 (3) It is now possible to give a list of domains (or IP addresses) to be
2278 looked up. The behaviour when one of the lookups defers can be
2279 controlled by a keyword.
2281 (4) It is now possible to specify the separator character for use when
2282 multiple records are returned.
2284 38. The dnslists ACL condition has been extended: it is now possible to supply
2285 a list of IP addresses and/or domains to be looked up in a particular DNS
2288 39. Added log_selector=+queue_time_overall.
2290 40. When running the queue in the test harness, wait just a tad after forking a
2291 delivery process, to get repeatability of debugging output.
2293 41. Include certificate and key file names in error message when GnuTLS fails
2294 to set them up, because the GnuTLS error message doesn't include the name
2295 of the failing file when there is a problem reading it.
2297 42. Allow both -bf and -bF in the same test run.
2299 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2301 44. The "Exiscan patch" is now merged into the mainline Exim source.
2303 45. Sometimes the final signoff response after QUIT could fail to get
2304 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2305 < 0 before doing a fflush(). This bug looks as though it goes back to the
2306 introduction of TLS in release 3.20, but "sometimes" must have been rare
2307 because the tests only now provoked it.
2309 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2310 (this can affect the format of dates).
2312 47. exim_tidydb, when checking for the continued existence of a message for
2313 which it has found a message-specific retry record, was not finding
2314 messages that were in split spool directories. Consequently, it was
2315 deleting retry records that should have stayed in existence.
2317 48. Steve fixed some bugs in eximstats.
2319 49. The SPA authentication driver was not abandoning authentication and moving
2320 on to the next authenticator when an expansion was forced to fail,
2321 contradicting the general specification for all authenticators. Instead it
2322 was generating a temporary error. It now behaves as specified.
2324 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2325 (the order specifies the preference for clients). The order is now AES256,
2326 AES128, 3DES, ARCFOUR128.
2328 51. Small patch to Sieve code - explicitly set From: when generating an
2331 52. Exim crashed if a remote delivery caused a very long error message to be
2332 recorded - for instance if somebody sent an entire SpamAssassin report back
2333 as a large number of 550 error lines. This bug was coincidentally fixed by
2334 increasing the size of one of Exim's internal buffers (big_buffer) that
2335 happened as part of the Exiscan merge. However, to be on the safe side, I
2336 have made the code more robust (and fixed the comments that describe what
2339 53. Now that there can be additional text after "Completed" in log lines (if
2340 the queue_time_overall log selector is set), a one-byte patch to exigrep
2341 was needed to allow it to recognize "Completed" as not the last thing in
2344 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2345 patch that reportedly fixes this has been added. I am not expert enough to
2346 create a test for it. This is what the patch creator wrote:
2348 "I found a little strange behaviour of ldap code when working with
2349 Windows 2003 AD Domain, where users was placed in more than one
2350 Organization Units. When I tried to give exim partial DN, the exit code
2351 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2352 But simultaneously result of request was absolutely normal ldap result,
2353 so I produce this patch..."
2355 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2356 REFERENCE, so I have modified the code to exclude the patch when that macro
2359 55. Some experimental protocols are using DNS PTR records for new purposes. The
2360 keys for these records are domain names, not reversed IP addresses. The
2361 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2362 leaves it alone. Component reversal etc. now happens only for IP addresses.
2365 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2367 57. Double the size of the debug message buffer (to 2048) so that more of very
2368 long debug lines gets shown.
2370 58. The exicyclog utility now does better if the number of log files to keep
2371 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2373 59. Two changes related to the smtp_active_hostname option:
2375 (1) $smtp_active_hostname is now available as a variable.
2376 (2) The default for smtp_banner uses $smtp_active_hostname instead
2377 of $primary_hostname.
2379 60. The host_aton() function is supposed to be passed a string that is known
2380 to be a valid IP address. However, in the case of IPv6 addresses, it was
2381 not checking this. This is a hostage to fortune. Exim now panics and dies
2382 if the condition is not met. A case was found where this could be provoked
2383 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2384 components; fortuitously, this particular loophole had already been fixed
2385 by change 4.50/55 above.
2387 If there are any other similar loopholes, the new check in host_aton()
2388 itself should stop them being exploited. The report I received stated that
2389 data on the command line could provoke the exploit when Exim was running as
2390 exim, but did not say which command line option was involved. All I could
2391 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2392 running as the user.
2395 61. There was a buffer overflow vulnerability in the SPA authentication code
2396 (which came originally from the Samba project). I have added a test to the
2397 spa_base64_to_bits() function which I hope fixes it.
2400 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2401 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2403 63. The daemon start-up calls getloadavg() while still root for those OS that
2404 need the first call to be done as root, but it missed one case: when
2405 deliver_queue_load_max is set with deliver_drop_privilege. This is
2406 necessary for the benefit of the queue runner, because there is no re-exec
2407 when deliver_drop_privilege is set.
2409 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2410 This has been fixed.
2412 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2413 was in use, was not putting the data itself into the right store pool;
2414 consequently, it could be overwritten for a subsequent message in the same
2415 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2418 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2420 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2421 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2422 the function were treating the return as a boolean value, which happened to
2423 work because 0=false and not-0=true, but is not correct code.
2425 68. The host_aton() function was not handling scoped IPv6 addresses (those
2426 with, for example, "%eth0" on the end) correctly.
2428 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2429 negated items (that is, ~something) in unsigned ints. Some compilers
2430 apparently mutter when there is no cast.
2432 70. If an address verification called from an ACL failed, and did not produce a
2433 user-specific message (i.e. there was only a "system" message), nothing was
2434 put in $acl_verify_message. In this situation, it now puts the system
2437 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2438 session; change 4.31/43 added the unwanted input to the log line - except
2439 that it did not do this in the start of session case. It now does.
2441 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2442 This is wrong and can cause the other end to generate a synchronization
2443 error if it is another Exim or anything else that does the synchronization
2444 check. A QUIT command is no longer sent after a timeout.
2446 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2447 during host lookups.
2449 74. The defer_ok option of callout verification was not working if it was used
2450 when verifying addresses in header lines, that is, for this case:
2452 verify = header_sender/callout=defer_ok
2454 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2455 those file descriptors could be used for SMTP connections. If anything
2456 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2457 could be sent to the SMTP client, causing chaos. The daemon now opens
2458 stdin, stdout, and stderr to /dev/null when it puts itself into the
2461 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2462 log by default. The user can override this with suitable Perl magic.
2464 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2465 the log message when discard triggers, was not working for the DATA ACL or
2466 for the non-SMTP ACL.
2468 78. Error message wording change in sieve.c.
2470 79. If smtp_accept_max_per_host was set, the number of connections could be
2471 restricted to fewer than expected, because the daemon was trying to set up
2472 a new connection before checking whether the processes handling previous
2473 connections had finished. The check for completed processes is now done
2474 earlier. On busy systems, this bug wouldn't be noticed because something
2475 else would have woken the daemon, and it would have reaped the completed
2478 80. If a message was submitted locally by a user whose login name contained one
2479 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2480 It caused a spool format error. I have fixed the spool reading code. A
2481 related problem was that the "from" clause in the Received: line became
2482 illegal because of the space(s). It is now covered by ${quote_local_part.
2484 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2487 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2489 83. Patch from Sieve maintainer to fix unterminated string problem in
2490 "vacation" handling.
2492 84. Some minor changes to the Linux configuration files to help with other
2493 OS variants using glibc.
2495 85. One more patch for Sieve to update vacation handling to latest spec.
2498 ----------------------------------------------------
2499 See the note above about the 4.44 and 4.50 releases.
2500 ----------------------------------------------------
2506 1. Change 4.43/35 introduced a bug that caused file counts to be
2507 incorrectly computed when quota_filecount was set in an appendfile
2510 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2511 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2514 3. Give more explanation in the error message when the command for a transport
2515 filter fails to execute.
2517 4. There are several places where Exim runs a non-Exim command in a
2518 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2519 was being done only for the command run by the queryprogram router. It is
2520 now done for all such subprocesses. The other cases are: ${run, transport
2521 filters, and the commands run by the lmtp and pipe transports.
2523 5. Some older OS have a limit of 256 on the maximum number of file
2524 descriptors. Exim was using setrlimit() to set 1000 as a large value
2525 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2526 systems. I've change it so that if it can't get 1000, it tries for 256.
2528 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2529 was an oversight, and furthermore, ever since the addition of extra
2530 controls (e.g. 4.43/32), the checks on when to allow different forms of
2531 "control" were broken. There should now be diagnostics for all cases when a
2532 control that does not make sense is encountered.
2534 7. $recipients is now available in the predata ACL (oversight).
2536 8. Tidy the search cache before the fork to do a delivery from a message
2537 received from the command line. Otherwise the child will trigger a lookup
2538 failure and thereby defer the delivery if it tries to use (for example) a
2539 cached ldap connection that the parent has called unbind on.
2541 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2542 of $address_data from the recipient verification was clobbered by the
2543 sender verification.
2545 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2546 was its contents. (It was OK if the option was not defined at all.)
2548 11. A "Completed" log line is now written for messages that are removed from
2549 the spool by the -Mrm option.
2551 12. $host_address is now set to the target address during the checking of
2552 ignore_target_hosts.
2554 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2555 being passed; this would have caused $sender_host_name to have been used if
2556 matching the list had actually called for a host name (not very likely,
2557 since this list is usually IP addresses). A host name is now passed as
2560 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2561 code that specifies a non-restarting handler (typically sigaction() in
2562 modern systems) in an attempt to fix a rare and obscure crash bug.
2564 15. Narrowed the window for a race in the daemon that could cause it to ignore
2565 SIGCHLD signals. This is not a major problem, because they are used only to
2566 wake it up if nothing else does.
2568 16. A malformed maildirsize file could cause Exim to calculate negative values
2569 for the mailbox size or file count. Odd effects could occur as a result.
2570 The maildirsize information is now recalculated if the size or filecount
2573 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2574 support for a long time. Removed HAVE_SYS_VFS_H.
2576 18. Updated exipick to current release from John Jetmore.
2578 19. Allow an empty sender to be matched against a lookup in an address list.
2579 Previously the only cases considered were a regular expression, or an
2582 20. Exim went into a mad DNS lookup loop when doing a callout where the
2583 host was specified on the transport, if the DNS lookup yielded more than
2586 21. The RFC2047 encoding function was originally intended for short strings
2587 such as real names; it was not keeping to the 75-character limit for
2588 encoded words that the RFC imposes. It now respects the limit, and
2589 generates multiple encoded words if necessary. To be on the safe side, I
2590 have increased the buffer size for the ${rfc2047: expansion operator from
2593 22. Failure to deliver a bounce message always caused it to be frozen, even if
2594 there was an errors_to setting on the router. The errors_to setting is now
2597 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2598 canonical form (fully expanded) before being placed in
2599 $sender_host_address.
2601 24. Updated eximstats to version 1.33
2603 25. Include certificate and key file names in error message when GnuTLS fails
2604 to set them up, because the GnuTLS error message doesn't include the name
2605 of the failing file when there is a problem reading it.
2607 26. Expand error message when OpenSSL has problems setting up cert/key files.
2610 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2611 (this can affect the format of dates).
2613 28. exim_tidydb, when checking for the continued existence of a message for
2614 which it has found a message-specific retry record, was not finding
2615 messages that were in split spool directories. Consequently, it was
2616 deleting retry records that should have stayed in existence.
2618 29. eximstats updated to version 1.35
2619 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2620 1.35 - bugfix such that pie charts by volume are generated correctly
2622 30. The SPA authentication driver was not abandoning authentication and moving
2623 on to the next authenticator when an expansion was forced to fail,
2624 contradicting the general specification for all authenticators. Instead it
2625 was generating a temporary error. It now behaves as specified.
2627 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2628 (the order specifies the preference for clients). The order is now AES256,
2629 AES128, 3DES, ARCFOUR128.
2631 31. Small patch to Sieve code - explicitly set From: when generating an
2634 32. Exim crashed if a remote delivery caused a very long error message to be
2635 recorded - for instance if somebody sent an entire SpamAssassin report back
2636 as a large number of 550 error lines. This bug was coincidentally fixed by
2637 increasing the size of one of Exim's internal buffers (big_buffer) that
2638 happened as part of the Exiscan merge. However, to be on the safe side, I
2639 have made the code more robust (and fixed the comments that describe what
2642 33. Some experimental protocols are using DNS PTR records for new purposes. The
2643 keys for these records are domain names, not reversed IP addresses. The
2644 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2645 leaves it alone. Component reversal etc. now happens only for IP addresses.
2648 34. The host_aton() function is supposed to be passed a string that is known
2649 to be a valid IP address. However, in the case of IPv6 addresses, it was
2650 not checking this. This is a hostage to fortune. Exim now panics and dies
2651 if the condition is not met. A case was found where this could be provoked
2652 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2653 components; fortuitously, this particular loophole had already been fixed
2654 by change 4.50/55 or 4.44/33 above.
2656 If there are any other similar loopholes, the new check in host_aton()
2657 itself should stop them being exploited. The report I received stated that
2658 data on the command line could provoke the exploit when Exim was running as
2659 exim, but did not say which command line option was involved. All I could
2660 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2661 running as the user.
2664 35. There was a buffer overflow vulnerability in the SPA authentication code
2665 (which came originally from the Samba project). I have added a test to the
2666 spa_base64_to_bits() function which I hope fixes it.
2669 36. The daemon start-up calls getloadavg() while still root for those OS that
2670 need the first call to be done as root, but it missed one case: when
2671 deliver_queue_load_max is set with deliver_drop_privilege. This is
2672 necessary for the benefit of the queue runner, because there is no re-exec
2673 when deliver_drop_privilege is set.
2675 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2676 was in use, was not putting the data itself into the right store pool;
2677 consequently, it could be overwritten for a subsequent message in the same
2678 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2681 38. Sometimes the final signoff response after QUIT could fail to get
2682 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2683 < 0 before doing a fflush(). This bug looks as though it goes back to the
2684 introduction of TLS in release 3.20, but "sometimes" must have been rare
2685 because the tests only now provoked it.
2691 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2692 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2693 Now it returns an integer. A number of calls were still expecting a T/F
2694 return. Fortuitously, in all cases, the tests worked in OK situations,
2695 which is the norm. However, things would have gone wrong on any write
2696 failures on the smtp file descriptor. This function is used when sending
2697 messages over SMTP and also when doing verify callouts.
2699 2. When Exim is called to do synchronous delivery of a locally submitted
2700 message (the -odf or -odi options), it no longer closes stderr before doing
2703 3. Implemented the mua_wrapper option.
2705 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2707 5. Implemented the functions header_remove(), header_testname(),
2708 header_add_at_position(), and receive_remove_recipient(), and exported them
2711 6. If an ACL "warn" statement specified the addition of headers, Exim already
2712 inserted X-ACL-Warn: at the start if there was no header name. However, it
2713 was not making this test for the second and subsequent header lines if
2714 there were newlines in the string. This meant that an invalid header could
2715 be inserted if Exim was badly configured.
2717 7. Allow an ACL "warn" statement to add header lines at the start or after all
2718 the Received: headers, as well as at the end.
2720 8. Added the rcpt_4xx retry error code.
2722 9. Added postmaster_mailfrom=xxx to callout verification option.
2724 10. Added mailfrom=xxxx to the callout verification option, for verify=
2727 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2728 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2730 12. Inserted some casts to stop certain compilers warning when using pointer
2731 differences as field lengths or precisions in printf-type calls (mostly
2732 affecting debugging statements).
2734 13. Added optional readline() support for -be (dynamically loaded).
2736 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2737 same clock tick as a message's arrival, so that its received time was the
2738 same as the "first fail" time on the retry record, and that message
2739 remained on the queue past the ultimate address timeout, every queue runner
2740 would try a delivery (because it was past the ultimate address timeout) but
2741 after another failure, the ultimate address timeout, which should have then
2742 bounced the address, did not kick in. This was a "< instead of <=" error;
2743 in most cases the first failure would have been in the next clock tick
2744 after the received time, and all would be well.
2746 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2747 being recognized when the domain list was tested by the match_domain
2748 condition in an expansion string.
2750 16. Added the ${str2b64: operator.
2752 17. Exim was always calling setrlimit() to set a large limit for the number of
2753 processes, without checking whether the existing limit was already
2754 adequate. (It did check for the limit on file descriptors.) Furthermore,
2755 errors from getrlimit() and setrlimit() were being ignored. Now they are
2756 logged to the main and panic logs, but Exim does carry on, to try to do its
2757 job under whatever limits there are.
2759 18. Imported PCRE 5.0.
2761 19. Trivial typo in log message " temporarily refused connection" (the leading
2764 20. If the log selector return_path_on_delivery was set and an address was
2765 redirected to /dev/null, the delivery process crashed because it assumed
2766 that a return path would always be set for a "successful" delivery. In this
2767 case, the whole delivery is bypassed as an optimization, and therefore no
2770 21. Internal re-arrangement: the function for sending a challenge and reading
2771 a response while authentication was assuming a zero-terminated challenge
2772 string. It's now changed to take a pointer and a length, to allow for
2773 binary data in such strings.
2775 22. Added the cyrus_sasl authenticator (code supplied by MBM).
2777 23. Exim was not respecting finduser_retries when seeking the login of the
2778 uid under which it was called; it was always trying 10 times. (The default
2779 setting of finduser_retries is zero.) Also, it was sleeping after the final
2780 failure, which is pointless.
2782 24. Implemented tls_on_connect_ports.
2784 25. Implemented acl_smtp_predata.
2786 26. If the domain in control=submission is set empty, Exim assumes that the
2787 authenticated id is a complete email address when it generates From: or
2788 Sender: header lines.
2790 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
2791 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
2792 chown and chgrp in /bin and hostname in /usr/bin.
2794 28. Exim was keeping the "process log" file open after each use, just as it
2795 does for the main log. This opens the possibility of it remaining open for
2796 long periods when the USR1 signal hits a daemon. Occasional processlog
2797 errors were reported, that could have been caused by this. Anyway, it seems
2798 much more sensible not to leave this file open at all, so that is what now
2801 29. The long-running daemon process does not normally write to the log once it
2802 has entered its main loop, and it closes the log before doing so. This is
2803 so that log files can straightforwardly be renamed and moved. However,
2804 there are a couple of unusual error situations where the daemon does write
2805 log entries, and I had neglected to close the log afterwards.
2807 30. The text of an SMTP error response that was received during a remote
2808 delivery was being truncated at 512 bytes. This is too short for some of
2809 the long messages that one sometimes sees. I've increased the limit to
2812 31. It is now possible to make retry rules that apply only when a message has a
2813 specific sender, in particular, an empty sender.
2815 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
2816 it possible to be selective about when SMTP synchronization is enforced.
2818 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
2820 32. Implemented hosts_connection_nolog.
2822 33. Added an ACL for QUIT.
2824 34. Setting "delay_warning=" to disable warnings was not working; it gave a
2827 35. Added mailbox_size and mailbox_filecount to appendfile.
2829 36. Added control = no_multiline_responses to ACLs.
2831 37. There was a bug in the logic of the code that waits for the clock to tick
2832 in the case where the clock went backwards by a substantial amount such
2833 that the microsecond fraction of "now" was more than the microsecond
2834 fraction of "then" (but the whole seconds number was less).
2836 38. Added support for the libradius Radius client library this is found on
2837 FreeBSD (previously only the radiusclient library was supported).
2843 1. When certain lookups returned multiple values in the form name=value, the
2844 quoting of the values was not always being done properly. Specifically:
2845 (a) If the value started with a double quote, but contained no whitespace,
2847 (b) If the value contained whitespace other than a space character (i.e.
2848 tabs or newlines or carriage returns) it was not quoted.
2849 This fix has been applied to the mysql and pgsql lookups by writing a
2850 separate quoting function and calling it from the lookup code. The fix
2851 should probably also be applied to nisplus, ibase and oracle lookups, but
2852 since I cannot test any of those, I have not disturbed their existing code.
2854 2. A hit in the callout cache for a specific address caused a log line with no
2855 reason for rejecting RCPT. Now it says "Previous (cached) callout
2856 verification failure".
2858 3. There was an off-by-one bug in the queryprogram router. An over-long
2859 return line was truncated at 256 instead of 255 characters, thereby
2860 overflowing its buffer with the terminating zero. As well as fixing this, I
2861 have increased the buffer size to 1024 (and made a note to document this).
2863 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2864 when Exim is waiting for an SMTP response from a remote server, Exim
2865 restarts its select() call on the socket, thereby resetting its timeout.
2866 This is not a problem when such interrupts are rare. Somebody set up a cron
2867 job to run exiwhat every 2 minutes, which is less than the normal select()
2868 timeout (5 or 10 minutes). This meant that the select() timeout never
2869 kicked in because it was always reset. I have fixed this by comparing the
2870 time when an interrupt arrives with the time at the start of the first call
2871 to select(). If more time than the timeout has elapsed, the interrupt is
2872 treated as a timeout.
2874 5. Some internal re-factoring in preparation for the addition of Sieve
2875 extensions (by MH). In particular, the "personal" test is moved to a
2876 separate function, and given an option for scanning Cc: and Bcc: (which is
2877 not set for Exim filters).
2879 6. When Exim created an email address using the login of the caller as the
2880 local part (e.g. when creating a From: or Sender: header line), it was not
2881 quoting the local part when it contained special characters such as @.
2883 7. Installed new OpenBSD configuration files.
2885 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2886 try to make them clearer.
2888 9. Callout options, other than the timeout value, were being ignored when
2889 verifying sender addresses in header lines. For example, when using
2891 verify = header_sender/callout=no_cache
2893 the cache was (incorrectly) being used.
2895 10. Added a missing instance of ${EXE} to the exim_install script; this affects
2896 only the Cygwin environment.
2898 11. When return_path_on_delivery was set as a log selector, if different remote
2899 addresses in the same message used different return paths and parallel
2900 remote delivery occurred, the wrong values would sometimes be logged.
2901 (Whenever a remote delivery process finished, the return path value from
2902 the most recently started remote delivery process was logged.)
2904 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2905 lines when AUTH and/or TLS are in use. This is the "received protocol"
2906 field. Exim used to use "asmtp" for authenticated SMTP, without any
2907 indication (in the protocol name) for TLS use. Now it follows the RFC and
2908 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2909 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2910 names appear in log lines as well as in Received: header lines.
2912 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2913 extensions, and comparison tests, and to fix some bugs.
2915 14. Changes to the "personal" filter test:
2917 (1) The test was buggy in that it was just doing the equivalent of
2918 "contains" tests on header lines. For example, if a user's address was
2919 anne@some.where, the "personal" test would incorrectly be true for
2921 To: susanne@some.where
2923 This test is now done by extracting each address from the header in turn,
2924 and checking the entire address. Other tests that are part of "personal"
2925 are now done using regular expressions (for example, to check local parts
2926 of addresses in From: header lines).
2928 (2) The list of non-personal local parts in From: addresses has been
2929 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2930 taken from the Sieve specification recommendations.
2932 (3) If the message contains any header line starting with "List-" it is
2933 treated as non-personal.
2935 (4) The test for "circular" in the Subject: header line has been removed
2936 because it now seems ill-conceived.
2938 15. Minor typos in src/EDITME comments corrected.
2940 16. Installed latest exipick from John Jetmore.
2942 17. If headers_add on a router specified a text string that was too long for
2943 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2944 of string_sprintf() is now avoided.
2946 18. $message_body_size was not set (it was always zero) when running the DATA
2947 ACL and the local_scan() function.
2949 19. For the "mail" command in an Exim filter, no default was being set for
2950 the once_repeat time, causing a random time value to be used if "once" was
2951 specified. (If the value happened to be <= 0, no repeat happened.) The
2952 default is now 0s, meaning "never repeat". The "vacation" command was OK
2953 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2954 (I found it when inspecting the code).
2956 20. There is now an overall timeout for performing a callout verification. It
2957 defaults to 4 times the callout timeout, which applies to individual SMTP
2958 commands during the callout. The overall timeout applies when there is more
2959 than one host that can be tried. The timeout is checked before trying the
2960 next host. This prevents very long delays if there are a large number of
2961 hosts and all are timing out (e.g. when the network connections are timing
2962 out). The value of the overall timeout can be changed by specifying an
2963 additional sub-option for "callout", called "maxwait". For example:
2965 verify = sender/callout=5s,maxwait=20s
2967 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2968 to the end before writing, but this should make it even safer).
2970 22. Exim was forgetting that it had advertised PIPELINING for the second and
2971 subsequent messages on an SMTP connection. It was also not resetting its
2972 memory on STARTTLS and an internal HELO.
2974 23. When Exim logs an SMTP synchronization error within a session, it now
2975 records whether PIPELINING has been advertised or not.
2977 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2978 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2979 rather than long int.
2981 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2983 26. Added the never_mail option to autoreply.
2989 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2990 crash if the getsockname() call failed; this can happen if a connection is
2991 closed very soon after it is established. The problem was simply in the
2992 order in which certain operations were done, causing Exim to try to write
2993 to the SMTP stream before it had set up the file descriptor. The bug has
2994 been fixed by making things happen in the correct order.
3000 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3001 before the connection was closed, thus losing the rejection response.
3003 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3004 some early Solaris releases, but causes trouble in current releases where
3005 socklen_t is defined.
3007 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3010 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3013 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3014 that releasing the top of it at the end releases what was used for sub-
3015 expansions (unless the block got too big). However, discard this block if
3016 the first thing is a variable or header, so that we can use its block when
3017 it is dynamic (useful for very large $message_headers, for example).
3019 6. Lookups now cache *every* query, not just the most recent. A new, separate
3020 store pool is used for this. It can be recovered when all lookup caches are
3021 flushed. Lookups now release memory at the end of their result strings.
3022 This has involved some general refactoring of the lookup sources.
3024 7. Some code has been added to the store_xxx() functions to reduce the amount
3025 of flapping under certain conditions.
3027 8. log_incoming_interface used to affect only the <= reception log lines. Now
3028 it causes the local interface and port to be added to several more SMTP log
3029 lines, for example "SMTP connection from", and rejection lines.
3031 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3033 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3035 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3036 could be overwritten at the end of the current message (or the start of a
3037 new message if it was set in a HELO ACL). The value is now preserved for
3038 the duration of the SMTP connection.
3040 12. If a transport had a headers_rewrite setting, and a matching header line
3041 contained an unqualified address, that address was qualified, even if it
3042 did not match any rewriting rules. The underlying bug was that the values
3043 of the flags that permit the existence of unqualified sender and recipient
3044 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3045 non-local messages, and by -bnq for local messages) were not being
3046 preserved with the message after it was received.
3048 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3049 "next input=" as part of the text comprising the host identity instead of
3050 the correct text. The code was using the same buffer for two different
3051 strings. However, depending on which order the printing function evaluated
3052 its arguments, the bug did not always show up. Under Linux, for example, my
3053 test suite worked just fine.
3055 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3056 doesn't work with some older versions of Perl. It has been changed to "my",
3057 which in any case is probably the better facility to use.
3059 15. A really picky compiler found some instances of statements for creating
3060 error messages that either had too many or two few arguments for the format
3063 16. The size of the buffer for calls to the DNS resolver has been increased
3064 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3065 for addresses that have a lot of PTR records. This alleviates a problem; it
3066 does not fully solve it.
3068 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3069 buffer now truncates the list and logs the incident, which is the same
3070 action as happens when Exim is looking up a host name and its aliases.
3071 Previously in this situation something unpredictable would happen;
3072 sometimes it was "internal error: store_reset failed".
3074 18. If a server dropped the connection unexpectedly when an Exim client was
3075 using GnuTLS and trying to read a response, the client delivery process
3076 crashed while trying to generate an error log message.
3078 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3079 string, for example:
3081 warn message = H1: something\nH2: something
3083 the text was added as a single header line from Exim's point of view
3084 though it ended up OK in the delivered message. However, searching for the
3085 second and subsequent header lines using $h_h2: did not work. This has been
3086 fixed. Similarly, if a system filter added multiple headers in this way,
3087 the routers could not see them.
3089 20. Expanded the error message when iplsearch is called with an invalid key to
3090 suggest using net-iplsearch in a host list.
3092 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3093 ACLs are no longer actually imposed (and a message to that effect is
3096 22. If a "gecos" field in a passwd entry contained escaped characters, in
3097 particular, if it contained a \" sequence, Exim got it wrong when building
3098 a From: or a Sender: header from that name. A second bug also caused
3099 incorrect handling when an unquoted " was present following a character
3100 that needed quoting.
3102 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3103 was not being matched caselessly.
3105 24. Arranged for all hyphens in the exim.8 source to be escaped with
3108 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3109 properly. Recipient callout cache records were still being keyed to include
3110 the sender, even when use_sender was set false. This led to far more
3111 callouts that were necessary. The sender is no longer included in the key
3112 when use_sender is false.
3114 26. Added "control = submission" modifier to ACLs.
3116 27. Added the ${base62d: operator to decode base 62 numbers.
3118 28. dnsdb lookups can now access SRV records.
3120 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3121 the configuration file.
3123 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3124 (-v) mode. This makes the output for a verbose queue run more intelligible.
3126 31. Added a use_postmaster feature to recipient callouts.
3128 32. Added the $body_zerocount variable, containing the number of binary zero
3129 bytes in the message body.
3131 33. The time of last modification of the "new" subdirectory is now used as the
3132 "mailbox time last read" when there is a quota error for a maildir
3135 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3137 35. Added +ignore_unknown as a special item in host lists.
3139 36. Code for decoding IPv6 addresses in host lists is now included, even if
3140 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3141 address was recognized as an IP address, but was then not correctly decoded
3142 into binary, causing unexpected and incorrect effects when compared with
3149 1. Very minor rewording of debugging text in manualroute to say "list of
3150 hosts" instead of "hostlist".
3152 2. If verify=header_syntax was set, and a header line with an unqualified
3153 address (no domain) and a large number of spaces between the end of the
3154 name and the colon was received, the reception process suffered a buffer
3155 overflow, and (when I tested it) crashed. This was caused by some obsolete
3156 code that should have been removed. The fix is to remove it!
3158 3. When running in the test harness, delay a bit after writing a bounce
3159 message to get a bit more predictability in the log output.
3161 4. Added a call to search_tidyup() just before forking a reception process. In
3162 theory, someone could use a lookup in the expansion of smtp_accept_max_
3163 per_host which, without the tidyup, could leave open a database connection.
3165 5. Added the variables $recipient_data and $sender_data which get set from a
3166 lookup success in an ACL "recipients" or "senders" condition, or a router
3167 "senders" option, similar to $domain_data and $local_part_data.
3169 6. Moved the writing of debug_print from before to after the "senders" test
3172 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3173 problems for message scanning, either using a data ACL, or using
3174 local_scan() because the Received: header was not generated till after they
3175 were called (in order to set the time as the time of reception completion).
3176 I have revised the way this works. The header is now generated after the
3177 body is received, but before the ACL or local_scan() are called. After they
3178 are run, the timestamp in the header is updated.
3184 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3185 before starting a queue runner without re-exec. This happened only when
3186 deliver_drop_privilege was set or when the Exim user was set to root. The
3187 effect of the bug was that timeouts during subsequent deliveries caused
3188 crashes instead of being properly handled. The handler is now left at its
3189 default (and expected) setting.
3191 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3192 message, again when deliver_drop_privilege is set or Exim is run as root.
3193 The bug described in (1) was not present in this case, but the tidying up
3194 of the other signals was missing. I have made the two cases consistent.
3196 3. The ignore_target_hosts setting on a manualroute router was being ignored
3197 for hosts that were looked up using the /MX notation.
3199 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3202 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3203 operated on the sender address. After changing the $sender_address to <>
3204 for the sender address verify, Exim was re-instated it as the original
3205 (before rewriting) address, but remembering that it had rewritten it, so it
3206 wasn't rewriting it again. This bug also had the effect of breaking the
3207 sender address verification caching when the sender address was rewritten.
3209 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3210 This has been changed so that if the ip literal address matches
3211 ignore_target_hosts, the router declines.
3213 7. Added expansion conditions match_domain, match_address, and match_local_
3214 part (NOT match_host).
3216 8. The placeholder for the Received: header didn't have a length field set.
3218 9. Added code to Exim itself and to exim_lock to test for a specific race
3219 condition that could lead to file corruption when using MBX delivery. The
3220 issue is with the lockfile that is created in /tmp. If this file is removed
3221 after a process has opened it but before that process has acquired a lock,
3222 there is the potential for a second process to recreate the file and also
3223 acquire a lock. This could lead to two Exim processes writing to the file
3224 at the same time. The added code performs the same test as UW imapd; it
3225 checks after acquiring the lock that its file descriptor still refers to
3226 the same named file.
3228 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3229 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3232 11. Added the smtp_active_hostname option. If used, this will typically be made
3233 to depend on the incoming interface address. Because $interface_address is
3234 not set up until the daemon has forked a reception process, error responses
3235 that can happen earlier (such as "too many connections") no longer contain
3238 12. If an expansion in a condition on a "warn" statement fails because a lookup
3239 defers, the "warn" statement is abandoned, and the next ACL statement is
3240 processed. Previously this caused the whole ACL to be aborted.
3242 13. Added the iplsearch lookup type.
3244 14. Added ident_timeout as a log selector.
3246 15. Added tls_certificate_verified as a log selector.
3248 16. Added a global option tls_require_ciphers (compare the smtp transport
3249 option of the same name). This controls incoming TLS connections.
3251 17. I finally figured out how to make tls_require_ciphers do a similar thing
3252 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3253 before starting the TLS session.
3255 18. Tabs are now shown as \t in -bP output.
3257 19. If the log selector return_path_on_delivery was set, Exim crashed when
3258 bouncing a message because it had too many Received: header lines.
3260 20. If two routers both had headers_remove settings, and the first one included
3261 a superfluous trailing colon, the final name in the first list and the
3262 first name in the second list were incorrectly joined into one item (with a
3263 colon in the middle).
3269 1. Added -C and -D options to the exinext utility, mainly to make it easier
3270 to include in the automated testing, but these could be helpful when
3271 multiple configurations are in use.
3273 2. The exinext utility was not formatting the output nicely when there was
3274 an alternate port involved in the retry record key, nor when there was a
3275 message id as well (for retries that were specific to a specific message
3276 and a specific host). It was also confused by IPv6 addresses, because of
3277 the additional colons they contain. I have fixed the IPv4 problem, and
3278 patched it up to do a reasonable job for IPv6.
3280 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3281 delivery, the log line now contains "pipelined" if PIPELINING was used.
3283 4. An SMTP transport process used to panic and die if the bind() call to set
3284 an explicit outgoing interface failed. This has been changed; it is now
3285 treated in the same way as a connect() failure.
3287 5. A reference to $sender_host_name in the part of a conditional expansion
3288 that was being skipped was still causing a DNS lookup. This no longer
3291 6. The def: expansion condition was not recognizing references to header lines
3292 that used bh_ and bheader_.
3294 7. Added the _cache feature to named lists.
3296 8. The code for checking quota_filecount in the appendfile transport was
3297 allowing one more file than it should have been.
3299 9. For compatibility with Sendmail, the command line option
3307 and sets the incoming protocol and host name (for trusted callers). The
3308 host name and its colon can be omitted when only the protocol is to be set.
3309 Note the Exim already has two private options, -pd and -ps, that refer to
3310 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3311 "s", but I don't think that's a major issue.
3313 10. A number of refactoring changes to the code, none of which should affect
3316 (a) The number of logging options was getting close to filling up the
3317 32-bit word that was used as a bit map. I have split them into two classes:
3318 those that are passed in the argument to log_write(), and those that are
3319 only ever tested independently outside of that function. These are now in
3320 separate 32-bit words, so there is plenty of room for expansion again.
3321 There is no change in the user interface or the logging behaviour.
3323 (b) When building, for example, log lines, the code previously used a
3324 macro that called string_cat() twice, in order to add two strings. This is
3325 not really sufficiently general. Furthermore, there was one instance where
3326 it was actually wrong because one of the argument was used twice, and in
3327 one call a function was used. (As it happened, calling the function twice
3328 did not affect the overall behaviour.) The macro has been replaced by a
3329 function that can join an arbitrary number of extra strings onto a growing
3332 (c) The code for expansion conditions now uses a table and a binary chop
3333 instead of a serial search (which was left over from when there were very
3334 few conditions). Also, it now recognizes conditions like "pam" even when
3335 the relevant support is not compiled in: a suitably worded error message is
3336 given if an attempt is made to use such a condition.
3338 11. Added ${time_interval:xxxxx}.
3340 12. A bug was causing one of the ddress fields not to be passed back correctly
3341 from remote delivery subprocesses. The field in question was not being
3342 subsequently used, so this caused to problems in practice.
3344 13. Added new log selectors queue_time and deliver_time.
3346 14. Might have fixed a bug in maildirsizefile handling that threw up
3347 "unexpected character" debug warnings, and recalculated the data
3348 unnecessarily. In any case, I expanded the warning message to give more
3351 15. Added the message "Restricted characters in address" to the statements in
3352 the default ACL that block characters like @ and % in local parts.
3354 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3355 Three changes have been made:
3357 (a) There was a serious bug; a negative response to MAIL caused the whole
3358 recipient domain to be cached as invalid, thereby blocking all messages
3359 to all local parts at the same domain, from all senders. This bug has
3360 been fixed. The domain is no longer cached after a negative response to
3361 MAIL if the sender used is not empty.
3363 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3366 (c) A new callout option, "use_sender" has been added for people who want
3367 the modified behaviour.
3373 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3376 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3377 indeed breaks things for older releases.
3379 3. Added additional logging to the case where there is a problem reading data
3380 from a filter that is running in a subprocess using a pipe, in order to
3381 try to track down a specific problem.
3383 4. Testing facility fudge: when running in the test harness and attempting
3384 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3385 getting "No route to host". Convert this to a timeout.
3387 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3390 6. Some OS don't have socklen_t but use size_t instead. This affects the
3391 fifth argument of getsockopt() amongst other things. This is now
3392 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3393 can be set for individual OS. I have set it for SunOS5, OSF1, and
3394 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3395 some earlier ones do not.
3397 7. Change 4.30/15 was not doing the test caselessly.
3399 8. The standard form for an IPv6 address literal was being rejected by address
3400 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3401 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3402 this, as well as the form without the "IPv6" on the front (but only when
3403 address literals are enabled, of course).
3405 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3407 10. Exim crashed if a message with an empty sender address specified by -f
3408 encountered a router with an errors_to setting. This could be provoked only
3409 by a command such as
3413 where an empty string was supplied; "<>" did not hit this bug.
3415 11. Installed PCRE release 4.5.
3417 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3418 remained set. It is now erased.
3420 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3421 times from message ids (which are base 36 rather than the normal 62).
3423 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3424 were being counted as actual protocol errors, and logged if the log
3425 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3426 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3427 and DATA following a set of rejected RCPTs do not count as protocol errors.
3428 In other words, Exim assumes they were pipelined, though this may not
3429 actually be the case. Of course, in all cases the client gets an
3430 appropriate error code.
3432 15. If a lookup fails in an ACL condition, a message about the failure may
3433 be available; it is used if testing the ACL cannot continue, because most
3434 such messages specify what the cause of the deferral is. However, some
3435 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3436 that caused an old message to be retained and used if a later statement
3437 caused a defer, replacing the real cause of the deferral.
3439 16. If an IP address had so many PTR records that the DNS lookup buffer
3440 was not large enough to hold them, Exim could crash while trying to process
3441 the truncated data. It now detects and logs this case.
3443 17. Further to 4.21/58, another change has been made: if (and only if) the
3444 first line of a message (the first header line) ends with CRLF, a bare LF
3445 in a subsequent header line has a space inserted after it, so as not to
3446 terminate the header.
3448 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3449 unnecessarily, used atoi() instead of strtol(), and didn't check the
3450 termination when getting file sizes from file names by regex.
3452 19. Completely re-implemented the support for maildirsize files, in the light
3453 of a number of problems with the previous contributed implementation
3454 (4.30/29). In particular:
3456 . If the quota is zero, the maildirsize file is maintained, but no quota is
3459 . If the maildir directory does not exist, it is created before any attempt
3460 to write a maildirsize file.
3462 . The quota value in the file is just a cache; if the quota is changed in
3463 the transport, the new value overrides.
3465 . A regular expression is available for excluding directories from the
3468 20. The autoreply transport checks the characters in options that define the
3469 message's headers; it allows continued headers, but it was checking with
3470 isspace() after an embedded newline instead of explicitly looking for a
3473 21. If all the "regular" hosts to which an address was routed had passed their
3474 expiry times, and had not reached their retry times, the address was
3475 bounced, even if fallback hosts were defined. Now Exim should go on to try
3478 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3479 equivalent code in the SMTP transport. Some hosts send humungous responses
3480 to HELO/EHLO, more than 1024 it seems.
3482 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3483 gives compiler warnings in some environments. I've now done it "properly",
3486 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3487 (because of problems with the built-in one) was declared to return uschar *
3488 instead of char *, causing compiler failure.
3490 25. Fixed a file descriptor leak when processing alias/forward files.
3492 26. Fixed a minor format string issue in dbfn.c.
3494 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3496 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3497 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3500 29. When a sender address is verified, it is cached, to save repeating the test
3501 when there is more than one recipient in a message. However, when the
3502 verification involves a callout, it is possible for different callout
3503 options to be set for different recipients. It is too complicated to keep
3504 track of this in the cache, so now Exim always runs a verification when a
3505 callout is required, relying on the callout cache for the optimization.
3506 The overhead is duplication of the address routing, but this should not be
3509 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3510 to be verified with callout=postmaster, and the main callout worked but the
3511 postmaster check failed, the verification correctly failed. However, if a
3512 subsequent RCPT command asked for sender verification *without* the
3513 postmaster check, incorrect caching caused this verification also to fail,
3516 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3517 it was not caching the DNS options (qualify_single, search_parents) that
3518 were used when the lookup failed. A subsequent lookup with different
3519 options therefore always gave the same answer, though there were cases
3520 where it should not have. (Example: a "domains = !$mx_any" option on a
3521 dnslookup router: the "domains" option is always processed without any
3522 widening, but the router might have qualify_single set.) Now Exim uses the
3523 cached value only when the same options are set.
3525 32. Added John Jetmore's "exipick" utility to the distribution.
3527 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3528 than a timeout (e.g. a certificate is required, and is not provided), an
3529 Exim server now closes the connection immediately. Previously it waited for
3530 the client to close - but if the client is SSL, it seems that they each
3531 wait for each other, leading to a delay before one of them times out.
3533 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3534 maintained 0.8.x compatibility because I don't think many are using it, and
3535 it is clearly obsolete.
3537 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3540 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3541 client certificate was expired. A simple patch fixes this, though I don't
3542 understand the full logic of why the verify callback is called multiple
3545 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3546 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3547 which causes problems with some clients (such as the Certicom SSL Plus
3548 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3549 disables the coutermeasure allowing Eudora to connect."
3551 38. Exim was not checking that a write() to a log file succeeded. This could
3552 lead to Bad Things if a log got too big, in particular if it hit a file
3553 size limit. Exim now panics and dies if it cannot write to a log file, just
3554 as it does if it cannot open a log file.
3556 39. Modified OS/Makefile-Linux so that it now contains
3558 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3560 The two -D definitions ensure that Exim is compiled with large file
3561 support, which makes it possible to handle log files that are bigger than
3564 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3565 instance) a domain was checked against a named list that involved a lookup,
3566 causing $domain_data to be set, then another domain was checked against the
3567 same list, then the first domain was re-checked, the value of $domain_data
3568 after the final check could be wrong. In particular, if the second check
3569 failed, it could be set empty. This bug probably also applied to
3572 41. The strip_trailing_dot option was not being applied to the address given
3573 with the -f command-line option.
3575 42. The code for reading a message's header from the spool was incrementing
3576 $received_count, but never initializing it. This meant that the value was
3577 incorrect (doubled) while delivering a message in the same process in which
3578 it was received. In the most common configuration of Exim, this never
3579 happens - a fresh exec is done - but it can happen when
3580 deliver_drop_privilege is set.
3582 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3583 it now includes up to 150 characters of the unexpected data in the log
3586 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3587 and building data strings. The size of both of these buffers was 10 000
3588 bytes - far larger than anybody would *ever* want, thought I. Needless to
3589 say, somebody hit the limit. I have increased the maximum line length to
3590 20 000 and the maximum data length of concatenated lines to 100 000. I have
3591 also fixed two bugs, because there was no checking on these buffers. Tsk,
3592 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3593 buffer is too small.
3595 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3596 lsearch lookups. Now it does.
3598 46. When parsing a route_list item in a manualroute router, a fixed-length
3599 buffer was used for the list of hosts. I made this 1024 bytes long,
3600 thinking that nobody would ever have a list of hosts that long. Wrong.
3601 Somebody had a whole pile of complicated expansion conditions, and the
3602 string was silently truncated, leading to an expansion error. It turns out
3603 that it is easier to change to an unlimited length (owing to other changes
3604 that have happened since this code was originally written) than to build
3605 structure for giving a limitation error. The length of the item that
3606 expands into the list of hosts is now unlimited.
3608 47. The lsearch lookup could not handle data where the length of text line was
3609 more than 4095 characters. Such lines were truncated, leading to shortened
3610 data being returned. It should now handle lines of any length.
3612 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3613 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3616 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3617 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3618 no longer happens in comment lines. A list of replacements is now placed
3619 at the head of all of the source files, except those whose only change is
3620 to replace PERL_COMMAND in the very first #! line.
3622 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3623 messages on the queue, with a bottom-up merge sort, using code contributed
3624 by Michael Haardt. This should make operations like -bp somewhat faster on
3625 large queues. It won't affect queue runners, except when queue_run_in_order
3628 51. Installed eximstats 1.31 in the distribution.
3630 52. Added support for SRV lookups to the dnslookup router.
3632 53. If an ACL referred to $message_body or $message_body_end, the value was not
3633 reset for any messages that followed in the same SMTP session.
3635 54. The store-handling optimization for building very long strings was not
3636 differentiating between the different store pools. I don't think this
3637 actually made any difference in practice, but I've tidied it.
3639 55. While running the routers to verify a sender address, $sender_address
3640 was still set to the sender address. This is wrong, because when routing to
3641 send a bounce to the sender, it would be empty. Therefore, I have changed
3642 it so that, while verifying a sender address, $sender_address is set to <>.
3643 (There is no change to what happens when verifying a recipient address.)
3645 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3646 target A or AAAA records (if not already returned) without resetting the
3647 qualify_single or search_parents options of the DNS resolver. These are
3648 inappropriate in this case because the targets of MX and SRV records must
3649 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3650 target that, when qualified, matched something in the local domain. These
3651 two options are now turned off when doing these lookups.
3653 57. It seems that at least some releases of Reiserfs (which does not have the
3654 concept of a fixed number of inodes) returns zero and not -1 for the
3655 number of available inodes. This interacted badly with check_spool_inodes,
3656 which assumed that -1 was the "no such thing" setting. What I have done is
3657 to check that the total number of inodes is greater than zero before doing
3658 the test of how many are available.
3660 58. When a "warn" ACL statement has a log_message modifier, the message is
3661 remembered, and not repeated. This is to avoid a lot of repetition when a
3662 message has many recipients that cause the same warning to be written.
3663 Howewer, Exim was preserving the list of already written lines for an
3664 entire SMTP session, which doesn't seem right. The memory is now reset if a
3665 new message is started.
3667 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3668 debugging output unless log_rewrite was also set.
3670 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3671 of (int)(handle) when we know that handle contains (void *)(-1).
3673 61. The Exim daemon panic-logs an error return when it closes the incoming
3674 connection. However "connection reset by peer" seems to be common, and
3675 isn't really an error worthy of noting specially, so that particular error
3678 62. When Exim is trying to find all the local interfaces, it used to panic and
3679 die if the ioctl to get the interface flags failed. However, it seems that
3680 on at least one OS (Solaris 9) it is possible to have an interface that is
3681 included in the list of interfaces, but for which you get a failure error
3682 for this call. This happens when the interface is not "plumbed" into a
3683 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3684 failure of the "get flags" call assumes that the interface is down.
3686 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3687 makes life easier for people who are doing arithmetic on fields extracted
3688 from dates, where you often get leading zeros that should not be
3689 interpreted as octal.
3691 64. Added qualify_domain to the redirect router, to override the global
3694 65. If a pathologically long header line contained very many addresses (the
3695 report of this problem mentioned 10 000) and each of them was rewritten,
3696 Exim could use up a very large amount of memory. (It kept on making new
3697 copies of the header line as it rewrote, and never released the old ones.)
3698 At the expense of a bit more processing, the header rewriting function has
3699 been changed so that it no longer eats memory in this way.
3701 66. The generation of the Received: header has been moved from the time that a
3702 message starts to be received, to the time that it finishes. The timestamp
3703 in the Received: header should now be very close to that of the <= log
3704 line. There are two side-effects of this change:
3706 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3707 logged header lines no longer include the local Received: line, because
3708 it has not yet been created. The same applies to a copy of the message
3709 that is returned to a non-SMTP sender when a message is rejected.
3711 (b) When a filter file is tested using -bf, no additional Received: header
3712 is added to the test message. After some thought, I decided that this
3715 This change does not affect the value of $received_for. It is still set
3716 after address rewriting, but before local_scan() is called.
3718 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3720 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3721 gave an unhelpful panic error message, and a defer error. I have managed to
3722 change this behaviour so that it now rejects any supplied certificate,
3723 which seems right, as the list of acceptable certificates is empty.
3725 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3726 gave an unhelpful defer error. I have not managed to make this reject any
3727 supplied certificates, but the error message it gives is "no certificate
3728 supplied", which is not helpful.
3730 70. exigrep's output now also includes lines that are not associated with any
3731 message, but which match the given pattern. Implemented by a patch from
3732 Martin Sluka, which also tidied up the Perl a bit.
3734 71. Recipient callout verification, like sender verification, was using <> in
3735 the MAIL FROM command. This isn't really the right thing, since the actual
3736 sender may affect whether the remote host accepts the recipient or not. I
3737 have changed it to use the actual sender in the callout; this means that
3738 the cache record is now keyed on a recipient/sender pair, not just the
3739 recipient address. There doesn't seem to be a real danger of callout loops,
3740 since a callout by the remote host to check the sender would use <>.
3741 [SEE ABOVE: changed after hitting problems.]
3743 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3744 temporary errors. However, in the case of such a code being given after
3745 the end of a data transmission (i.e. after ".") Exim was failing to write
3746 a retry record for the message. (Yes, there was some broken host that was
3747 actually sending 8xx at this point.)
3749 73. An unknown lookup type in a host list could cause Exim to panic-die when
3750 the list was checked. (An example that provoked this was putting <; in the
3751 middle of a list instead of at the start.) If this happened during a DATA
3752 ACL check, a -D file could be left lying around. This kind of configuration
3753 error no longer causes Exim to die; instead it causes a defer errror. The
3754 incident is still logged to the main and panic logs.
3756 74. Buglet left over from Exim 3 conversion. The message "too many messages
3757 in one connection" was written to the rejectlog but not the mainlog, except
3758 when address rewriting (yes!) was being logged.
3760 75. Added write_rejectlog option.
3762 76. When a system filter was run not as root (that is, when system_filter_user
3763 was set), the values of the $n variables were not being returned to the
3764 main process; thus, they were not subsequently available in the $sn
3767 77. Added +return_path_on_delivery log selector.
3769 78. A connection timeout was being treated differently from recipients deferred
3770 when testing hosts_max_try with a message that was older than the host's
3771 retry timeout. (The host should not be counted, thus allowing all hosts to
3772 be tried at least once before bouncing.) This may have been the cause of an
3773 occasionally reported bug whereby a message would remain on the queue
3774 longer than the retry timeout, but would be bounced if a delivery was
3775 forced. I say "may" because I never totally pinned down the problem;
3776 setting up timeout/retry tests is difficult. See also the next item.
3778 79. The ultimate address timeout was not being applied to errors that involved
3779 a combination of host plus message (for example, a timeout on a MAIL
3780 command). When an address resolved to a number of possible hosts, and they
3781 were not all tried for each delivery (e.g. because of hosts_max_try), a
3782 message could remain on the queue longer than the retry timeout.
3784 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
3787 81. Fixed an obscure SMTP outgoing bug which required at least the following
3788 conditions: (a) there was another message waiting for the same server;
3789 (b) the server returned 5xx to all RCPT commands in the first message so
3790 that the message was not completed; (c) the server dropped the connection
3791 or gave a negative response to the RSET that Exim sends to abort the
3792 transaction. The observed case was a dropped connection after DATA that had
3793 been sent in pipelining mode. That is, the server had advertised PIPELINING
3794 but was not implementing it correctly. The effect of the bug was incorrect
3795 behaviour, such as trying another host, and this could lead to a crash.
3801 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
3802 and daemon.c were passed as pointers to ints; they should have been
3803 pointers to socklen_t variables (which are typically unsigned ints).
3805 2. Some signed/unsigned type warnings in the os.c file for Linux have been
3808 3. Fixed a really odd bug that affected only the testing scheme; patching a
3809 certain fixed string in the binary changed the value of another string that
3810 happened to be identical to the end of the original first string.
3812 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
3813 name", it returns that address as the IP address. On some operating
3814 systems (e.g. Solaris), it also passes back the IP address string as the
3815 "host name". However, on others (e.g. Linux), it passes back an empty
3816 string. Exim wasn't checking for this, and was changing the host name to an
3817 empty string, assuming it had been canonicized.
3819 5. Although rare, it is permitted to have more than one PTR record for a given
3820 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
3821 all the names associated with an address, because they do in Solaris.
3822 However, it seems that they do not in Linux for data that comes from the
3823 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3824 I found this out when I moved to a new Linux workstation and tried to run
3825 the Exim test suite.
3827 To get round this problem I have changed the code so that it now does its
3828 own call to the DNS to look up PTR records when searching for a host name.
3829 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3830 addresses that are only in /etc/hosts are still found.
3832 This behaviour is, however, controlled by an option called host_lookup_
3833 order, which defaults to "bydns:byaddr". If people want to use the other
3834 order, or indeed, just use one or the other means of lookup, they can
3835 specify it in this variable.
3837 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3838 some operating systems, this comes back from gethostbyaddr() as an empty
3839 string, and this is what Exim used to test for. However, it seems that in
3840 other systems, "." is yielded. Exim now tests for this case too.
3842 7. The values of check_spool_space and check_log_space are now held internally
3843 as a number of kilobytes instead of an absolute number of bytes. If a
3844 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3845 kilobyte. This means that much larger values can be stored.
3847 8. Exim monitor: an attempt to get the action menu when not actually pointing
3848 at a message produces an empty menu entitled "No message selected". This
3849 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3850 no entries in it ("Shell widget menu has zero width and/or height"). So I
3851 have added a single, blank menu entry in this case.
3853 9. Added ${quote_local_part.
3855 10. MIME decoding is now applied to the contents of Subject: header lines when
3858 11. Now that a reference to $sender_host_address automatically causes a reverse
3859 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3860 host lookup before query-style lookups in lists that might use this
3861 variable. This has therefore been abolished, and the "net-" prefix is no
3862 longer necessary for query-style lookups.
3864 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3865 have been a typo for LFLAGS, so it has been changed.
3867 13. The install script calls Exim with "-C /dev/null" in order to find the
3868 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3869 to be output. Howeve, since Exim outputs its version number before the
3870 error, it didn't break the script. It just looked ugly. I fixed this by
3871 always allowing "-C /dev/null" if the caller is root.
3873 14. Ignore overlarge ACL variable number when reading spool file - insurance
3874 against a later release with more variables having written the file.
3876 15. The standard form for an IPv6 address literal was being rejected by EHLO.
3877 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3878 this, as well as the form without the "IPv6" on the front.
3880 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3881 OS/Makefile-Darwin file.
3883 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3884 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3886 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
3887 variable. However, some OS use ints for inodes. I've added cast to long int
3888 to get rid of the compiler warning.
3890 19. I had forgotten to lock out "/../" in configuration file names when
3891 ALT_CONFIG_PREFIX was set.
3893 20. Routers used for verification do not need to specify transports. However,
3894 if such a router generated a host list, and callout was configured, Exim
3895 crashed, because it could not find a port number from the (non-existent)
3896 transport. It now assumes port 25 in this circumstance.
3898 21. Added the -t option to exigrep.
3900 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3901 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3902 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3903 them is set, LOOKUP_LSEARCH is forced.
3905 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3907 24. Added sender and host information to the "rejected by local_scan()" log
3908 line; previously there was no indication of these.
3910 25. Added .include_if_exists.
3912 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3913 Linux. It turns out that not all file systems support this. Apparently some
3914 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3915 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3916 not supported on the file descriptor, is now ignored when Exim is trying to
3917 sync a directory. This applies only to Linux.
3919 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3921 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3922 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3925 29. Added support for maildirsize files from supplied patch (modified a bit).
3927 30. The use of :fail: followed by an empty string could lead Exim to respond to
3928 sender verification failures with (e.g.):
3930 550 Verification failed for <xxx>
3931 550 Sender verify failed
3933 where the first response line was missing the '-' that indicates it is not
3934 the final line of the response.
3936 31. The loop for finding the name of the user that called Exim had a hardwired
3937 limit of 10; it now uses the value of finduser_retries, which is used for
3938 all other user lookups.
3940 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3943 33. Exim was neglecting to zero errno before one call of strtol() when
3944 expanding a string and expecting an integer value. On some systems this
3945 resulted in spurious "integer overflow" errors. Also, it was casting the
3946 result into an int without checking.
3948 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3949 did not work. The code looks as if it has *never* worked, though it appears
3950 to have been documented since at least releast 1.62. I have made it work.
3952 35. The "timeout_DNS" error in retry rules, also documented since at least
3953 1.62, also never worked. As it isn't clear exactly what this means, and
3954 clearly it isn't a major issue, I have abolished the feature by treating it
3955 as "timeout", and writing a warning to the main and panic logs.
3957 36. The display of retry rules for -brt wasn't always showing the error code
3960 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3961 timeout_connect_A, timeout_connect_MX.
3963 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3964 to the empty sender.
3966 39. The daemon was not analysing the content of -oX till after it had closed
3967 stderr and disconnected from the controlling terminal. This meant that any
3968 syntax errors were only noted on the panic log, and the return code from
3969 the command was 0. By re-arranging the code a little, I've made the
3970 decoding happen first, so such errors now appear on stderr, and the return
3971 code is 1. However, the actual setting up of the sockets still happens in
3972 the disconnected process, so errors there are still only recorded on the
3975 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3976 connections (as happens on some IP stacks) was logged at start up time as
3977 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3978 it from "IPv6 and IPv4", which means that two separate sockets are being
3981 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3982 says whether AF_INET or AF_INET6 was passed as an argument.
3984 42. Exiwhat output was messed up when time zones were included in log
3987 43. Exiwhat now gives more information about the daemon's listening ports,
3988 and whether -tls-on-connect was used.
3990 44. The "port" option of the smtp transport is now expanded.
3992 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3993 silently ignored. Now an error message is written to the main and panic
3996 46. There's a new ACL modifier called "logwrite" which writes to a log file
3997 as soon as it is encountered.
3999 47. Added $local_user_uid and $local_user_gid at routing time.
4001 48. Exim crashed when trying to verify a sender address that was being
4004 49. Exim was recognizing only a space character after ".include". It now also
4005 recognizes a tab character.
4007 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4008 extracting the relevant information from the specification. The man page no
4009 longer contains scrambled data for the -d option, and I've added a section
4010 at the front about calling Exim under different names.
4012 51. Added "extra_headers" argument to the "mail" command in filter files.
4014 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4017 53. Installed eximstats 1.29.
4019 54. Added transport_filter_timeout as a generic transport option.
4021 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4022 Cc: header lines. This was required by RFC 822, but it not required by RFC
4025 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4026 incoming messages that did not have them. Now it does so only if the
4027 message originates locally, that is, if there is no associated remote host
4028 address. When Resent- header lines are present, this applies to the Resent-
4029 lines rather than the non-Resent- lines.
4031 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4032 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4034 58. Messages for configuration errors now include the name of the main
4035 configuration file - useful now that there may be more than one file in a
4036 list (.included file names were always shown).
4038 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4039 for those rare installations that do not start the daemon as root or run it
4040 setuid root. I've cut out the call to initgroups() if the daemon is not
4043 60. The Exim user and group can now be bound into the binary as text strings
4044 that are looked up at the start of Exim's processing.
4046 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4048 62. Added $mailstore_basename variable.
4050 63. Installed patch to sieve.c from Michael Haardt.
4052 64. When Exim failed to open the panic log after failing to open the main log,
4053 the original message it was trying to log was written to stderr and debug
4054 output, but if they were not available (the usual case in production), it
4055 was lost. Now it is written to syslog before the two lines that record the
4056 failures to open the logs.
4058 65. Users' Exim filters run in subprocesses under the user's uid. It is
4059 possible for a "deliver" command or an alias in a "personal" command to
4060 provoke an address rewrite. If logging of address rewriting is configured,
4061 this fails because the process is not running as root or exim. There may be
4062 a better way of dealing with this, but for the moment (because 4.30 needs
4063 to be released), I have disabled address rewrite logging when running a
4064 filter in a non-root, non-exim process.
4070 1. The buildconfig auxiliary program wasn't quoting the value set for
4071 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4072 not defined. This bug was masked in 4.22 by the effect that was fixed in
4075 2. Some messages that were rejected after a message id was allocated were
4076 shown as "incomplete" by exigrep. It no longer does this for messages that
4077 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4079 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4080 have allow_domain_literals set, the ID did not get logged in the <= line.
4081 Domain literals are now always recognized in Message-ID: header lines.
4083 4. The first argument for a ${extract expansion item is the key name or field
4084 number. Leading and trailing spaces in this item were not being ignored,
4085 causing some misleading effects.
4087 5. When deliver_drop_privilege was set, single queue runner processes started
4088 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4089 same command in the process it spins off) were not dropping privilege.
4091 6. When the daemon running as "exim" started a queue runner, it always
4092 re-executed Exim in the spun-off process. This is a waste of effort when
4093 deliver_drop_privilege is set. The new process now just calls the
4094 queue-runner function directly.
4100 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4103 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4104 ignored. Though the use of -oP was forcing the writing of a pid file, it
4105 was always written to the default place.
4107 3. If the message "no IP address found for host xxxx" is generated during
4108 incoming verification, it is now followed by identification of the incoming
4109 connection (so you can more easily find what provoked it).
4111 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4113 5. Added some features to "harden" Exim a bit more against certain attacks:
4115 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4116 be put in Local/Makefile. This is like the never_users runtime option,
4117 but it cannot be overridden. The default setting is "root".
4119 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4120 prefix string with which any file named in a -C command line option
4123 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4124 is retained for -C and -D only if the caller of Exim is root. Without
4125 it, the exim user may also use -C and -D and retain privilege.
4127 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4128 command line option is disabled.
4130 6. Macro names set by the -D option must start with an upper case letter, just
4131 like macro names defined in the configuration file.
4133 7. Added "dereference=" facility to LDAP.
4135 8. Two instances of the typo "uknown" in the source files are fixed.
4137 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4138 the Configure-Makefile script screwed up while processing it.
4140 10. Incorporated PCRE 4.4.
4142 11. The SMTP synchronization check was not operating right at the start of an
4143 SMTP session. For example, it could not catch a HELO sent before the client
4144 waited for the greeting. There is now a check for outstanding input at the
4145 point when the greeting is written. Because of the duplex, asynchronous
4146 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4147 way, but not yet received, when the check is performed.
4149 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4150 on TCP/IP sockets, because this apparently causes some broken clients to
4153 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4154 unchanged) from the Cygwin maintainer.
4156 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4157 was supported instead of testing for mbx. Effectively a typo.
4159 15. The spa authenticator server code was not checking that the input it
4160 received was valid base64.
4162 16. The debug output line for the "set" modifier in ACLs was not showing the
4163 name of the variable that was being set.
4165 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4167 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4168 lookup on its own. Something else was needed to trigger the lookup. For
4169 example, a match in host_lookup or the need for a host name in a host list.
4170 Now, if $sender_host_name is referenced and the host name has not yet been
4171 looked up, a lookup is performed. If the lookup fails, the variable remains
4172 empty, and $host_lookup_failed is set to "1".
4174 19. Added "eqi" as a case-independent comparison operator.
4176 20. The saslauthd authentication condition could segfault if neither service
4177 nor realm was specified.
4179 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4180 error message that was logged was misleading, and incoming SMTP
4181 connections were dropped. The message is now more accurate, and temporary
4182 errors are given to SMTP connections.
4184 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4185 (except RSET) with a 421 error, until QUIT is received. However, it was
4186 failing to send a response to QUIT.
4188 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4189 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4190 ACL is now run after the helo_try_verify_hosts code.
4192 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4193 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4194 case-independent, so other case variants are also recognized). Apparently
4195 some systems use these upper case variants.
4197 25. If more than two messages were waiting for the same host, and a transport
4198 filter was specified for the transport, Exim sent two messages over the
4199 same TCP/IP connection, and then failed with "socket operation on non-
4200 socket" when it tried to send the third.
4202 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4204 27. The extern definition of crypt16() in expand.c was not being excluded when
4205 the OS had its own crypt16() function.
4207 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4208 as a preferred synonym for return_size_limit, both as an option and as an
4211 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4213 30. Changed the default configuration ACL to relax the local part checking rule
4214 for addresses that are not in any local domains. For these addresses,
4215 slashes and pipe symbols are allowed within local parts, but the sequence
4216 /../ is explicitly forbidden.
4218 31. SPA server authentication was not clearing the challenge buffer before
4221 32. log_message in a "warn" ACL statement was writing to the reject log as
4222 well as to the main log, which contradicts the documentation and doesn't
4223 seem right (because no rejection is happening). So I have stopped it.
4225 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4226 However, I am unable to do any testing of this.
4228 34. Fixed an infelicity in the appendfile transport. When checking directories
4229 for a mailbox, to see if any needed to be created, it was accidentally
4230 using path names with one or more superfluous leading slashes; tracing
4231 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4233 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4234 contents are added to the log line that is written for every discarded
4235 recipient. (Previously a log_message setting was ignored.)
4237 36. The ${quote: operator now quotes the string if it is empty.
4239 37. The install script runs exim in order to find its version number. If for
4240 some reason other than non-existence or emptiness, which it checks, it
4241 could not run './exim', it was installing it with an empty version number,
4242 i.e. as "exim-". This error state is now caught, and the installation is
4245 38. An argument was missing from the function that creates an error message
4246 when Exim fails to connect to the socket for saslauthd authentication.
4247 This could cause Exim to crash, or give a corrupted message.
4249 39. Added isip, isip4, and isip6 to ${if conditions.
4251 40. The ACL variables $acl_xx are now saved with the message, and can be
4252 accessed later in routers, transports, and filters.
4254 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4255 strings in the file are not string-expanded.
4257 42. If a MAIL command specified a SIZE value that was too large to fit into an
4258 int variable, the check against message_size_limit failed. Such values are
4259 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4260 day this will have to be increased, but I don't think I want to be around
4261 when emails are that large.
4268 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4269 iconv() is not standard in FreeBSD.
4271 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4272 IPv6 enabled. The observed symptom was a segmentation fault on return from
4273 the function os_common_find_running_interfaces() in src/os.c.
4275 3. In the check_special_case() function in daemon.c I had used "errno" as an
4276 argument name, which causes warnings on some systems. This was basically a
4277 typo, since it was named "eno" in the comments!
4279 4. The code that waits for the clock to tick (at a resolution of some fraction
4280 of a second) so as to ensure message-id uniqueness was always waiting for
4281 at least one whole tick, when it could have waited for less. [This is
4282 almost certainly not relevant at current processor speeds, where it is
4283 unlikely to ever wait at all. But we try to future-proof.]
4285 5. The function that sleeps for a time interval that includes fractions of a
4286 second contained a race. It did not block SIGALRM between setting the
4287 timer, and suspending (a couple of lines later). If the interval was short
4288 and the sigsuspend() was delayed until after it had expired, the suspension
4289 never ended. On busy systems this could lead to processes getting stuck for
4292 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4293 process, before it forks any delivery processes. The open lookup caching
4294 mechanism meant that the open file or database connection was passed into
4295 the delivery process. The problem was that delivery processes always tidy
4296 up cached lookup data. This could cause a problem for the next delivery
4297 process started by the queue runner, because the external queue runner
4298 process does not know about the closure. So the next delivery process
4299 still has data in the lookup cache. In the case of a file lookup, there was
4300 no problem because closing a file descriptor in a subprocess doesn't affect
4301 the parent. However, if the lookup was caching a connection to a database,
4302 the connection was closed, and the second delivery process was likely to
4303 see errors such as "PGSQL: query failed: server closed the connection
4304 unexpectedly". The problem has been fixed by closing all cached lookups
4305 in a queue runner before running a delivery process.
4307 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4308 seem to have the "const" qualifier which it has on other OS. I've
4311 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4312 *of the same type* (client or server) with the same public name that an
4313 error should be diagnosed.
4315 9. When Exim looked up a host name for an IP address, but failed to find the
4316 original IP address when looking up the host name (a safety check), it
4317 output the message "<ip address> does not match any IP for NULL", which was
4318 confusing, to say the least. The bug was that the host name should have
4319 appeared instead of "NULL".
4321 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4322 user that is built into the binary, and the -C or -D options is used, root
4323 privilege is dropped before the configuration file is read. In addition,
4324 logging is switched to stderr instead of the normal log files. If the
4325 configuration then re-defines the Exim user, the unprivileged environment
4326 is probably not what is expected, so Exim logs a panic warning message (but
4329 However, if deliver_drop_privilege is set, the unprivileged state may well
4330 be exactly what is intended, so the warning has been cut out in that case,
4331 and Exim is allowed to try to write to its normal log files.
4337 1. smtp_return_error_details was not giving details for temporary sender
4338 or receiver verification errors.
4340 2. Diagnose a configuration error if two authenticators have the same public
4343 3. Exim used not to create the message log file for a message until the first
4344 delivery attempt. This could be confusing when incoming messages were held
4345 for policy or load reasons. The message log file is now created at the time
4346 the message is received, and an initial "Received" line is written to it.
4348 4. The automatically generated man page for command line options had a minor
4349 bug that caused no ill effects; however, a more serious problem was that
4350 the procedure for building the man page automatically didn't always
4351 operate. Consequently, release 4.20 contains an out-of-date version. This
4352 shouldn't happen again.
4354 5. When building Exim with embedded Perl support, the script that builds the
4355 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4356 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4358 6. The freeze_tell option was not being used for messages that were frozen on
4359 arrival, either by an ACL or by local_scan().
4361 7. Added the smtp_incomplete_transaction log selector.
4363 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4364 was accepting AUTH without a new EHLO.
4366 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4367 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4370 10. Logging of TCP/IP connections (when configured) now happens in the main
4371 daemon process instead of the child process, so that the TCP/IP connection
4372 count is more accurate (but it can never be perfect).
4374 11. The use of "drop" in a nested ACL was not being handled correctly in the
4375 outer ACL. Now, if condition failure induced by the nested "drop" causes
4376 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4377 or "require"), the connection is dropped.
4379 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4380 that yield "discard" can now be used with an "accept" or a "discard" verb,
4381 but an error is generated for any others (because I can't see a useful way
4382 to define what should happen).
4384 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4385 are now processed in the same way as lines in the Exim configuration file.
4386 In particular, continuation lines are supported.
4388 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4390 15. Added -ti meaning -t -i.
4392 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4393 domains, and warn by logging if others are found.
4395 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4396 returned from GIFCONF when Exim is trying to find the list of interfaces on
4397 a host. The code in os.c has been modified to copy each ifreq to an aligned
4398 structure in all cases.
4400 Also, in some cases, the returned ifreq's were being copied to a 'struct
4401 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4402 means the last couple of bytes of an IPv6 address could be chopped if the
4403 ifreq contained only a normal sockaddr (14 bytes storage).
4405 18. Named domain lists were not supported in the hosts_treat_as_local option.
4406 An entry such as +xxxx was not recognized, and was treated as a literal
4409 19. Ensure that header lines added by a DATA ACL are included in the reject log
4410 if the ACL subsequently rejects the message.
4412 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4413 MD5 (which is deprecated).
4415 21. When testing a filter file using -bf, Exim was writing a message when it
4416 took the sender from a "From " line in the message, but it was not doing so
4417 when it took $return_path from a Return-Path: header line. It now does.
4419 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4420 with a valid header line field name (a series of printing characters
4421 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4423 23. Changed "disc" in the source to "disk" to conform to the documentation and
4424 the book and for uniformity.
4426 24. Ignore Sendmail's -Ooption=value command line item.
4428 25. When execve() failed while trying to run a command in a pipe transport,
4429 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4430 could be confused with a return value of 69 from the command itself. This
4431 has been changed to 127, the value the shell returns if it is asked to run
4432 a non-existent command. The wording for the related log line suggests a
4433 non-existent command as the problem.
4435 26. If received_header_text expands to an empty string, do not add a Received:
4436 header line to the message. (Well, it adds a token one on the spool, but
4437 marks it "old" so that it doesn't get used or transmitted.)
4439 27. Installed eximstats 1.28 (addition of -nt option).
4441 28. There was no check for failure on the call to getsockname() in the daemon
4442 code. This can fail if there is a shortage of resources on the system, with
4443 ENOMEM, for example. A temporary error is now given on failure.
4445 29. Contrary to the C standard, it seems that in some environments, the
4446 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4447 program. Exim now does this explicitly; it affects the formatting of
4448 timestamps using strftime().
4450 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4451 complaints. I've now initialized all the variables, to avoid this.
4453 32. Header lines added by a system filter were not being "seen" during
4454 transport-time rewrites.
4456 33. The info_callback() function passed to OpenSSL is set up with type void
4457 (*)(SSL *, int, int), as described somewhere. However, when calling the
4458 function (actually a macro) that sets it up, the type void(*)() is
4459 expected. I've put in a cast to prevent warnings from picky compilers.
4461 34. If a DNS black list lookup found a CNAME record, but there were no A
4462 records associated with the domain it pointed at, Exim crashed.
4464 35. If a DNS black list lookup returned more than one A record, Exim ignored
4465 all but the first. It now scans all returned addresses if a particular IP
4466 value is being sought. In this situation, the contents of the
4467 $dnslist_value variable are a list of all the addresses, separated by a
4470 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4471 to accept a host name and all its aliases if the forward lookup for any of
4472 them yielded the IP address of the incoming connection. Now it accepts only
4473 those names whose forward lookup yields the correct IP address. Any other
4474 names are discarded. This closes a loophole whereby a rogue DNS
4475 administrator could create reverse DNS records to break through a
4476 wildcarded host restriction in an ACL.
4478 37. If a user filter or a system filter that ran in a subprocess used any of
4479 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4480 the wrong values were passed to the pipe command ($thisaddress had the
4481 value of $0, $0 had the value of $1, etc). This bug was introduced by
4482 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4484 38. Improved the line breaking for long SMTP error messages from ACLs.
4485 Previously, if there was no break point between 40 and 75 characters, Exim
4486 left the rest of the message alone. Two changes have been made: (a) I've
4487 reduced the minimum length to 35 characters; (b) if it can't find a break
4488 point between 35 and 75 characters, it looks ahead and uses the first one
4489 that it finds. This may give the occasional overlong line, but at least the
4490 remaining text gets split now.
4492 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4493 file descriptors might be low, and that setting 1000 would always raise it.
4494 It turns out that in some environments, the limit is already over 1000 and
4495 that lowering it causes trouble. So now Exim takes care not to decrease it.
4497 40. When delivering a message, the value of $return_path is set to $sender_
4498 address at the start of routing (routers may change the value). By an
4499 oversight, this default was not being set up when an address was tested by
4500 -bt or -bv, which affected the outcome if any router or filter referred to
4503 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4504 log only when "message" or "log_message" are set. However, if one of the
4505 conditions was an address verification, or a call to a nested ACL, the
4506 messages generated by the underlying test were being passed through. This
4507 no longer happens. The underlying message is available in $acl_verify_
4508 message for both "message" and "log_message" expansions, so it can be
4509 passed through if needed.
4511 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4512 new expansion $bh_ to give the encoded byte string without charset
4513 translation. Translation happens only if iconv() is available; HAVE_ICONV
4514 indicates this at build time. HEADERS_CHARSET gives the charset to
4515 translate to; headers_charset can change it in the configuration, and
4516 "headers charset" can change it in an individual filter file.
4518 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4519 that creates RFC 2047 encoded "words" labels them as that charset instead
4520 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4521 expansion operator; (ii) when Exim creates a From: line for a local
4522 message; (iii) when a header line is rewritten to include a "phrase" part.
4524 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4525 buggy, causing it to skip the first lines of messages whose message ID
4526 ended in 'D'. This would not have bitten before Exim release 4.14, because
4527 message IDs were unlikely to end in 'D' before then. The effect was to have
4528 incorrect size information for certain domains.
4530 45. #include "config.h" was missing at the start of the crypt16.c module. This
4531 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4534 46. If there was a timeout during a "random" callout check, Exim treated it as
4535 a failure of the random address, and carried on sending RSET and the real
4536 address. If the delay was just some slowness somewhere, the response to the
4537 original RCPT would be taken as a response to RSET and so on, causing
4538 mayhem of various kinds.
4540 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4541 when I implemented it. It didn't allow for the fact that some option values
4542 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4543 the right test for positive values.
4545 48. Domain names in DNS records are case-independent. Exim always looks them up
4546 in lower case. Some resolvers return domain names in exactly the case they
4547 appear in the zone file, that is, they may contain uppercase letters. Not
4548 all resolvers do this - some return always lower case. Exim was treating a
4549 change of case by a resolver as a change of domain, similar to a widening
4550 of a domain abbreviation. This triggered its re-routing code and so it was
4551 trying to route what was effectively the same domain again. This normally
4552 caused routing to fail (because the router wouldn't handle the domain
4553 twice). Now Exim checks for this case specially, and just changes the
4554 casing of the domain that it ultimately uses when it transmits the message
4557 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4560 50. If a filter generated a file delivery with a non-absolute name (possible if
4561 no home directory exists for the router), the forbid_file option was not
4564 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4565 the existing equality matching.
4567 52. Exim was using ints instead of ino_t variables in some places where it was
4568 dealing with inode numbers.
4570 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4571 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4572 called TMPDIR, and if it finds it is different, it changes its value.
4574 54. The smtp_printf() function is now made available to local_scan() so
4575 additional output lines can be written before returning. There is also an
4576 smtp_fflush() function to enable the detection of a dropped connection.
4577 The variables smtp_input and smtp_batched_input are exported to
4580 55. Changed the default runtime configuration: the message "Unknown user"
4581 has been removed from the ACL, and instead placed on the localuser router,
4582 using the cannot_route_message feature. This means that any verification
4583 failures that generate their own messages won't get overridden. Similarly,
4584 the "Unrouteable address" message that was in the ACL for unverifiable
4585 relay addresses has also been removed.
4587 56. Added hosts_avoid_esmtp to the smtp transport.
4589 57. The exicyclog script was not checking for the esoteric option
4590 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4591 will work only if exicyclog is run under the appropriate euid.
4593 58. Following a discussion on the list, the rules by which Exim recognises line
4594 endings on incoming messages have been changed. The -dropcr and drop_cr
4595 options are now no-ops, retained only for backwards compatibility. The
4596 following line terminators are recognized: LF CRLF CR. However, special
4597 processing applies to CR:
4599 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4600 nor a local message in the state where . is a terminator.
4602 (ii) If a bare CR is encountered in a header line, an extra space is added
4603 after the line terminator so as not to end the header. The reasoning
4604 behind this is that bare CRs in header lines are most likely either
4605 to be mistakes, or people trying to play silly games.
4607 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4608 was being incorrectly given as 18 bytes larger than it should have been.
4609 This is a VOB (very old bug).
4611 60. This may never have affected anything current, but just in case it has:
4612 When the local host is found other than at the start of a list of hosts,
4613 the local host, those with the same MX, and any that follow, are discarded.
4614 When the list in question was part of a longer list of hosts, the following
4615 hosts (not currently being processed) were also being discarded. This no
4616 longer happens. I'm not sure if this situation could ever has previously
4619 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4622 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4623 header. This is something that is recommended in a new Internet Draft, and
4624 is something that is documented as being done by Sendmail. There are two
4625 possible values. For messages generated by the autoreply transport, Exim
4628 Auto-Submitted: auto-replied
4630 whereas for all other generated messages (e.g. bounces) it adds
4632 Auto-Submitted: auto-generated
4634 63. The "personal" condition in filters now includes a test for the
4635 Auto-Submitted: header. If it contains the string "auto-" the message it
4636 not considered personal.
4638 64. Added rcpt_include_affixes as a generic transport option.
4640 65. Added queue_only_override (default true).
4642 66. Added the syslog_duplication option.
4644 67. If what should have been the first header line of a message consisted of
4645 a space followed by a colon, Exim was mis-interpreting it as a header line.
4646 It isn't of course - it is syntactically invalid and should therefore be
4647 treated as the start of the message body. The misbehaviour could have
4648 caused a number of strange effects, including loss of data in subsequent
4649 header lines, and spool format errors.
4651 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4652 client host had authenticated. This control can now be exercised by an ACL
4653 for more flexibility.
4655 69. By default, callouts do not happen when testing with -bh. There is now a
4656 variant, -bhc, which does actually run the callout code, including
4657 consulting and updating the callout cache.
4659 70. Added support for saslauthd authentication, courtesy of Alexander
4662 71. If statvfs() failed on the spool or log directories while checking their
4663 size for availability, Exim confusingly gave the error "space shortage".
4664 Furthermore, in debugging mode it crashed with a floating point exception.
4665 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4666 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4667 serious problem, Exim now writes to the main and panic logs when this
4668 happens, with details of the failure. It then refuses to accept the
4669 incoming message, giving the message "spool directory problem" or "log
4670 directory problem" with a 421 code for SMTP messages.
4672 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4673 0, 1, and 2 exist, because some OS complain for execs without them (see
4674 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4675 descriptors. However, the code omitted to check that the open succeeded,
4676 causing mysterious errors if for some reason the permissions on /dev/null
4677 got screwed. Now Exim writes a message to the main and panic logs, and
4678 bombs out if it can't open /dev/null.
4680 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4681 interact so that it is all more flexible. It is supposed to remain
4682 backwards compatible. Also added extra_local_interfaces.
4684 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4685 to bomb out with an assertion failure - to the client this appears as a
4686 connection drop. This problem occurs in the part of the code that was taken
4687 from the Samba project. Fortunately, the assertion is in a very simple
4688 function, so I have fixed this by reproducing the function inline in the
4689 one place where it is called, and arranging for authentication to fail
4690 instead of killing the process with assert().
4692 75. The SPA client code was not working when the server requested OEM rather
4693 than Unicode encoding.
4695 76. Added code to make require_files with a specific uid setting more usable in
4696 the case where statting the file as root fails - usually a non-root-mounted
4697 NFS file system. When this happens and the failure is EACCES, Exim now
4698 forks a subprocess and does the per-uid checking as the relevant uid.
4700 77. Added process_log_path.
4702 78. If log_file_path was not explicitly set, a setting of check_log_space or
4703 check_log_inodes was ignored.
4705 79. If a space check for the spool or log partitions fails, the incident is now
4706 logged. Of course, in the latter case the data may get lost...
4708 80. Added the %p formatting code to string_format() so that it can be used to
4709 print addresses in debug_print(). Adjusted all the address printing in the
4710 debugging in store.c to use %p rather than %d.
4712 81. There was a concern that a line of code in smtp_in.c could overflow a
4713 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4714 initially expressed, the concern was not well-founded, because trailing
4715 spaces are removed early. However, if the trailing spaces were followed by
4716 a NULL, they did not get removed, so the overflow was possible. Two fixes
4719 (a) I re-wrote the offending code in a cleaner fashion.
4720 (b) If an incoming SMTP command contains a NULL character, it is rejected
4723 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4724 runs initgroups(), so that if the Exim user is in any additional groups,
4725 they will be used during message reception.
4731 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob